live_tv
Livestream Starting Soon
00
Hours
:
00
Minutes
:
00
Seconds
Up next in 10
This week, with special guest Christina Warren: The surprise drop of AirPods Max 2, a deep dive on OpenClaw and agentic coding, and our final thoughts after a full week with the Studio Display XDR, MacBook Neo and iPhone 17e!
Produced by Extra Ordinary for Cult of Mac
Music composed by Will Davenport, arranged by D. Griffin Jones
Chapters:
0:00 - Intro
4:04 - AirPods Max 2
28:21 - OpenClaw and agentic coding
54:27 - MacBook Neo
1:08:21 - Studio Display XDR
1:15:59 - iPhone 17e
Follow us!
Threads: https://www.threads.net/@cultofmac
Mastodon: https://mastodon.social/@cultofmac
Instagram: https://instagram.com/cultofmac/
X: https://x.com/cultofmac
Facebook: https://www.facebook.com/cultofmac
Show More Show Less View Video Transcript
0:00
Coming up, surprise drop of AirPods Max
0:02
2, a deep dive on Open Core and Agent
0:05
Coding, and our final thoughts after a
0:07
full week with the Studio Display XDR,
0:10
MacBook Neo, and iPhone 17e. And a
0:13
special guest, Christina Warren, aka
0:15
Film Girl.
0:19
Hello, welcome to the Cult of Mac
0:20
podcast. I'm your host, Leander Kaney.
0:22
Joining me today, we have D. Griffin
0:24
Jones. Hey, Griffin.
0:25
>> Good evening. It's been a fun week with
0:27
all these Apple products,
0:29
>> hasn't it? Indeed, hasn't it? And hello,
0:31
Louiswis. How are you?
0:32
>> Hey, great. I can almost talk. Very
0:33
excited.
0:36
>> Welcome develop. That's a welcome
0:37
development. And we have a very special
0:38
guest today, Christina Warren.
0:40
>> Hey there. Thank you so much for having
0:41
me.
0:41
>> You're welcome. I can't believe you got
0:43
joined, you know, agreed to join our
0:45
crummy podcast, but we're we're very
0:47
very grateful for having you.
0:48
>> It's um, you know, Christina, I have a
0:50
question for you, which I've always been
0:52
very curious about. Where does Film Girl
0:54
come from? Great question. And so it's
0:57
funny because you know a non- diploma
0:59
that you have like kind of a nickname.
1:01
You're signing up for a Twitter account.
1:02
You're in college and you're like, I
1:04
like film. That's what I'm studying.
1:05
That's going to be my username. You
1:07
don't realize it's going to become your
1:08
professional identity and even your work
1:10
email address um across across multiple
1:14
companies. So yeah, it just uh I like
1:17
film. I'm a girl. And it was 2007 and I
1:20
needed a a username uh for for a little
1:22
website called twitter.com. That's it's
1:24
it's it's that it's that simple.
1:26
>> And it's stuck with you. So, you know,
1:28
three decades since
1:30
>> two two let's let's let's not
1:34
let's not make this worse than it is.
1:36
No. Um but no ex Exactly. And it just
1:38
kind of became one of those things where
1:39
at a certain point probably I don't know
1:41
about six or seven years ago I kind of
1:43
looked at I was like should I try to to
1:45
drop that? Should I you know try to you
1:47
know come up with another nickname or
1:48
just go by my name or whatnot. And I
1:50
realized for better or worse, this is
1:51
how a lot of people know me and will
1:53
probably always know me. Even to your
1:55
point, they don't know the origins of
1:56
the name. They just know that this is my
1:58
name. Um,
1:59
>> at a where where I work, we we refer to
2:02
each other by our GitHub handles. And my
2:04
my GitHub handle
2:05
>> has always been film girl. It was before
2:07
I I joined. And so it's it's a frequent
2:09
question. People, but the the fun thing
2:11
about that culture is that
2:13
>> everybody has weird handles for the most
2:15
part. um uh one one of my favorites is
2:18
um um DJ Hamilton Beach and um uh you
2:22
know Jerry which is great and it's just
2:24
that that that's just his his GitHub
2:26
handle right and that's just how you
2:27
refer to people and so there are a lot
2:28
of people who have no idea my you know
2:30
what my background was what I studied in
2:32
school or why that exists but it just is
2:35
>> I was going to say we should we should
2:36
um just talk a little bit about your
2:37
background I mean you know Christina
2:39
you've been a longtime um Apple uh media
2:42
uh reporter worked uh where did you
2:44
start at Um, I was um um I was worked my
2:47
very first job when I was in college. I
2:48
worked at the unofficial Apple web blog
2:50
uh TAW back in the day. RIP. Um and uh
2:54
and then
2:55
>> almost a worse name for a blog than cult
2:56
of Mac.
3:00
>> Not quite though. Not quite as bad. But
3:03
then you went to Gizmodo.
3:04
>> No. Then I was at Mashable. I was at a
3:05
website called Mashable for a long time.
3:07
Um and and then I was at Gizmodo. And
3:09
then um I switched careers and I went
3:11
into um um uh software development
3:13
developer uh relations um and Microsoft
3:16
uh GitHub uh Google DeepMind and now I'm
3:19
back at GitHub.
3:20
>> Okay. And what's your job at GitHub?
3:22
>> So I'm I'm um a a senior developer
3:24
advocate. So I I work on our developer
3:26
relations team. So I uh we do a lot of
3:27
stuff with um you know um uh software
3:30
development, coding, our platforms
3:32
community um that sort of thing. But for
3:34
the last I guess five years, you know,
3:36
AI has really been the the biggest thing
3:39
in the world and and and GitHub copilot
3:41
was one of the first um AI kind of
3:44
coding um tools and assistants and so um
3:48
you know that's not the only thing that
3:49
I that I work on, but that's a big part
3:51
of it is is GitHub copilot stuff.
3:54
>> We're hoping to talk about open claw and
3:56
uh max minis local AI all that kind of
4:00
stuff. We'll get to that in a minute
4:01
though.
4:02
>> Absolutely. I can't wait.
4:04
>> All right, cool. Let's talk about uh
4:05
before, you know, the first thing uh we
4:07
should talk about is this surprise
4:08
announcement from Apple this last week.
4:10
They dropped a bomb. Uh no rumors, no
4:12
leaks, no nothing. And it doesn't even
4:14
have an arrival date, which is even
4:16
crazier.
4:17
>> So, Lewis, why don't you tell us about
4:18
the AirPods Max?
4:20
>> AirPods Max 2. Thank god Apple's using
4:22
numbers.
4:23
>> Uh yeah, new headphones, biggest upgrade
4:26
since they were released in 2020. Uh
4:29
main difference seems to be that it's
4:31
powered by Apple's H2 chip, which is not
4:33
exactly new considering it's been around
4:35
since AirPods Pro 2. Uh but, you know,
4:39
better in all ways. 1.5 times more
4:42
effective noise cancellation, tons of
4:44
other smart features that people with
4:45
AirPods Pro 2 or three already have. Uh
4:49
you know, so that's a decent part of it.
4:51
Um, there's also something about a new
4:53
high I don't hear too many people
4:54
talking about this and if you read
4:55
Apple, you know, new high dynamic range
4:58
amplifier intended to deliver cleaner
5:00
audio without changing the headphones
5:02
acoustic signature. So that's I guess
5:05
that's new hardware must be. Sounds
5:08
cool.
5:08
>> I guess the most the most you know the
5:10
the one exclusive thing is this uh
5:12
lossless audio over a USBC cable.
5:16
>> Well, you could do that with the last
5:17
version.
5:17
>> Yeah, that's not new. That's the
5:19
problem. I
5:19
>> They said something something about, you
5:21
know, you can do you can do uh
5:23
personalized spatial audio.
5:24
>> Yeah, you could do that with the other
5:25
ones, too.
5:26
>> Sorry, we, Oddly enough, while we were
5:29
uh doing the show, you know, show prep.
5:31
Uh I was getting that corrected on the
5:32
website, Apple introduced an update last
5:36
year that that brought that stuff to
5:39
regular AirPods Max. The ones with USBC.
5:41
>> The ones with USBC. Yeah. Not not not
5:42
the OGs, but the the ones that I I I
5:45
bought two pairs. That's it. I won't buy
5:47
a third, but
5:48
>> Oh, wow. Only two.
5:51
new ones.
5:52
>> Absolutely not. Absolutely not. Um, no.
5:54
>> But these ones actually have new
5:55
features.
5:56
>> I know. But but I spent $550 on a pair
5:59
that I really didn't want to buy a year
6:00
and a half ago, and I'm not going to
6:02
reward myself with bad decisions by
6:04
spending even more money. I'm not doing
6:06
it. I'm not doing it.
6:08
>> I don't know. This is Apple. This is,
6:10
you know, there's a world of bad
6:11
decisions. One bad decision after
6:12
another. More and more money.
6:14
>> This is true. But you could buy a laptop
6:15
with this, right? Like that's that's the
6:18
difference. Um. Yeah,
6:20
>> I saw a theory that this, you know, the
6:21
AirPods Max are aren't really a real
6:24
product for Apple. What they are is a
6:25
price anchoring strategy is that they
6:28
it's a it it makes the AirPods Pro look
6:30
cheap in comparison, right,
6:32
>> to the high-end headphones.
6:34
>> Well, they good, better, best.
6:37
>> And and in truth, I mean, this is I am
6:39
glad that they released this like like
6:40
all kind of like jokes aside, um I I
6:42
think one of the joke, one of the
6:44
theories I heard was that they literally
6:45
just don't want to make the H1 anymore.
6:47
And so they just took the same colors,
6:49
the same shells, everything and just
6:51
said, "Okay, we we're going to put the
6:52
H2 chips that we've been making for four
6:54
years now um in in these, which you know
6:57
makes sense." Um know why they couldn't
6:59
have done that um a year and a half ago
7:02
when they released the like the the USBC
7:04
variant, I don't know. I guess they
7:05
probably still maybe had a bunch of, you
7:08
know, H1s lying around.
7:10
That's the only thing I I can think of.
7:12
We do know that Apple development teams
7:14
are kind of small. So, it's possible
7:15
that, you know, around that time where
7:18
they would have been developing these
7:19
AirPods Max 2 a year and a half ago,
7:21
they everybody was working on updated
7:22
AirPods instead.
7:23
>> And then after they shipped the AirPods
7:25
Pro 3, maybe they just it was it wasn't
7:27
until then that they started working on
7:28
these,
7:29
>> which is why they came out a year later.
7:31
>> The bad thing I mean, I'm glad that
7:32
these at least are now on par with what
7:34
you could get for the AirPods. But the
7:36
the the bad thing is specifically about
7:38
the ones that the USBC version which
7:39
again like I'm going to have to clown
7:41
myself cuz I bought them um is is that
7:43
you know like the the sound quality was
7:47
a little bit better um you know just
7:49
because they're they're overear rather
7:50
than in-ear. But in terms of the noise
7:52
cancelling and and some of the other
7:54
kinds of of you know features like the
7:56
AirPods Pro 2os were really really good.
7:58
Uh to the point that to your point like
8:00
as an pricing anchoring strategy, it was
8:03
wor like I the reason I bought the
8:05
second pair. I bought the first pair
8:06
because they were the first time they'd
8:07
come out. I like headphones. I was going
8:10
to buy them. I think that they're too
8:11
heavy. I don't like that they don't have
8:13
a power button. I don't like the case.
8:15
Um I there are a lot of things I don't
8:16
like about them. The reason I bought a
8:18
second pair was because the one thing um
8:21
is that if you live in the Apple
8:22
ecosystem, the instant pairing and
8:25
switching across devices is clutch. And
8:28
even though I have better I have much
8:30
better headphones for traveling. I have
8:31
much better headphones for other
8:32
listening purposes. Like in my opinion,
8:35
sleeping on a on a plane with AirPods um
8:37
uh Pro Maxes is awful. Um or AirPods Max
8:41
is is awful. It's a terrible experience.
8:42
Um Sonos, Sony, um Bose all are superior
8:45
to that. But if you are working across
8:48
multiple devices, say you have two Macs
8:50
and you have an iPad and you have a
8:52
phone, maybe you have a second phone,
8:53
and you want to be able to seamlessly
8:55
pair across those things in the course
8:56
of a workday,
8:58
the battery life is is not great. It's
9:01
actually pretty terrible compared to
9:02
other headphones in its class, but it's
9:04
10 hours, so you can get through a full
9:06
day that way for meetings. What I should
9:09
have done and and this would be probably
9:12
my recommendation to people unless you
9:13
just really like the like the look or
9:15
whatnot was just buy two pairs of AirPod
9:17
Pros and then just switch between the
9:20
two if when you run out of battery
9:22
because that was the only reason I
9:23
bought a second pair. Um but
9:25
>> for the battery life well the battery
9:27
wouldn't last.
9:27
>> Well well yeah well the it's just
9:28
because like I said I would be on so
9:30
many different devices in the course of
9:31
a workday switching between different
9:33
things and I didn't want to have to
9:34
unplug my headphones or do the repairing
9:36
dance. Um, but for, you know, three and
9:39
a half hours isn't going to last you all
9:41
day um on AirPod Pros. What I should
9:43
have done though is just bought a second
9:45
pair of those um and then just switched
9:47
off rather than than buying the the the
9:49
Maxes in my opinion. But mistakes were
9:51
made.
9:51
>> It sounds like you have two of
9:52
everything. You mentioned two Macs, two
9:54
phones, two headphones.
9:56
>> Yeah. Yeah.
9:57
>> It's like a lot. Is that really So, do
10:00
you use two Macs simultaneously? Yeah,
10:02
because I have my my work machine and
10:03
then I have my personal machine and so
10:05
um you know there might be some things
10:07
that uh there for various reasons like
10:08
you're you're doing for you know
10:10
personal stuff something comes in um and
10:12
sometimes you know there are like
10:13
corporate restrictions on things you or
10:15
you can or can't do or whatnot. So you
10:17
have like usually I have like one
10:18
computer plugged into one monitor one
10:20
into another um and then yeah I have a
10:22
work phone and I have a personal phone.
10:24
Um uh I have I have too many Macs
10:27
though. I have a number of personal
10:28
Macs. So, uh, but yeah. But yeah,
10:31
honestly, the switching, but who likes
10:33
it? Come on. This is this is cult of
10:34
Mac. We all do.
10:36
>> That's right. I know. Yeah. Who doesn't?
10:37
Who doesn't? Um, so the one weird thing
10:40
about this, another weird thing about
10:41
these AirPods Max is that there's no
10:42
release date.
10:44
>> And why is that?
10:46
>> I Ed Hardy decoded this, right? It's
10:48
because uh all these features require
10:53
uh what is it? iOS 26.4 four and we
10:55
don't know exactly when that's coming
10:57
out. So, um
10:58
>> there you go.
10:59
>> Well, we do now though, don't we?
11:00
>> Well, I think we have a good guess. I
11:02
can't remember what that good guess is,
11:03
but uh
11:04
>> they they put out their release
11:05
candidate yesterday, I think. So,
11:07
probably next. So, probably 264 is
11:11
coming next Monday or maybe the week
11:12
after that. But it's weird because app
11:14
because Apple says, you know, AirPods
11:16
Max 2 are available to order starting
11:19
next Wednesday, March 25th, and will be
11:21
available early April. But they don't
11:23
they don't say when at all. It was so
11:26
bizarre.
11:27
>> When are they going to get it?
11:28
>> And all the uh sort of downsides that
11:32
you were talking about, Christina, those
11:34
are all still there. Like nothing else
11:35
changed.
11:36
>> No.
11:37
>> Uh they're still heavy. They still don't
11:39
have an off switch, which is to me is
11:42
insane. Yes.
11:43
>> Uh they still have that smart case,
11:45
which Leander, you mentioned that uh you
11:47
actually see people in San Francisco all
11:49
the time wearing uh AirPods. Have you
11:51
ever seen anybody toing around a smart
11:53
case?
11:54
>> I haven't seen it. No.
11:55
>> God, no.
11:56
>> I was going to ask Christina, I mean, do
11:58
you see them in in Seattle at all?
12:00
>> Yeah. Yeah. No, I mean, not not as much
12:01
as in San Francisco or New York. And I
12:03
will say that it's been less as I've
12:05
been like the last year I've probably
12:07
seen fewer of them than maybe I did like
12:09
say two years ago. But yeah, no, you see
12:12
them all the time. People like them. I
12:13
even see people at the gym with them
12:14
which I think is objectively insane
12:16
because these are a they're heavy. B
12:19
they are not good. Like many people have
12:20
had them die because of condensation
12:22
buildup on the inside of the of the ear
12:24
pads. Um so they are not headphones you
12:26
want to wear at the gym at all. Like
12:27
like like that's what the that's what
12:29
the the Beats are for. um or or just
12:31
frankly like you know regular AirPods.
12:33
Um but yeah, no, I mean I see them
12:35
people people seem to like them. I I
12:37
think they're heavy. Um I think that you
12:39
can do better for the money. I do feel
12:42
like if you buy them for people I've had
12:44
people try to justify that to me like,
12:45
"Oh, well they're perfect travel
12:46
headphones. If you traveled a lot, you
12:48
would appreciate them." And then I laugh
12:50
and I go, "Okay, I don't travel as much
12:51
as I used to, but I used to travel about
12:53
100,000 120,000 miles a year." So I know
12:56
airplanes pretty well. And again, you
12:59
want things that will last longer than
13:01
these do for their battery life, which
13:03
they claim 20 hours. It's it's less than
13:05
that. It's about 10. Um, you want
13:07
something that is, you know, lighter
13:09
weight, that is not going to be, if
13:11
you're going to be on your side, if
13:12
you're in a lay flat seat or if you're,
13:13
you know, even if you're in a in a
13:15
regular seat, you don't want it kind of
13:16
knocking around. The the weight, I
13:18
think, is a problem. They're not good
13:20
travel headphones. They're just
13:21
>> They don't fold up like these headphones
13:22
that I have right now. You know, I can I
13:24
can rotate them and fold each cup. You
13:26
can rotate them.
13:27
>> Exactly. Well, and then to your point
13:29
about the case,
13:30
>> I lost the case of of my of my first
13:32
pair.
13:33
>> And the pro Right. Well, right. And and
13:36
and you if you don't want to spend $60
13:37
or $70 or whatever that the fee is to
13:39
get another one. The problem is people
13:42
say, "Oh, well, because you don't need a
13:43
power button, they'll they'll just
13:45
automatically cycle to low power."
13:46
That's not really true. The the the
13:48
magnet will put them into a deeper sleep
13:50
mode and let them at least hold a
13:53
battery longer. But if you don't have
13:54
them in the case and you're just using
13:56
them outside of the case, you're gonna
13:58
have to charge them every single day
14:01
because when you need them, they will be
14:04
dead. And and and because that that that
14:06
that low power mode is is just give us
14:09
an off switch. Like it I I I I fully
14:12
understand for for AirPods for the in
14:14
ears not having a switch. I think that's
14:15
a good decision, but you've got all this
14:17
real estate. Just give us an off button.
14:19
It would be so much better for everyone,
14:21
right? like you I'm not even saying that
14:23
you couldn't even have some sort of
14:24
sensor on where you could maybe detect
14:26
something to automatically turn on or
14:29
whatever the case may be. But but I
14:31
think even that tradeoff most people
14:33
myself I would certainly much rather
14:35
have to press a button and and and power
14:37
it on the way I do with every other pair
14:39
of headphones than have to say oh I
14:42
haven't used these in three days.
14:44
They've been in the smart case. Let's
14:45
hope that I still have, you know, some
14:47
battery life and I don't have to charge
14:49
them for five minutes before I can
14:50
actually, you know, listen to music.
14:52
>> I have like a meat thermometer in my
14:53
kitchen that turns itself off when you
14:55
fold it up and like, you know, collapse
14:57
the little metal metal prong thing. If
14:59
AirPods Max folded up, then they could
15:01
just do that. Like, oh, as soon as you
15:02
turn the ears, clearly it's not on your
15:04
head.
15:05
>> Oh, that would be brilliant. Yeah. No,
15:07
you're you're dead on. That's all they
15:08
have to do. And and then they would
15:09
actually be portable, right? To your
15:11
point, it' be like the headphones you're
15:12
wearing. You could actually fold them
15:13
up, put them in your bag. Cuz that's the
15:15
other problem, too. Like, they take up
15:16
so much. Like, I always have them around
15:18
my neck almost all the time when I'm
15:20
wearing them. And it's not even a
15:21
fashion statement. It's I don't have any
15:23
room to put them anywhere else. I'm not
15:24
going to take up space in my bag, you
15:27
know, for the headphones.
15:32
>> Yeah, you should. I mean, in my opinion,
15:34
that that's that would be the that's
15:36
what I would do. And I think you
15:37
instinctively kind of do that. There
15:38
there are two buttons. There's the
15:39
digital crown and then there's um
15:41
another button um that cycles through um
15:44
the the different modes or for pairing
15:47
or whatever. So, they do have buttons on
15:48
them. That's the dumb thing. It's just
15:50
you can't power them up. It's it's like
15:51
the studio display. Why can't Why do I
15:53
have to have a smart um uh uh plug? I
15:57
literally have this to power cycle my
15:58
monitor. I have to have a smart plug
16:00
that I have connected to Home Assistant
16:02
and that way I can I can basically issue
16:04
a command and it will turn my it'll
16:06
power cycle my monitor. That's so dumb.
16:08
That's so dumb. Just give me a power
16:09
button.
16:11
>> Did you ever try that hack where um I
16:13
saw someone put their a pair of AirPods
16:16
Pro inc mode in their ears and then they
16:19
put over them a a pair of AirPods Max
16:23
also in ANC mode. So they had double,
16:26
you know, active noise cancelling. I was
16:28
always kind of curious like I I I I'm
16:31
sure it won't work, but I'm curious to
16:33
know if it does.
16:34
>> I don't know.
16:34
>> I guess I should try it. I I mean that
16:37
that's that sounds dumb, but that also
16:39
sounds like the sort of thing I would
16:40
do. I mean, hey, if you really wanted to
16:42
block out all the noise, I guess you
16:44
could do that, right? Because because I
16:45
mean I I in addition to them probably
16:49
running out of H1 chips, it was a little
16:52
embarrassing that you've had like two
16:53
successive versions of the AirPod Pros
16:55
that have had better noise cancelling
16:57
than the than the $550 AirPods Max's,
17:00
right? which has been a problem because
17:02
the the Pro Twos I think were just
17:04
probably one of the best like in-ear
17:05
headphones ever to exist and and then
17:08
the threes are better noise cancelling.
17:10
I don't like the sound signature as much
17:11
but you know the battery life is better.
17:12
There are other benefits there and then
17:14
it's like okay you're now going to be um
17:17
like you're but but your flagship like
17:19
the whole reason that people buy them is
17:21
because oh these are the best. They're
17:22
the most expensive. They're the biggest.
17:24
No. Now, at least, I mean, I still would
17:27
tell people you you can do better unless
17:29
you really love that instant pairing or
17:31
you just really love the design. I think
17:32
you can do better for your money, but at
17:33
least now it's not a complete
17:35
embarrassment.
17:36
>> Every time somebody would say to me, oh,
17:38
you know, cuz I have like one friend who
17:40
has AirPods Maxes, and so like other
17:42
friends would occasionally say to me,
17:43
"Oh, you know, I kind of want some of
17:45
those, too." And I would say, "Oh, well,
17:47
you know, I know you're not like a major
17:49
audio file. You probably wouldn't be
17:51
able to appreciate like the incredible
17:53
like sound system thing. you should just
17:54
get like, you know, you can get like
17:56
Sony or Bose headphones for like half
17:57
the price. They'd be good enough for
17:59
you. And they just kind of like go quiet
18:01
for like a week or two and then they
18:03
say,
18:04
>> "I'm kind of interested in those AirPods
18:06
Maxes."
18:07
>> Yes. Yes. I I Yes. I've had the exact
18:10
same thing.
18:10
>> Are you Are you tempted to buy them,
18:11
Louis? I mean, because you like them,
18:13
don't you? Would you, if you had the
18:14
money, go get them?
18:15
>> I mean, you know, I reviewed them for
18:17
Cult of Mac when they came out. Yeah. I
18:19
I I thought they sounded awesome. uh you
18:23
know they were comfortable enough. I I
18:25
never I you know I wrote the review you
18:27
know the day they came out or something
18:28
right I didn't have like a month and and
18:30
I gave them right back to you Leander so
18:32
I never traveled with them. Uh so I
18:34
don't really know about that aspect of
18:36
it. Um that's about the only time I that
18:38
and this podcast is about the only time
18:40
I ever wear big headphones overear
18:43
headphones because I have you know
18:45
AirPods Pro 2 in my pocket. I've got
18:48
like a skull ring and my jeans from
18:51
where my AirPods Pro 2 thing is. Uh I
18:54
carry them with me everywhere, you know,
18:55
and I use them every single day for
19:00
hours and hours, multiple things. Uh you
19:02
know, whether it's listening to podcasts
19:04
or taking phone calls. I mean, I swear
19:06
to God, every time the phone rings, I'm
19:08
scrambling for the AirPods. Put them in
19:09
my ear because it's so much more
19:11
convenient, so much easier, sounds so
19:13
much better. So, I I absolutely love
19:14
AirPods Pro. I I mean, if I had all the
19:16
money in the world, sure, I'd buy some
19:18
AirPods Max, but uh it's it's not I I
19:23
just don't sit around and listen to
19:24
music with headphones. I I just don't do
19:26
it.
19:26
>> Uh the only time I would wear them would
19:28
be on flights and uh even then, I mean,
19:32
I have some overear headphones, noise
19:34
cancelling ones that I would take on
19:37
overseas trips and, you know, it's kind
19:39
of worth it. But then again, you know,
19:41
just having the AirPods Pro, they're
19:43
really good and they fit in my pocket.
19:46
So,
19:47
>> I it's and even even headphones that do
19:50
fold up take up, you know, a fair amount
19:52
of space if you're trying to travel
19:53
light. So,
19:54
>> yeah, that's that's the one thing about
19:56
the the Max is just they're just too big
19:59
and bulky, you know, like the whole
20:01
thing. And especially you get that case
20:03
which I you know looks aside I mean it
20:06
just seems like the one of the silliest
20:08
things I've ever seen you know like
20:09
protects
20:10
>> I don't know I quite like very minimal.
20:13
>> Well but but that's kind of the problem
20:14
like you like the the the band on the
20:16
top um if we're airing our grievances on
20:18
on the Maxis since they changed since
20:20
they changed nothing with them and since
20:21
I've bought them twice and and so again
20:23
I I keep stressing this because I'm
20:25
dumb. Like I'm not a bright person when
20:27
it comes to this. like I really did just
20:28
reward bad behavior with like bad more
20:30
bad decisions is that the the mesh um
20:33
kind of loses it its shape a little bit
20:35
if you wear them. Um, but what I've
20:36
always been worried, but what I've
20:38
worried about,
20:39
>> um, and I haven't had any punctures or
20:41
anything like I I think it is strong
20:42
enough for that. But yeah, this is like
20:44
my issue with the case. Minimal is fine,
20:46
but you're protecting, you know, like
20:48
the the the ear cups fine, which is
20:50
okay, but then you have the the top of
20:52
it like this this not super rigid, you
20:54
know, like mesh structure. Like I've
20:56
always been half afraid that I have them
20:58
in my bag. if you have, you know, say
21:00
even an Apple Pencil or something that
21:02
hits it the wrong way, you know, it
21:03
could slice through the top and and cut
21:05
into that mesh, you know, it just it
21:07
just feels like at least
21:09
um Sony when they they redesigned their
21:11
their um flagship overear headphones a
21:13
couple of years ago and they made them
21:14
less f foldable, which was not a good
21:16
thing, but the case at least was still
21:18
fairly minimal. And and Sonos's
21:20
headphones, which um I did not pay full
21:22
price for. I would not have paid full
21:23
price for, but for the price I did pay,
21:25
I think they were they were excellent.
21:26
They don't fold up super small, but at
21:28
least it is like a full, you know, case
21:30
that you can um uh it's it's relatively
21:34
thin that you can kind of fit in it. It
21:35
protects the whole body. But like the
21:37
AirPods, it's just like, okay, I'm now
21:39
I'm just going to have these loose in my
21:40
bag and and now I'm worried about them
21:42
kind of getting, you know, jumbled
21:44
around and
21:46
nothing's ever happened. And and I've
21:48
worn them many many many times all over
21:50
the world. Although, like I said, I
21:51
usually wear other headphones on on a
21:53
plane because they're just better
21:54
options for that. But yeah, I just it's
21:57
it's I I hate that they they had they're
22:00
calling them two and it's like we
22:02
changed the chips so that they are now
22:04
actually not outdated. So that's a good
22:06
thing. So So you like the design and if
22:08
you want and you have the money for it,
22:09
I think this is actually a much better
22:11
deal than the ones that existed last
22:12
week. But I I don't know. I I feel like
22:17
if if if I were to do it all over again,
22:19
I would just buy a second pair of of
22:21
regular AirPods and cycle through them
22:23
um for that purpose. Um because you
22:26
could do that. You could get two of
22:27
those for for one pair of AirPod Maxes.
22:29
Really, I think the real thing to do
22:30
would be like buy a pair of AirPods and
22:32
get a really nice pair of o overear
22:34
headphones on sale from from Bose or
22:36
Sony. Um
22:38
>> well, people are praising the Soundcore
22:40
ones. Um what are they called? The
22:41
Soundcore uh something pro.
22:43
>> Space Pro.
22:44
>> Space Pro. Yeah.
22:45
>> Yeah, they're good. I mean, I I don't
22:47
think they still, you know, I I can't a
22:50
them because I don't have the AirPods
22:51
Max anymore, but uh I I don't think the
22:54
sound is is good. You know, it's to me
22:57
there was something very special and and
22:59
lively about the sound of those AirPods
23:01
Max. It made the music just really come
23:02
alive. Uh and and I value that and it
23:06
was great. But, you know, you can pretty
23:08
much
23:08
>> That's that's kind of feature number one
23:10
of for a pair of headphones, isn't it?
23:11
>> Well, I don't know. I mean, again, when
23:14
was the last time you sat around in your
23:16
house and listened to music? I I don't
23:17
do it. I I never do it. I
23:20
>> I mean, again, the problem with every
23:21
other pair of like regular Bluetooth
23:22
headphones is that, you know, oh, I have
23:25
to go through this whole dance of like,
23:26
you know, like tapping and holding a
23:28
button and god forbid I want to switch
23:30
it from my iPhone to my Nintendo Switch,
23:32
but then it just keeps automatically
23:33
pairing to my iPhone and I need to
23:34
figure out a way to like, oh, I'm just
23:36
going to like forget the device and I
23:37
have to do do the whole song and dance
23:39
again. And what battery percentage is it
23:41
at? It doesn't have a screen. It's a
23:43
pair of headphones. So, you just have to
23:44
wait for them to die and then it'll
23:46
interrupt your music and say battery low
23:47
like every two minutes. And
23:49
>> can't you put a you could put a widget
23:51
on your phone, can't you? To to show the
23:53
battery
23:53
>> for the AirPod for the AirPods Max, but
23:55
not for you know whatever Bluetooth Sony
23:58
can no yeah. Yeah. Um it it's in the
24:02
same section. Um even without the H1
24:04
chip, it's just in the same like
24:05
Bluetooth section as anything else. It's
24:07
not going to maybe be in your Apple
24:08
battery widget, but if you open up like
24:10
the Bluetooth area like like my Sonos's
24:12
for instance, yeah, I can see um you
24:14
know what the battery life is on those.
24:16
And then they have a they have an app
24:17
which you don't have to use their app,
24:18
but you can you know use use that app if
24:20
if you need to see something um as well.
24:22
But no, you can you can see the battery
24:24
percentage. The the nice thing um about
24:27
those though is like again there's an
24:28
off button and so I can have those
24:30
powered off for like a month and a half
24:31
and I can turn them on and it's like,
24:33
"Oh, you still have 40% battery." And
24:34
I'm like, "Oh, excellent." Yeah, I guess
24:37
it varies by Bluetooth heads because
24:39
I've had some before that definitely
24:40
don't warn you until it just starts
24:42
interrupting your podcast and
24:44
>> Right. And and then you're like, "Okay,
24:45
I'm I'm I'm at a loss." Yeah. No, you're
24:47
right. I mean, there are some very nice
24:48
quality of life things, especially if
24:50
you are almost entirely in on Apple. I
24:52
will say where the the multi-pairing
24:54
thing falls down if you are wanting to
24:56
use it with something like a Nintendo
24:58
Switch or if you're wanting to use it
24:59
with, you know, if you have a a Windows
25:01
device or a Linux device or or, you
25:03
know, something else um that that once
25:06
you want to, you know, pair your
25:07
headphones with, then that like instant
25:10
pairing aspect of the of the AirPods
25:12
goes away a little bit. But um I don't
25:15
think that's usually the target market
25:16
for for I think most people who buy
25:18
AirPods Max are either heavily in the
25:20
Apple ecosystem are buying it for the
25:23
fashion component, whatever that may be.
25:26
Um or or just, you know, and and and
25:29
that's really what it is because you're
25:30
you can get better sounding headphones
25:31
for the money. You can get better
25:34
practical headphones for the money, but
25:35
there is something unfortunately about
25:38
where they fit in kind of that sweet
25:39
spot where like if you have the money
25:40
and you use a lot of Apple devices and
25:43
you want overear headphones and you're
25:45
okay with all the tradeoffs, you know,
25:48
>> $550.
25:50
>> Do Do you use uh spatial audio? Do do
25:53
you guys ever listen to special audio?
25:55
>> I do, but I turn head tracking off.
25:58
>> Oh, you do? Why?
25:59
>> Because it changes Well, it depends. So
26:02
sometimes like cuz cuz I do like how
26:03
Apple Music um I mean all the music
26:05
services have done like the spatial
26:06
audio stuff but but Apple Music I think
26:08
has done a fairly good job. Sometimes it
26:10
depends on like if they've added it
26:12
after the fact. I feel like it I don't
26:14
like the effect of the the you know
26:16
headtracking aspect with it. Like if I'm
26:18
watching TV or something I think it's
26:19
fine. Um it can be good if I'm listening
26:21
to music. Um I I don't necessarily like
26:24
how sometimes I guess the the sounds
26:26
sound um in my ears by moving things
26:28
around. So, I usually have it like a
26:30
fixed track, but I do like the simulated
26:33
surround sound for sure.
26:34
>> Yeah, it's it's a it's a it's a it's a
26:36
it could be really striking. It's a very
26:38
striking effect, but I'm not really sure
26:39
what it adds to the experience.
26:41
Uh, you know, like is it better uh to
26:44
have it on or or not. I I'm kind of
26:46
undecided. Um, you know, I think you
26:50
tend to kind of I the one thing I like
26:51
about it is it it makes you listen to
26:53
the music. It makes you concentrate on
26:54
the music. um which you know uh is I
26:57
think a huge benefit because you can you
27:00
know often if you
27:03
it could be like background isn't it you
27:04
know but the spatial audio because
27:06
you're listening to that effect
27:08
>> makes you pay attention to what you're
27:09
listening to. Yeah, I think it totally
27:11
depends on the mastering because there's
27:12
some albums that I'll listen to that I
27:13
can tell they just probably put it
27:15
through some sort of filter to add
27:16
support for it and it takes me out of it
27:18
because the songs don't sound the same
27:20
that I was expecting them to sound
27:22
sound. And maybe that was mastering,
27:23
maybe it's just you're so used to
27:24
hearing a song a certain way that that
27:26
it it's just off. But then there are
27:28
some where like I can tell, okay, the
27:30
the mastering engineer did a really good
27:31
job and they did make this sound more
27:33
enhanced and so this does, you know,
27:35
sound like a more surround sound type of
27:37
experience. like if I had a really nice,
27:39
you know, um audio setup because um
27:42
there was a there was a format called
27:43
like a DVD audio that really didn't work
27:45
um that kind of tried to introduce like
27:47
multi- channelannel, you know, to to
27:49
regular CD stuff and and we haven't
27:51
really had that since then. It's been
27:52
primarily, you know, just just just two
27:54
channel audio. And so I think the effect
27:56
can be really good. I think for movie
27:58
watching, you know, stuff it can be
27:59
really really good like for the the
28:01
Dolby stuff. Um but I think it just
28:04
depends on how it's implemented. Mhm.
28:06
>> We've come a long way from the Beatles
28:07
just deciding, let's put George and
28:09
Ringo on the left and John and Paul on
28:11
the right for just no arbitrary no
28:14
reason whatsoever.
28:15
>> God. Yeah. Well, so they could so they
28:17
can sell multiple copies.
28:18
>> Mhm.
28:21
All right, let's talk about let's talk
28:22
about Open Claw. Um uh you know it's
28:26
this this uh Jensen Wang I think he gave
28:29
a keynote speech to me um last week and
28:31
he said that this is open claw running
28:34
on a Mac mini or well he was actually
28:36
touting in the cloud but he's saying
28:37
this is the personal computer this is
28:39
the paradigm the modern paradigm this is
28:41
what the personal computer is going to
28:42
turn into and of course Christina GitHub
28:43
you know this is the most popular
28:44
project on GitHub ever yeah
28:47
>> yeah I mean it's been crazy so I mean I
28:48
think that like I think um
28:51
uh Pete started it in um I think
28:55
December and and January started really
28:57
gaining momentum and um yeah it's the
28:59
most popular project in GitHub history.
29:01
I mean and and just the the number of
29:02
people who are using it uh is it's blown
29:05
away. I mean I remember looking at it uh
29:07
when the the name changed. So originally
29:08
it was called um um uh uh something else
29:12
and uh because of Claudebot. Yeah. And
29:14
because um Anthropic didn't like uh uh
29:18
that then then it became um something uh
29:21
it went Claudebot then it went
29:24
Then it was moltbot uh and and then um
29:27
openclaw. And so it's had a couple of
29:28
name changes. I think openclaw is a been
29:30
a a good one, but it's definitely um I
29:33
mean so the whole idea for anybody who's
29:35
not familiar is that basically this is a
29:37
system that um uh someone that's kind of
29:41
hooked up where you can connect um an
29:43
LLM of your choice. It could be locally
29:45
running on your machine or it could be
29:46
uh one of the models from OpenAI or
29:48
Enthropic or or anybody else. Um you
29:51
know Google um and you can basically
29:53
connect a chat interface to it using
29:56
iMessage if you're running it on a Mac
29:58
mini or you could also use Telegram or
30:00
WhatsApp um through bridges and other
30:02
things and you can basically kind of
30:03
communicate with your agent and say I
30:04
want you to do these tasks for me and
30:06
you're giving it full access to your
30:08
system or whatever access you're wanting
30:10
to give it to. This is why most people
30:11
are using dedicated machines. And it can
30:13
then do a genetic tasks. Um, so it
30:16
could, you know, do research for you. It
30:17
could sort through your email. Um, it
30:19
could start, you know, uh, uh, creating,
30:21
you know, uh, projects for you for
30:22
software development. Um, based on what
30:25
you're working with it back and forth
30:26
and then it will kind of give you the
30:27
results. You can kind of check in
30:28
remotely on what's happening. Um, and so
30:31
it's it's uh, very very cool. It can
30:34
also get very expensive because in
30:36
addition to you don't need to have a Mac
30:38
Mini for it. A lot of people get a
30:39
dedicated Mac Mini because they're
30:40
inexpensive and they want to isolate the
30:43
the uh AI agent from the rest of their
30:46
work and say, "Okay, well, if if this
30:48
goes a foul, if something goes wrong, um
30:50
all my personal documents and everything
30:52
else won't be impacted." Um although,
30:55
you know, there could still be security.
30:57
>> There was that woman from uh Facebook,
30:59
right? She she let it loose and it
31:02
deleted half of her emails. Yeah, I I
31:05
think she stopped it early enough. But
31:06
yeah, I mean that's kind of the problem
31:07
is that, you know, these these things
31:09
will work really well and then I can
31:10
kind of get started and I'm on a task
31:11
and you have to kind of be like abort
31:13
abort, you know, stop stop what you're
31:14
doing uh before it goes too far. Uh so
31:16
you want these things to be um able to
31:18
be more in a sandbox environment where
31:20
it will, you know, test things um first.
31:23
Uh there there are a lot of potential
31:25
security problems depending on on how
31:27
you if you just let this run loose, but
31:28
it can be really really um fun. And then
31:31
the nice thing about it is that like it
31:33
remembers your preferences and it
31:35
remembers, you know, the things that you
31:36
like to do. Um, and so it it's um
31:40
there's a huge community for it. Not
31:42
just, you know, the number of stars on
31:44
on GitHub. I think it let me look at the
31:45
number. It's it's 325,000, which is just
31:49
ridiculous, especially for a project
31:50
that's a couple of months old. And then
31:53
um you have a huge ecosystem of people
31:55
building plugins and and other stuff for
31:57
it. And um it's uh now like you know um
32:02
um Pete uh got a job at at at OpenAI. Um
32:05
there have been uh there was a a social
32:08
network uh called Claude um a molt book
32:11
uh MT B K which was like a social
32:14
network just for the agents uh to talk
32:17
too didn't it?
32:18
>> Yeah that um those two um uh guys were
32:21
were aqua hired by by Meta um I think
32:23
two weeks ago. Um, and so there's this,
32:25
you know, uh, Nvidia announced, um,
32:28
their own Nemo Claw, which is like a a
32:30
secure, what they're calling a secure
32:32
kind of cloud hosted version of Open
32:34
Claw. Um, Perplexity is doing something
32:36
very similar, um, with their computer.
32:38
So like there are a lot of uh, people
32:41
who are really, you know, keen on this
32:44
idea. Um, but it can get expensive
32:46
because if you're using it with the the
32:49
host of models, especially the anthropic
32:51
models, which are very very good, it can
32:53
get expensive really quickly. So, what a
32:54
lot of people do is they have like their
32:56
or what they were doing, Anthropic has
32:59
cut this off is they would use their
33:00
$200 a month clawed max plans with
33:03
OpenClaw, but um but Anthropic is like,
33:06
"No, we don't want you to do that." um
33:07
OpenAI is very happy to let you use your
33:10
your $200 a month um uh chatbt, you
33:13
know, codeex plan with it, but the
33:15
results aren't going to quite be as
33:16
good. But, um the the downside with some
33:20
of these models is it can be very
33:21
powerful. Like you can literally just
33:22
say, you know, um go through my emails,
33:24
find the the the 10 most important
33:26
unread things and and write, you know,
33:28
uh create a a summary of what needs to
33:31
be done um or or respond, you know, to
33:33
to to these people um with this task. um
33:36
and and check in and then you can kind
33:38
of get, you know, status updates about
33:39
what it's doing or do a research report
33:41
on I'm going to give you a list of these
33:42
links. I want you to do a report on this
33:44
and then and then come back to me. Um
33:46
and that's really powerful and you can
33:47
do really great things with it, but you
33:49
don't know exactly how much all of those
33:52
system calls are going to cost. And so
33:55
it can it can you know it can get
33:58
expensive,
33:58
>> run up a pretty hefty bill. Yeah.
34:00
>> Yeah. Yeah. I think Feder Rico when he
34:02
wrote like the for Max stories I think
34:03
like he and this was two months ago I
34:05
think at this point you know he spent
34:07
you know over $1,000 probably you know
34:10
just in in AI credits and I don't know
34:11
if if he had to spend that himself or if
34:13
he had something else you know offered
34:15
to him that he could get out of that but
34:17
>> and I think at the end he he ended up
34:19
disconnecting his email from it because
34:20
he was scared that I have an open email
34:22
address on the internet I don't want
34:24
somebody to prompt inject open claw by
34:27
like sending me an email that then has a
34:30
my files reply with all my credit card
34:32
information.
34:33
>> No, and which which is a great point.
34:35
Yeah, I mean this is this is the
34:36
interesting thing is that like you have
34:37
this whole ecosystem around security and
34:39
and obviously you know the project
34:40
itself like they're trying to take in
34:42
patches and try to make things more
34:43
secure but then you have like ironclaw
34:45
which is like oh this is a more secure
34:46
version of openclaw and then Nvidia is
34:48
like well we have this hosted thing and
34:49
and you have some of the you don't have
34:51
to use a Mac you could use any sort of
34:53
you know um isolated environment you
34:55
could use a you know um a Raspberry Pi
34:58
um if you wanted to um but uh but but
35:01
the the Macs are are great if you want
35:03
to give it access to all your files. But
35:05
yeah, Federico's point.
35:07
>> Great power comes great responsibility.
35:08
I have it running. I do not give it
35:11
access to my email for those same types
35:13
of reasons. I'm like there I I don't
35:15
want somebody to be able to find some
35:16
sort of flaw in something that I've
35:18
misconfigured and find a way to prompt
35:20
and then all of a sudden now they have
35:22
access to all my emails.
35:23
>> There was a report that uh hackers are
35:25
using invisible unicode to prompt uh
35:30
machines. Only the machine can see it.
35:31
Even if you
35:33
>> carefully pass it and read uh whatever
35:35
message has been given, they're hidden
35:37
uh in in Unicode. So only the machine
35:39
can see it. And even pass some kind of
35:42
uh checks too, some machine checks you
35:44
can run it through. I'm not sure the
35:45
details, but it seemed very nefarious.
35:48
But what about uh Mac hardware? You
35:49
know, like Mac uh Apple Silicon uh you
35:53
know, they've done some amazing stuff,
35:55
you know, and there's all this talk
35:55
about Apple losing the AI race, but it
35:57
seems to me that that's not true at all.
35:59
Well, I think it's interesting because
36:00
there's like I guess two different
36:01
components here because on the one end
36:03
like Apple definitely in terms of their
36:05
own models is not even part of the
36:08
conversation, right? Which is this is
36:09
why they've they've had to have a
36:11
partnership with with
36:12
>> Right. Well, right. But this is now they
36:14
had to have the partner they've signed a
36:15
partnership with with um uh uh DeepMind
36:18
um to to run Gemini um and make that
36:20
kind of their power um provider. And
36:22
obviously I think they're probably still
36:23
working on their own frontier models
36:25
internally. How that will work, we don't
36:27
know. But the hardware itself because
36:29
there are all these openweight models um
36:32
that uh you know from from various
36:34
companies. Um Google has some um uh Quan
36:37
is a big one. Um uh Mr. Lake there are a
36:40
number of you know companies out there
36:41
who are doing these things that can be
36:43
pretty powerful and can um you can get a
36:45
lot out of them. And if you go to
36:46
HuggingFace, uh, which is like a a great
36:49
resource just to find people who've
36:50
tuned local kind of customized AI models
36:53
that you can run in a cloud or on your
36:55
own machine. Um, one of the best ways
36:58
kind of out of the box without having to
37:00
configure things to get started with
37:02
that sort of stuff is to get a Mac Mini
37:03
or a Mac Studio or even a MacBook Pro.
37:06
uh and um you know because uh GPUs are
37:10
are expensive and they can be more
37:12
powerful but for what a lot of tasks a
37:14
lot of people want to do um what's great
37:17
is yeah I think you're right like Apple
37:19
has maybe they're behind in terms of
37:20
their own models but the the tooling for
37:23
the things you can actually do in within
37:25
Mac OS um is is actually pretty robust
37:29
because there's so many people who love
37:31
the hardware for other reasons and
37:32
they're like okay we want to figure out
37:33
how we can get the the custom AI models
37:36
to run on these things as well as
37:37
possible. And so that's opened up a
37:39
whole new world.
37:40
>> Well, last night I was reading about a
37:41
guy who got the Quen 3.5, which is a
37:43
Frontier model that was released, I
37:45
think, back in February, running on his
37:46
MacBook, which is astonishing. And he
37:48
said that he did it uh he was able to do
37:50
it partly because of the the
37:51
architecture of the model. um it's it's
37:54
built in such a way that it um it's I
37:56
think it's a almost a 400 billion
37:58
parameter model but it's architected in
38:01
such a way that when you're using it it
38:03
only uses 17 billion parameters which
38:05
apparently can be
38:06
>> and so he said that Apple silicon
38:08
because of uh the the architecture of
38:11
Apple Silicon the way that the GPU and
38:13
the CPU and the SSD controller all share
38:16
the same memory he's able to pull all
38:19
the data it needs off of the SSD
38:22
at like 14 gigabytes per second. It's
38:25
fast enough to be able to feed the
38:26
machine off the SSD to run a Frontier
38:29
model on a MacBook. And uh and
38:33
he's saying that it's the caching as
38:34
well, the caching system in the the the
38:37
uh on the on the chip
38:39
>> um enables this. And he said it was all
38:41
kind of like a a it was a a culmination
38:44
of all these different things that abled
38:46
enabled him to do this project. On the
38:47
one hand, it was Apple silicon. On the
38:49
one hand, it was this highly optimized
38:50
model. And the third thing was um
38:53
Kapathi loop being he he he just uh got
38:58
called code to run its own experiments
39:01
to create its own software running
39:03
experiments trying different things to
39:05
see what worked and he came back in the
39:07
morning and built him this system that
39:09
ran this Frontier model on his MacBook.
39:11
It's amazing the fact that yeah, you can
39:12
do this for relatively um I mean, you
39:14
know, RAM prices are only going through
39:17
the roof, but at least right now, at
39:18
least on the Mac side, like relatively
39:19
inexpensively, you can do all this and
39:22
yeah, to your point, the fact that
39:23
you're able to use the models themselves
39:25
to create their own test cases and their
39:27
own iterations and improve themselves,
39:29
that's that's been the biggest thing I
39:30
think over the last, you know, four or
39:32
five months that we've seen with like
39:33
the way that the AI models have improved
39:35
so much is that the way that they're
39:37
able to improve themselves and kind of
39:39
do things more autonomously
39:41
um and get really good results has
39:43
really really improved. Uh and uh yeah,
39:46
the fact that he was able to get like a
39:48
model that size running on a MacBook is
39:50
is is wild and and is incredibly
39:53
impressive. Um and yeah, a lot of that
39:54
is because of the hardware because they
39:56
have that high memory bandwidth. They
39:58
have the the fast enough SSDs, they have
40:00
the good caching. Um you know, the way
40:02
that the system handles, you know, the
40:03
unified memory works really well. I have
40:05
I have a framework desktop which uses an
40:07
AMD stricks Halo machine which also has
40:09
unified memory. Um but its memory
40:11
bandwidth is not as is not as good um as
40:13
as what you get on say like a Mac
40:15
Studio. I think it's it's similar to
40:17
what you might get on a Mac Mini. Um and
40:20
uh even though it's unified memory just
40:22
because of how various operating systems
40:24
deal with memory like you can allocate
40:26
like of the 128 GB of RAM you can
40:28
allocate like you know 96 or something
40:31
to to just the AI models. So, it's not
40:33
quite as seamless as what you can do on
40:36
the Apple silicon stuff. Um, and then,
40:38
you know, again, because so many people
40:41
love their Macs for so many other
40:42
reasons, they're going to be a whole
40:45
kind of ecosystem of of people kind of
40:46
figuring out ways, okay, how can I get
40:48
this model running? How can I distill it
40:50
down and maybe make it, you know, um,
40:51
parameize it so I can make it smaller
40:53
and get the the the big benefits out of
40:55
it? There's a there's, um, a group
40:58
called Unsloth, Unsloth.ai, AI and they
41:01
really work hard on being able to get
41:03
the the model sizes um uh uh quant uh
41:07
parameized down so that you can uh have
41:09
a much bigger model that can run um in
41:12
in less space um and still run very
41:14
efficiently and um like they they have a
41:17
guide for how to how to run Quinn 3.5
41:20
locally um and so and and Quinn 3.5 uh
41:24
the 35 uh billion and 27 billion moni um
41:27
models work on a 22 gigabyte uh Mac RAM
41:30
device and like you know that mean you
41:33
so you a MacBook Air I wouldn't
41:35
recommend it on a MacBook Air because it
41:36
doesn't have a fan. Um, so it'll
41:38
throttle at a certain point, but you but
41:39
you you could on a MacBook Air run like,
41:43
you know, one of the latest like Quinn
41:44
series models, which is really
41:46
impressive depending on.
41:47
>> Do you think that a big advantage Apple
41:49
has is that Apple only makes like four
41:52
computers? And so it's really easy to
41:54
develop for, you know, the Mac compared
41:56
to like the infinite number of ways that
41:58
you can build a PC just because like
42:00
there's a limited amount of hardware
42:01
that you need to test on.
42:02
>> Yeah, I I think that's part of it. Um, I
42:04
will also give Apple credit for for MLX,
42:06
which is their kind of interface um for
42:09
being able to um access um uh AI models.
42:13
And that this was this basically it's
42:14
like um like a a machine learning um
42:16
kind of like framework. And when Apple
42:19
released this, I guess it was two and a
42:22
half years ago, um a lot of us were
42:24
surprised because this is not a thing
42:25
that it's on GitHub. It's open source.
42:27
That's not typically a thing that that
42:29
Apple does. Um but they did release it
42:32
and um a lot of people were able to then
42:34
quickly take the existing AI models and
42:37
either add MLX support and now Apple has
42:39
started to add you know do more work to
42:40
add that support there too. So I think I
42:42
think you're right. I think part of it
42:43
is the fact that there are fewer devices
42:45
so there are fewer targets like if
42:47
people really want to build the tooling
42:49
you know all the chipsets families have
42:51
a lot of similarities. If we target M1,
42:53
M2, M3 so on we we can get similar
42:55
results. Um but even then I mean like uh
42:58
we were talking about this before the
42:59
show. Nvidia for many many many years
43:01
had uh almost like a a monopoly on the
43:05
type of AI tooling when it comes to
43:06
training and and running sort of data um
43:09
with a interface they have called CUDA
43:11
cuda and um AMD has been trying for for
43:15
years to to get people to kind of build
43:17
similar tooling for for their GPUs
43:19
because their GPUs are are pretty good
43:22
but it just hasn't happened. Um, and so
43:24
even when you have really powerful
43:26
integrated hardware like the stricks
43:28
Halo,
43:29
>> the tooling just, you know, it it it's
43:32
not quite there. It's better in some
43:33
ways in Linux, some ways in in Windows,
43:35
but like you it's hard to kind of find
43:37
that that that sweet spot. Whereas
43:38
Apple, I think to your point, Griffin,
43:40
like there's one chip and there are
43:43
millions and millions and millions of
43:44
them. So, I think that it's it's
43:46
probably a combination of fewer SKs and
43:48
then the fact that you have like an
43:50
order of magnitude more of those
43:52
machines available than you do for what
43:55
those other configurations are, right?
43:57
Because even if you had like if if if
44:00
AMD was was was killing it, like they're
44:02
not going to be selling as many units
44:04
with like the exact same type of, you
44:06
know, chip architecture in the wild as
44:09
as Apple does, right? Like just in terms
44:11
of main consumer things. So
44:14
where if if you're a developer, where
44:15
are you going to who's trying to solve a
44:16
problem? Where are you going to, you
44:18
know, go after that? Like Nvidia has
44:19
invested a lot of of money and effort in
44:21
into their ecosystem and there's a
44:23
whole, you know, thing around that. Um,
44:26
but but if you're just being like, "Oh,
44:27
I want to get the latest model running
44:28
on my GPU." Yeah, I think you make a
44:30
great point. there are going to be, you
44:33
know, several million uh or not several
44:35
million, tens of millions of, you know,
44:36
Macs out there that can do it versus,
44:38
okay, I don't know how many people, you
44:41
know, bought the specific, you know,
44:43
exact model, you know, graphics card
44:45
that that runs with this CPU and and how
44:47
can I make sure that this is all going
44:48
to going to work the way that I want it
44:50
to?
44:50
>> Since you work at GitHub, you're
44:51
probably on the front lines of this.
44:53
What's your take on uh people using you
44:55
know AI coding agents like you know
44:57
cloud and codeex overr running a lot of
45:00
these open-source projects with pull
45:04
requests that they've you know that
45:05
they've done just because like the speed
45:06
is so much faster now. No, it's a great
45:09
point and and is a qu is is probably
45:10
it's a thing that that we are very
45:12
focused on trying to find a happy
45:13
balance on and um and this is a thing
45:16
that like we've seen like just with the
45:17
kind of rise of agentic coding the last
45:20
several months um and and the power the
45:21
fact that like the latest models are
45:23
really really good like we've just seen
45:25
um explosive growth of of usage on our
45:27
platform and that's great but that also
45:29
comes with the downsides because what
45:31
this means is that way more people than
45:33
ever can just like you said they can
45:34
turn their their um you know AI agents
45:37
onto an open source a project and say,
45:38
"Hey, I have a great idea for a feature
45:40
and I'm gonna have my bot code it and
45:42
I'm going to submit the pull request."
45:44
And maybe it works, but maybe it
45:46
doesn't. And if you're someone who's
45:48
maintaining, you know, usually unpaid an
45:50
open source project, having to triage
45:52
all those issues that are coming in.
45:54
Sometimes people file false security
45:56
reports, some people file false bug
45:58
reports, things that haven't been
45:59
validated, you know, pull requests for
46:01
people to make changes. this now adds
46:03
human um um you know labor where how how
46:05
do we you know figure out how we can um
46:08
you know look through all this stuff and
46:10
figure out what is real and what's not
46:12
and so finding the right balance of like
46:14
not unilaterally saying we're going to
46:17
cut everything off because this is just
46:19
a way people develop but but I think
46:21
giving um uh maintainers um and and you
46:24
know uh project leaders like more
46:25
control over what types of things they
46:27
will accept and won't like we recently
46:29
introduced the ability to even turn off
46:32
um pull requests at all on a repo if if
46:33
someone wants to and and and that was
46:35
something that you know there was some
46:38
resistance to because you don't want to
46:39
discourage collaboration but at the same
46:41
time my personal philosophy has always
46:43
been if someone chooses to make
46:44
something available but they don't want
46:45
to accept pull requests and then that
46:47
should be their right I might not think
46:49
that's the best way of developing
46:50
software but that should be something
46:52
that they can do but you can also for
46:53
instance turn off um you know pull
46:55
requests from anyone except for like a a
46:57
group of you know contributors like so
46:59
if you've if you've already been allowed
47:00
to contribute to the then you can create
47:02
a pull request but otherwise you can't.
47:04
Um but it's it's an ongoing problem that
47:06
we're we're looking at because again
47:08
just the velocity of of what can happen
47:11
and many times it comes from good
47:12
intentions. Sometimes it is spam
47:13
sometimes it is even malicious and that
47:15
can be scary too but sometimes it does
47:17
come from good intentions whereas like
47:18
oh I I have a great idea for this new
47:20
feature for this you know project. Um I
47:23
I am just going to have Claude do it for
47:25
me and submit the pull request.
47:28
What people don't think about is okay,
47:31
but what impact is this going to have on
47:33
the real human on the other side who now
47:34
has to review this and test this and do
47:36
I have the skill to actually make sure
47:39
that my thing is validated and is the
47:41
direction that I want to go in the
47:43
direction the project wants to go in.
47:44
Right? Because those things aren't
47:45
always aligned. And so,
47:46
>> well, thank god there's still some human
47:48
involved in this whole thing.
47:49
>> Yeah. No, there has to be. There there
47:51
has to be. I mean, and and and
47:52
ironically, one of the ways that you can
47:53
maybe make this better is by using AI to
47:56
help kind of determine what are good and
47:59
what are are bad intents, but but that
48:00
seems like a band-aid. I think some of
48:02
this is also just a cultural thing. I
48:04
mean, I I I always try to say to people,
48:06
you know, like fork fork should not be a
48:08
four-letter word. Meaning a lot of times
48:10
I think people think like the only way I
48:12
can um get something done if I want to
48:14
customize it myself is I I have to have
48:16
this this contribution be accepted by
48:18
the main you know project maintainer
48:20
into the project. But if I just want to
48:22
make a modification just for myself I
48:24
can just have it running like this is
48:25
the beauty of open source. I can just
48:27
have my own variant of it running. I
48:28
don't have to have it in the main
48:30
project and then you know have that
48:31
burden on on the others out there.
48:34
>> Is there anything on GitHub that isn't
48:35
AIdriven or
48:37
>> AI focused? I mean, definitely. I mean,
48:39
you know, but but but AI, I think, is
48:41
just kind of like the a big thing for
48:43
for better or worse. Um, I would say on
48:45
the whole better, but we'll see. Um, of
48:47
where software development is just
48:49
going. So, it's it's a big part of
48:50
things. You know, agents are are a big
48:52
part of things. Code review um uh there
48:55
just AI fits the tentacles into a lot of
48:57
places. But, I mean, but yeah,
48:58
fundamentally at the end of the day, you
48:59
know, the things that you're doing on
49:00
GitHub, where you're storing your code,
49:02
um how you're contributing to projects,
49:04
these are all very human problems,
49:05
right? So like that's kind of the
49:06
interesting you question you posed like
49:08
okay how do we like come up with
49:10
solutions for like the the um amount of
49:16
work and activity that can now happen
49:18
like orders of magnitude that we we
49:20
didn't expect before. How do we how do
49:22
we protect you know human time is
49:24
finite. So how do how do we you know um
49:27
balance that because it's it's a problem
49:29
and it's it's not getting better.
49:31
thinking long term there will always be
49:32
a need for humans to generate you know
49:35
there'll always be some market for human
49:36
generated code because the LLMs need
49:38
something to train on in the first
49:39
place.
49:40
>> Yeah.
49:40
>> And well although
49:41
>> unless you want the to just slowly
49:43
deolve over time.
49:44
>> Well I mean there they they do actually
49:45
kind of generate their own code now
49:47
which is in their own training sets
49:48
which is sort of interesting. Um uh
49:50
which is you know and who knows like how
49:52
good that will be but yeah we need
49:54
things to train on. I would also say
49:55
just having like how a lot of people who
49:58
are very very very good engineers use AI
50:00
for for coding is really as a task
50:02
manager and working with for planning
50:04
asking questions back and forth and then
50:06
being very directional. This is what I
50:08
want created. This is how I want it. Now
50:10
we're going to test this and see if this
50:11
fits my requirements. If you know what
50:13
you're doing, you can then maybe see
50:15
some of the problems and and work back
50:16
and forth and and then if you're, you
50:18
know, uh do a security audit and make
50:19
sure that things are up to snuff. if you
50:21
don't know what you're doing, which is
50:22
the bigger kind of kind of question
50:25
mark, it can be like you don't know how
50:28
um good that that code is going to be.
50:30
But um yeah, which I I guess is a long
50:33
way of saying like if you're a security
50:34
uh uh person uh you probably have some
50:36
job security for now,
50:38
>> right?
50:39
>> More than anybody else,
50:41
>> you know, it seems like that there is
50:42
definitely an ongoing need for human
50:44
oversight. I I I don't see that going
50:46
away. it even if it's perfect, someone's
50:49
got to look over it and make sure that
50:50
it's doing what it's supposed to do and
50:52
not doing what it's not supposed to do.
50:53
And Jess Wang, you know, I get this
50:56
morning, I think, gave an interview
50:57
where he said that any company that is
50:59
firing people because AI can now do, you
51:02
know, these people's jobs. That's a
51:04
failure of imagination because um what
51:07
they should really be doing is
51:08
empowering those workers with AI to make
51:10
them do more. You know, these companies
51:12
could be if they if they retain their
51:14
workforce and train them to use AI,
51:16
these companies could be doing so much
51:17
more with their workforce rather than
51:19
getting rid of people. Getting rid of
51:21
people is like an unimaginative response
51:24
to this wave of automation, which I
51:28
don't know that kind of, you know, that
51:29
made sense to me and and gave me some
51:31
optimism. But of course, you know,
51:33
>> a lot of companies who are laying off
51:34
people would have laid them off anyway
51:35
and they're just using AI as an excuse.
51:38
>> Yeah.
51:38
>> Right.
51:39
>> That's true, too. Yeah, that's true,
51:40
too. You talk about, you know, companies
51:42
laying people off. Um, which is
51:44
obviously happening for for a variety of
51:45
reasons, but it's like part of me is
51:47
like, okay, but how are you going to
51:50
continue to how how will, you know, your
51:52
business be sustained if we have more
51:54
and more uh people who who don't have,
51:57
you know, good paying jobs, right? It's
51:59
like that that becomes kind of a kind of
52:01
a problem, too.
52:02
>> Um, but no.
52:03
>> How do you feel? Are you pess are you
52:05
pessimistic or optimistic?
52:06
>> I think I'm probably somewhere in the
52:08
middle. I mean obviously look I work in
52:09
the field and and I've
52:12
which sometimes I'm like am I am I
52:14
aiding in my own demise or not right
52:15
like I have like that question a lot
52:17
right but but but I but I think it's
52:20
tremendously exciting and I think that
52:21
there's tremendous potential I think
52:23
that it is probably I think the people
52:26
who are like this is gamechanging
52:28
nothing will ever be the same again and
52:30
and and this is the the greatest thing
52:31
that's ever happened I think are
52:32
probably wrong. I also think that the
52:34
doomsday prophecers who are like a this
52:36
does nothing. This is just a waste of
52:37
energy and and we never needed any of
52:39
this. I think they're wrong too. I think
52:40
that there's there's some ground in the
52:42
middle which is to say this is truly
52:43
transformative and disruptive. Um that
52:46
has good and bad effects just as any
52:48
sort of you know transformation and
52:49
disruption does. Um but but as a you
52:52
know nerd like this is exciting. um even
52:56
when it's scary, right? Just the fact
52:57
that, you know, you can take a project
53:00
that you know was was created 3 months
53:02
ago and run it on a Mac Mini and
53:04
essentially kind of have it kind of do a
53:06
lot of tasks for you. Um and not even
53:08
just you know computer tasks, just even
53:09
things like just like organizing your
53:11
life like that this is now an accessible
53:12
thing is remarkable, right? that that
53:15
that people are are building um
53:17
solutions for themselves all the time
53:19
which are really great. Even when we
53:22
look at uh things that are not even the
53:23
in the programming realm like just even
53:25
if you think about things like um you
53:26
know the way that voice to um to text um
53:30
>> has improved so much and and and um uh
53:33
uh like the transcription sort of stuff
53:35
whatnot like it opens up a lot of things
53:37
for accessibility and translations that
53:39
we just didn't have before which is
53:41
really really exciting. I mean this is
53:43
going to make it much much easier for
53:44
people. It already is making it easier
53:46
for people for people to communicate but
53:47
when you have talk about like real-time
53:49
translation stuff which you know is
53:51
something that Apple you know was kind
53:52
of teased in the past. feel like we're
53:54
not at that point where the local models
53:57
are good enough for that to really work,
53:59
but we're not far off from it. Where,
54:01
you know, that idea of like the the, you
54:02
know, the Star Trek kind of, you know,
54:04
communicator where, you know, we could
54:05
all kind of understand one another, um,
54:08
you know, regardless of what language
54:10
we're speaking, like that's exciting.
54:12
Like there there are all these really
54:14
interesting potential things. is I just
54:16
hope that people don't get so um
54:18
shortsighted by the the potentials for
54:21
harm or the the potentials for what they
54:23
what they think is going to be, you
54:25
know, something that that might not be
54:26
possible.
54:27
>> So, you know, Griffin, we did a hands-on
54:28
last week uh a kind of first look at the
54:30
the new MacBook Neo and the iPhone 17e
54:33
and the um XDR display, the studio
54:35
display XDR,
54:37
spent uh a week with them and did full
54:39
reviews, you know. So, Griffin, what do
54:40
you Let's Let's hear what you think
54:42
about them, you know, finally after
54:43
spending a bit bit of quality time with
54:45
these machines.
54:46
>> So, the one thing I I definitely wanted
54:48
to do on the Neo that I didn't have time
54:49
to for the show last week after only a
54:51
day is all of the speed tests that I do.
54:53
And, you know, other people just run,
54:55
you know, Geekbench benchmarks, which
54:57
have their place, but I want to I want
54:59
to be more hands-on with it. And so, I I
55:01
threw at it all of like the regular
55:02
tasks that I do all the time, like I
55:05
edited last week's show entirely on the
55:07
MacBook Neo. I, you know, did audio
55:09
processing, audio transcriptions, built
55:12
it a, you know, Xcode projects and, uh,
55:15
ran some data binding.
55:16
>> Some real world stuff. Yeah.
55:17
>> Yeah. Real real world stuff that I that
55:18
I do using, you know, actual actual
55:20
work. And so, in the in those five tests
55:23
that I do, about three out of the five,
55:26
comparing the MacBook Neo to my M2 Pro
55:29
Mac Mini, performance was just about
55:32
identical. Um, for the bigger tasks like
55:35
video rendering and building an app in
55:37
Xcode, things that really benefit from
55:39
having like just the greater number of
55:41
CPU and GPU cores, the Mac Mini came out
55:43
a little bit ahead. But again, for a lot
55:46
of these tests, like anybody who's
55:47
concerned about the MacBook Neo not
55:50
being fast enough just because it's an
55:51
iPhone chip, they forget just how
55:53
powerful iPhone chips are. Like this is
55:55
this is a $500 computer
55:57
>> that outperforms the $2,300 Mac Mini
55:59
that I bought. So, excellent excellent
56:01
performance out of these things. The 20
56:03
watt charger that comes with it is a
56:05
little small. I I I drained one of them
56:09
down to like 3%. It took a lot of time
56:11
because it has a long battery life. And
56:13
then I then when I plugged it in, it
56:14
said that it estimated like four and a
56:16
half hours to get back to 100% which
56:18
isn't terribly fast. But again, this
56:20
thing has such long battery life that
56:22
it's only really a question of like, you
56:24
know, where do I plug it in overnight
56:26
and how long is that going to take?
56:27
Yeah, it'll be fine.
56:29
>> Not not that big of an issue. Um,
56:31
>> or you could get a battery. You could
56:33
use a battery with it. Like you an
56:34
external battery.
56:35
>> Oh, yeah. You probably could. Yeah,
56:37
>> that's a good point. Yeah. But you said
56:39
it works. It It does charge faster with
56:40
a 30 watt. Um,
56:42
>> it Yeah. So, if I was mostly using it on
56:44
my sofa, which is the same in the same
56:46
corner that I use the Vision Pro. Uh,
56:49
and that has a uh that came with a 30
56:52
watt charger. And, you know, so it'll
56:54
charge a little faster with that. Um,
56:56
but 30 watts is as big as it goes. if
56:59
you can you can plug it in with like a
57:00
you know 45 or 60 watt charger but it'll
57:02
only charge at a rate of 30 watts. Um
57:05
>> okay
57:05
>> it doesn't have it doesn't have MagSafe
57:07
which is kind of sad because again this
57:08
this is going to be a computer that I
57:10
imagine a lot of people are going to buy
57:11
for their kids and kids are clumsy and
57:14
they have no respect for technology and
57:15
I imagine a lot of MacBook Neos are
57:17
going to you know tumble off of tables
57:18
and get stepped on and stuff but at
57:21
least it's very repairable so they they
57:24
looking for
57:25
>> Exactly. Exactly.
57:27
>> Good point. I wonder if people will
57:29
bring back the the magnetic uh like kind
57:31
of like USBC charger things that we see
57:33
that would like kind of break off. Like
57:34
I wonder if we'll see a resurgence of
57:35
those again with with the Neo. That will
57:38
be a thing people buy
57:39
>> because for a while that wasn't really a
57:41
thing because we had Mag Safe again, but
57:42
I wonder if we'll see more accessory
57:44
makers making that.
57:45
>> Yeah, maybe maybe they'll even be color
57:46
matched to the
57:49
um it doesn't have an ambient light
57:50
sensor, but Apple does claim that it
57:52
matches the brightness of a room. I
57:54
don't know exactly how. My my guess
57:56
based on observing it is that I think
57:57
anytime you open the lid, it might like
58:00
activate the camera for like a split
58:01
second and try and determine it because
58:02
I notice when you do that, like the
58:04
first time that you open it, it makes
58:06
like a little adjustment, but then it it
58:08
doesn't monitor it continuously. So, you
58:11
know, if you carry it from a bright room
58:12
to a dark room, like you're going to
58:13
have to be clicking the the keyboard
58:15
brightness button on it.
58:17
>> Overall, how do you feel about it
58:18
overall?
58:19
>> Overall, it's it's an excellent
58:20
computer. I like I know this is going to
58:22
be a lot of people's first computer and
58:25
um I I think the interesting way that
58:27
Apple has positioned it is that if you
58:30
have the money to buy a MacBook Air,
58:32
there is literally no reason for you to
58:35
buy a Neo. It is that is very clearly
58:37
not better than the MacBook Air in any
58:39
respect. It's not smaller. It's not
58:40
lighter. So, if you already have a
58:42
MacBook Air, I know there are a lot of
58:44
people who are like, I kind of want it
58:46
because it's cool, but it's like an
58:47
objectively worse computer than the
58:48
MacBook Air in every way. there's no
58:50
reason it's better other than the
58:52
colors. Uh so like I'm sure this came up
58:54
in how when they were designing the
58:56
product like they very intentionally
58:58
didn't make it any thinner or any
58:59
lighter and so they're losing like a
59:01
little bit of sales there for people who
59:02
already own a MacBook Air. They have no
59:04
reason to buy this instead. But I think
59:06
the trade-offs are that, you know, it's
59:08
built with more screws inside. It has
59:11
like a smaller battery, so it's
59:13
incredibly repairable. It's really, you
59:16
know, it's it's an excellent computer.
59:18
You know, nobody was nobody was
59:19
expecting it to be only $600. So,
59:22
>> right. Well, your friend Sam Henry Gold
59:24
wrote a very interesting perspective on
59:25
this, didn't he? He's saying that that
59:26
the limitations are kind of the point.
59:28
>> Yeah.
59:29
>> Because it forces the user to overcome
59:33
them
59:33
>> and
59:34
>> and it and it doesn't matter if it has
59:35
limitations because you know what? It's
59:36
it's so much people are going to get
59:38
this and they are going to push it to
59:39
its limits no matter what. I mean, I was
59:41
doing video editing on an Intel Celeron
59:43
processor like
59:45
>> Well, I read a really interesting story.
59:46
you know, Rockstar Games, a GTA
59:49
franchise. Um, they started in
59:51
Edinburgh, I think, and and Edinburgh
59:52
was a a center for game excellence. You
59:55
like a lot of game companies came out of
59:57
Edinburgh in this is years ago, 20 years
59:59
ago or more. Um, and they were excellent
1:00:02
at optimizing for uh, you know, cutting
1:00:06
edge hardware, but they learned it on
1:00:07
limited hardware.
1:00:08
>> Mhm.
1:00:09
>> And it was because it was a they were
1:00:10
poor. They had crappy machines. They had
1:00:13
to program for these very limited uh
1:00:16
devices, but it turned out they learned
1:00:18
all these amazing techniques that, you
1:00:19
know, came in super handy when they when
1:00:22
they were, you know, when it when it
1:00:23
came time to make games for PlayStations
1:00:24
and Xboxes.
1:00:25
>> Christine, did you get a Neo as well?
1:00:27
>> No. No. I mean, I'm considering it. The
1:00:29
The problem is kind of to your point, I
1:00:31
don't need it. So, if I bought it, it
1:00:33
would strictly just be it's because I
1:00:34
want a pink laptop. And um and and
1:00:37
that's definitely I'm not going to lie,
1:00:38
like there is a lot of appeal to that.
1:00:40
Um, I think that for but for for me
1:00:43
personally the challenge is it's like
1:00:45
okay if you can get it with education
1:00:47
discount it's the same price as a Mac
1:00:49
Mini and I feel like for my use cases a
1:00:51
Mac Mini is a lot more applicable but um
1:00:54
I will say so um for Christmas of 2025
1:00:59
or 2024 I got my mom um an M3 MacBook
1:01:03
Air to replace her 2018 MacBook Air that
1:01:06
I got her in 2018 like the the Retina
1:01:08
MacBook Air which was a terrible
1:01:10
laptop. Um, and uh, you know, they
1:01:13
brought them the Air back and it had a
1:01:14
terrible chip in it and and that was a
1:01:15
disappointing computer. And I got a
1:01:17
really good deal on on that machine for
1:01:19
her from Best Buy. I think it was $800
1:01:21
or something. Um, and then for Christmas
1:01:24
this year, I got my sister who had a
1:01:25
10-year-old MacBook um, Air. I got her
1:01:28
an M2 Air that I think was $600, the 16
1:01:32
gig version on like a a Cyber Monday
1:01:34
Best Buy sale. So, great great machine.
1:01:36
Has Maxafe has all those things. I will
1:01:38
say like I if I were to do this again, I
1:01:41
would still get my sister the the M2 Air
1:01:43
that she got even though like the
1:01:45
performance would be close because I
1:01:46
just feel like it's you know to the
1:01:47
points that you kind of you know um uh
1:01:50
just just said like there are better
1:01:52
things the build quality and like the
1:01:54
the screen and the Mag Safe and the the
1:01:55
port selection um that I think that her
1:01:58
M2 Airspec the fact that it's the same
1:02:00
price was a better deal. But if I were
1:02:02
to buy my mom a computer tomorrow, like
1:02:04
this is like the perfect type of
1:02:07
computer in in in that range. Um I and
1:02:10
and if my dad needed a computer
1:02:12
tomorrow, even though he does not like
1:02:13
Mac OS, this is the sort of thing that I
1:02:15
would buy for him. And so I I feel like
1:02:18
for all the the niggles that I might
1:02:20
personally have about some of the
1:02:21
decisions they made about, you know,
1:02:23
where to cut corners and whatnot, um I
1:02:25
feel like I've kind of come around where
1:02:27
I'm like the price is so good that it
1:02:30
doesn't matter. Especially looking at
1:02:32
your benchmarks and your review and
1:02:33
whatnot, right? Like I do worry 3 years
1:02:36
from now, what is it going to actually
1:02:39
be like to have 8 gigabytes of RAM? I do
1:02:41
actually really worry for that to be
1:02:42
honest with you. And so I I I my my my
1:02:46
slight concern for for some folks is
1:02:47
just be like don't expect like because
1:02:49
people have kind of the general
1:02:50
knowledge oh max lasts forever and at
1:02:52
least with this model I'm kind of like
1:02:54
okay we are literally right I think at
1:02:56
that edge just at least in terms of the
1:02:57
the RAM amounts where we're it's it's
1:03:01
it's certainly usable but I don't know
1:03:03
like several years from now if this is
1:03:05
going to be a machine that you you're
1:03:06
not even going to know why it's giving
1:03:08
you issues but it but it's might might
1:03:10
be pushing those boundaries. Um, but but
1:03:12
I feel like just the price is just so
1:03:14
good and and I think, you know, uh,
1:03:16
>> sounds like everyone's going to be
1:03:17
getting a MacBook Neo for Christmas in
1:03:19
your house.
1:03:20
>> Well, maybe. I mean, look, they all have
1:03:21
enough laptops. So, they're they're
1:03:22
good, but if I were trying to buy one
1:03:23
for like a like my my, you know, my
1:03:25
mother-in-law or somebody else who
1:03:26
needed one, like, it's just I think that
1:03:27
this is a really good go-to if you have
1:03:31
certain needs, if you have the budget,
1:03:32
get the Air and and get a refurb if you
1:03:34
can from Apple because you can get
1:03:36
really good deals on those. But
1:03:38
otherwise the the Neo even though like
1:03:41
like I said I have some concerns about
1:03:42
like okay will this actually be a viable
1:03:44
computer in 5 years there's also a
1:03:46
question to be made maybe we don't need
1:03:48
it.
1:03:48
>> We published an opinion please this week
1:03:49
arguing that um it's do uh the 8
1:03:52
gigabytes of RAM is doing everybody a
1:03:53
favor because Apple will not we'll try
1:03:57
to optimize Mac OS so that it'll run
1:04:00
yeah
1:04:01
>> well on this machine and and so that you
1:04:03
know uh that limitation is going to do
1:04:06
everyone a favor. I'm using an old M1 uh
1:04:08
iMac right now which I bought back in I
1:04:11
don't I don't know when is it 2001 maybe
1:04:14
uh 21 sorry yeah 2021 and it's perfectly
1:04:17
fine. I mean it chokes you know if I
1:04:19
have 300 Chrome tabs open which I you
1:04:21
know tend to do. Um but otherwise it's
1:04:23
it's it's huming along great you know it
1:04:26
no it it you know quite happily handles
1:04:29
Slack and Chrome and Riverside and all
1:04:32
these other applications that I've got
1:04:34
open. No problem at all really. And it's
1:04:36
only got 8 gigs of RAM. Sorry. That's an
1:04:38
important point.
1:04:39
>> Yeah. No totally totally. I mean, yeah.
1:04:41
I I think that that um I did want to
1:04:42
know, Griffin, like from your
1:04:43
perspective, like when you're testing
1:04:44
things, the the storage is a lot slower
1:04:48
than what we saw what we've seen on
1:04:49
other Macs. Um did did that impact
1:04:52
anything like in your usability or
1:04:53
anything? I mean, again, I don't think
1:04:54
for the target audience of this laptop,
1:04:56
it matters, but just curious what your
1:04:57
perspective was.
1:04:58
>> The storage speed I didn't really notice
1:05:00
as much, but I did notice that I mean, I
1:05:01
had two of them. I have the two because
1:05:03
I I simply cannot be stopped. I I bought
1:05:06
both the 256 and the 512 one
1:05:08
>> and the 512. Yeah. um the 256 like if
1:05:11
you're doing uh I mean the the two worst
1:05:14
apps for storage space are Final Cut Pro
1:05:16
and Xcode. Both of those will just eat
1:05:18
up no matter how much storage you have.
1:05:20
You can get like the 4 TB or 8 TB ones
1:05:23
and it'll because you know then Xcode
1:05:24
will just like keep downloading uh new
1:05:27
SDKs for each new point release of iOS
1:05:29
and each of those takes up 10 gigabytes
1:05:31
and if you never cleaned it out then
1:05:32
that'll just like eat up over an an
1:05:34
entire terabyte of your storage. or
1:05:36
Final Cut Pro. It's really bad in that
1:05:38
in its default settings, anytime you
1:05:41
import media into a library in Final Cut
1:05:43
Pro, it makes a copy of it and that'll
1:05:46
just eat up hundreds and hundreds and
1:05:47
hundreds of gigabytes of space. So,
1:05:49
import like importing one episode of the
1:05:52
Cultcast into Final Cut Pro while having
1:05:54
like a bunch of other like pro apps just
1:05:56
downloaded on them ate up the entire 256
1:05:59
GB of storage pretty quick.
1:06:01
>> So, I
1:06:03
>> one day, huh?
1:06:04
>> Yeah. I I recognize like the 256 is like
1:06:07
the the the storage capacity is much
1:06:09
bigger limitation than the storage
1:06:10
speed. I mean it's an SSD. It's still
1:06:12
incredibly fast. Apple has high
1:06:13
standards for storage.
1:06:14
>> That's good to know.
1:06:15
>> A super cheap like SSD off of like
1:06:19
Amazon. Like some of them are really
1:06:21
sketchy. Like they they promise like a
1:06:22
terabyte of SSD storage for only like 50
1:06:24
bucks. And it's th those are only going
1:06:27
to be barely faster than a spinning hard
1:06:28
drive.
1:06:30
But those little Kingston ones now, the
1:06:31
little tiny ones you can plug into the
1:06:33
into the side, you know, they're they're
1:06:35
one terabyte, I think. Um they they're
1:06:37
plenty fast enough for for um for a lot
1:06:40
of most tasks, I believe.
1:06:41
>> Yeah. Um
1:06:42
>> and it's pretty inexpensive, too. I
1:06:43
think it's 60 bucks.
1:06:45
>> Mhm. I mean, just make sure that if you
1:06:46
get one of those on a MacBook Neo, plug
1:06:48
it into the port in the back so you get
1:06:49
the full
1:06:51
I was going to say, yeah, plug it into
1:06:53
the right port. Um did you use it with
1:06:55
external monitors? I know that it can
1:06:56
only outplay to to 4K and I know that we
1:06:58
we're about to talk about your u monitor
1:07:00
review, but um just curious like what
1:07:02
the performance was like um if you were
1:07:04
using it with a with this um an external
1:07:07
display.
1:07:08
>> So, this is kind of interesting. I
1:07:09
plugged it into an external 4K monitor
1:07:12
that I like a basic one from LG and
1:07:15
>> Mhm.
1:07:15
>> it because it's a limitation of the
1:07:18
bandwidth and not actually of like the
1:07:19
compute power. Even though it's a 4K
1:07:21
monitor, I could get more than 4K
1:07:23
resolution by using it in a scaling
1:07:25
mode. So, I scaled it up to like, you
1:07:27
know, the equivalent of like a, you
1:07:29
know, a 27 inch 5K monitor, but on a 4K
1:07:32
space and it could it could still do
1:07:34
that. And that's what I did when I used
1:07:35
that 4K monitor as my regular, you know,
1:07:37
one and only screen before I got a 5K
1:07:40
screen.
1:07:40
>> That's not as big of a deal. Like, you
1:07:42
can plug it into a a studio display and
1:07:44
it'll be, you know, 1K fuzzier. So it'll
1:07:48
be 20% like a little blurriier but
1:07:51
>> yeah I
1:07:52
>> yeah no I guess my my question was more
1:07:53
about the and about the scaling thing
1:07:55
was because I know that on on some
1:07:56
machines like there have been like uh
1:07:58
you know scaling like because it's it's
1:08:00
doing that stuff like there might be
1:08:02
performance hindrances and I did did you
1:08:03
notice anything with that if you were
1:08:05
when you were kind of doing it in that
1:08:06
scaled motion was was everything still
1:08:08
fluid and and working well for you?
1:08:10
>> Yeah. Yeah, it was still fluid. It
1:08:11
actually it took like you know two or
1:08:13
three tries plugging it in but it might
1:08:15
just be a dodgy cable that I used
1:08:16
because it was like an HDMI to USBC
1:08:18
cable that might have been cheap.
1:08:19
>> Got it.
1:08:20
>> But speaking of the display, we want to
1:08:22
talk about that a little bit.
1:08:24
>> Yeah. Yeah. Let's talk about the XDR
1:08:25
display. I think a lot of people are
1:08:26
interested in this because this is, you
1:08:27
know, this is a a great display, isn't
1:08:29
it? You know, fantastic uh performance
1:08:31
for half the price of what it used to.
1:08:32
>> And I was really surprised that I don't
1:08:34
think Apple seeded any review units of
1:08:36
the XDR to anybody. I haven't seen any
1:08:38
reviews of the XDR from, you know, any
1:08:41
of any of our peers, but um yeah, I I
1:08:44
bought this and, you know, I I do plan
1:08:47
on returning it. I can't just buy a
1:08:48
$3,300 monitor, but you know, I I'm
1:08:51
really enjoying my time with it. Um so,
1:08:54
four years ago, everyone celebrated the
1:08:56
launch of the Studio Display and laughed
1:08:59
at the then overpriced and four-year-old
1:09:02
Pro Display XDR. But now, the tables
1:09:05
have turned. It's the studio display
1:09:06
that's overpriced and underfeatured in
1:09:08
its class because you can get a 5K 27
1:09:11
inch 60 Hz display from ASUS, BenQ or
1:09:14
VSic for like in some cases literally
1:09:16
half the price. Apple reinvented its
1:09:18
high-end display as a product that makes
1:09:20
much smarter trade-offs and it's
1:09:22
somewhere between 55 to 66% cheaper. And
1:09:26
$ 32.99 is still a lot of money, but at
1:09:28
least now it's a price that makes some
1:09:30
sort of sense because, you know, show me
1:09:32
the other 5K Retina 120 Hz display with
1:09:37
HDR. There isn't one. It's the only one.
1:09:39
So, Apple has now decided like this is
1:09:41
how much it costs and, you know, they
1:09:43
own that market now. I think the the XDR
1:09:46
makes much more sense than the regular
1:09:48
studio display.
1:09:50
>> Still a lot of money, though. This is,
1:09:52
you know, isn't they going to sell a lot
1:09:53
of these or uh because monitors used to
1:09:56
be so people used to obsess over
1:09:58
monitors? I remember working at Macworld
1:09:59
and they'd have special editions where
1:10:01
it was devoted to monitors, monitor,
1:10:02
monitors and then everyone who was
1:10:04
working in print, film, anything visual,
1:10:07
you know, had to have the highest end
1:10:09
monitor. That must still be true
1:10:10
obviously, right? And you know, people
1:10:13
can be I've seen those editing bays in
1:10:15
Hollywood where they have 12 monitors,
1:10:18
>> all of them Pro Display XDRs. I do
1:10:20
wonder if the fact that when you speak
1:10:22
of Hollywood, I do wonder if like just
1:10:24
the fact that this only works with Macs
1:10:27
and a Macs of a certain vintage if that
1:10:29
might actually inhibit some of the
1:10:31
adoption because so many of those render
1:10:33
farms and things are are not using Mac
1:10:35
OS. I mean, some of them are, but some
1:10:37
of them are not.
1:10:38
>> Well, you can still use it with with a
1:10:40
PC. Like uh it actually even though
1:10:42
Apple doesn't say it, it it does work
1:10:44
with an Intel Mac if it has Thunderbolt
1:10:46
3, but you don't get all of the nice
1:10:48
features. So, you probably don't get
1:10:49
like the single specific reference
1:10:51
modes, but you can still plug it into a
1:10:52
PC.
1:10:53
>> Oh, okay. Well, okay. I didn't know
1:10:54
that. That's great to know. Okay. So,
1:10:55
that that actually makes a lot better
1:10:56
because I was under the impression that
1:10:58
that it wouldn't work with with Intel
1:11:00
stuff at all. And and that's kind of a a
1:11:03
deal breaker in my opinion if you're
1:11:04
spending $3,300 on a monitor that you
1:11:07
can only use with certain types of
1:11:09
hardware. like it's bad enough, you
1:11:11
know, on your um studio regular, you
1:11:13
know, $1,600 studio display that you
1:11:15
have to go through many many generations
1:11:16
to get it to work with, you know, a game
1:11:19
console for instance. Um although a
1:11:21
Steam Deck works pretty well with it,
1:11:22
but um uh Okay, that's that's good to
1:11:25
know. So, it won't have all the maybe
1:11:26
the features, but it will still work
1:11:28
with with a with a um Intel machine as
1:11:30
long as it has a Thunderbolt 3. I I was
1:11:32
curious about this because they they on
1:11:35
all of their other products that run 120
1:11:36
Hz, the iPhones, the iPads, the MacBook
1:11:38
Pros, they all call that prootion, but
1:11:41
they specifically didn't say the XDR has
1:11:43
prootion. They said it has 120 Hz with
1:11:46
adaptive sync. And I was curious what
1:11:48
the difference was. And uh you know,
1:11:50
their their technical white paper does
1:11:52
say that um adaptive sync is like the
1:11:54
more industry-wide standard version of
1:11:57
it. And it even enables like a special
1:11:59
mode where you can specify in the Studio
1:12:02
Display XDR, you know, based on your
1:12:04
workflow whether you want it to optimize
1:12:06
for the lowest latency possible or the
1:12:09
highest image quality possible. Like,
1:12:12
and it it's it makes a difference down
1:12:13
to like the milliseconds over like when
1:12:15
the display changes the image or the
1:12:17
refresh rates and stuff like that. It it
1:12:20
it's like really technical stuff, but
1:12:22
like you know this is a truly
1:12:24
professional display that they're that
1:12:27
they're targeting to these markets like
1:12:28
you know by discontinuing the Pro
1:12:30
Display XDR which I know had like at
1:12:32
least a little bit of you know
1:12:34
penetration into the uh creative
1:12:36
industries like you know
1:12:38
>> you know in movie studios and things
1:12:40
like I was worried whether are they
1:12:41
going to lose that market. It doesn't
1:12:42
seem like they will um because you know
1:12:45
this is a superior display. It just
1:12:46
happens to be a little bit smaller, but
1:12:48
it's half the price. So, I I think that
1:12:51
>> it was half the price. You can get two
1:12:52
of them. Yeah. And and and um Yeah. And
1:12:54
yeah, I'm looking at the white paper
1:12:56
now. even saying like they can basically
1:12:58
have it set by default to be either the
1:13:00
North American broadcast, you know, kind
1:13:02
of standard or or European standard that
1:13:04
you can set, which um so I I I think I
1:13:07
actually do think now um now that I know
1:13:09
that this can be used with non-Mac hosts
1:13:12
that yeah, I would expect especially at
1:13:14
this price probably to see this in a lot
1:13:15
of a lot of movie studios. um even if
1:13:18
just because it's difficult. I I don't
1:13:21
know if they would necessarily need the
1:13:22
the high refresh rate to be honest with
1:13:24
you. But um you're certainly not going
1:13:26
to find anything of this resolution or
1:13:27
this quality for this price because the
1:13:29
panel just doesn't exist. We're finally
1:13:31
seeing the the 5K, you know, 60 Hz
1:13:33
panels in a lot of displays that are
1:13:35
well below, like you said, the the
1:13:37
current studio display pricing. Um, and
1:13:38
we're finally also starting to see a lot
1:13:40
of the 6K 32 inch panels. I think there
1:13:42
are two um very uh um two different
1:13:44
panels that are available from a a
1:13:46
variety of manufacturers. And that's
1:13:48
great, but we haven't seen anybody do
1:13:51
120. All we've seen I think it's the
1:13:54
high end would be in 4K mode it could
1:13:56
run at like 120 Hz, but at 5K or higher
1:13:59
resolution it would still be, you know,
1:14:01
60 Hz. So, the fact that this is
1:14:04
>> and that's before you even get to HDR,
1:14:05
right? This display is gorgeous. Like
1:14:08
even
1:14:09
>> like I knew we would do HDR videos and
1:14:10
shows and all that, but like the number
1:14:12
of just regular pictures in my photo
1:14:14
library, those are all HDR. Pictures
1:14:15
people send me like they all look
1:14:17
absolutely incredible. Um I was also
1:14:20
amazed.
1:14:20
>> Can you say a big notable difference to
1:14:22
like, you know, a cheap monitor?
1:14:23
>> Oh yeah, absolutely. I mean, just
1:14:25
because it's it's glossy as well, so it
1:14:27
just has like that much more contrast
1:14:28
built into it. Like it this thing is
1:14:30
absolutely gorgeous. I I normally have
1:14:32
my display pointing away from my office
1:14:34
window so I don't have to normally worry
1:14:36
about reflections, but to test this I
1:14:38
set it up on the opposite side of the
1:14:40
room and I like directly compared like
1:14:42
this with the uh V sonic display which
1:14:45
is just regular 5K 60 Hz that I used
1:14:47
most of the time. And yeah, there were
1:14:50
like a few reflections in it much less
1:14:53
dramatic than I was expecting. Like I
1:14:55
could still see the display perfectly
1:14:57
clearly, just like a little hint of a
1:14:58
reflection. But then I looked over at
1:15:00
the Vison Sonic monitor and like those
1:15:02
blacks were much more washed out. Like
1:15:04
the thing about a matte display is that
1:15:06
yeah, you don't see a clear reflection,
1:15:08
but it it still has that light falling
1:15:11
on it. It just distributes it more and
1:15:12
that washes out all the blacks and the
1:15:14
colors. Um, so this thing absolutely
1:15:17
gorgeous. I was also amazed by the
1:15:19
speakers. Uh, like you know, I queued up
1:15:21
an episode of Doctor Who to play and
1:15:23
like just the theme song like the rich
1:15:25
bass in it. It was astonishing. Like it
1:15:28
honestly the the speakers sound maybe
1:15:31
even better than like the fulls size,
1:15:33
you know, speaker that I use on my
1:15:35
computer the rest of the time. Like if I
1:15:37
got those display, I probably wouldn't
1:15:38
need it anymore. I don't just use the
1:15:39
built-in speakers when I'm not
1:15:40
>> Did you watch those two Lost episodes
1:15:42
they just found?
1:15:44
>> They haven't been released yet. They're
1:15:45
they're debuting, I think, on uh Easter
1:15:48
weekend or like the first a weekend of
1:15:50
April around April 4th, I think. But I'm
1:15:52
I'm looking forward to those uh as as a
1:15:55
side note. Um
1:15:58
>> the iPhone 17. Why don't you just run
1:16:00
real quick, you know, just speedrun this
1:16:03
one?
1:16:04
>> Yeah. Um I spent more time with the
1:16:05
camera on it. I was taking a lot of more
1:16:07
pictures with it. Um they look super
1:16:09
sharp. You know, I I kind of have like
1:16:11
an existential breakdown in the middle
1:16:13
of our review. Like I'm I'm supposed to
1:16:15
take a bunch of pictures and like come
1:16:16
up with some sort of feelings on it, but
1:16:19
we all know what iPhone pictures look
1:16:20
like. And this takes iPhone pictures.
1:16:22
They look great. It's simultaneously
1:16:25
amazing that we all just carry this
1:16:27
photographic power with us in our
1:16:28
pockets wherever we go. But then, you
1:16:31
know, but it's it's the 17th version of
1:16:34
that. And, you know, they're just not
1:16:36
that special. Like, if if you have an
1:16:39
iPhone, you know, 17 Pro, you've already
1:16:40
seen the pictures that it takes, which
1:16:42
are which are very very good, but uh you
1:16:45
know, it's only that one lens. So, I I
1:16:47
took all of the pictures from my other
1:16:49
reviews on the 17E. They look great.
1:16:52
that they look really beautiful. They
1:16:54
were really really nice pictures.
1:16:55
>> Um until you zoom into 2x like yeah when
1:16:58
you take a 2x it's they say it's like a
1:17:00
full 12 megapixel resolution image but
1:17:02
they get really smudgy at that size.
1:17:04
It's just the limit of physics. Um I do
1:17:06
kind of miss the physical 2x zoom lens
1:17:09
for that.
1:17:09
>> Mhm.
1:17:10
>> It doesn't have macro mode. Uh I'm very
1:17:12
surprised because this camera system has
1:17:14
been designed not to automatically
1:17:16
switch to like a.5x lens when you get
1:17:19
super close to your subject. It lets you
1:17:21
get a lot closer just with the regular
1:17:22
1x lens. Like noticeable difference.
1:17:25
Even though like a lot of the young
1:17:26
people like the 0.5x lens, I'm not a big
1:17:28
fan of it. I was thinking about this in
1:17:31
terms of um hardware features. The
1:17:33
iPhone Air, which has a very similar
1:17:35
camera setup, uh only the one rear lens,
1:17:38
basically the same processor. All of the
1:17:40
other iPhone 17 generation models
1:17:43
support dual capture where you can
1:17:45
simultaneously record on the front and
1:17:46
back camera. The 17E doesn't. That's a
1:17:50
software feature, not a hardware
1:17:51
feature, right? Like why it has the same
1:17:53
processor in it. I guess that's
1:17:55
>> Well, it's also a different frontfacing
1:17:57
camera.
1:17:57
>> That's true.
1:17:58
>> The frontacing camera is So, so I wonder
1:18:00
I mean it I'm sure it's a
1:18:02
differentiator. Um but like that that's
1:18:04
been my thought is that the big the
1:18:06
whole reason they're going to try to get
1:18:07
people to spend the money on the actual
1:18:09
17 versus the 17e is the selfie cam.
1:18:12
>> Yeah, I had thought about that. Maybe
1:18:13
the Yeah, maybe it is like the center
1:18:14
stage camera that's needed for dual
1:18:16
capture compared to my iPhone 16 Pro.
1:18:19
Like I know that of all the changes like
1:18:21
the N1 chip is maybe the least exciting
1:18:22
ever, but it actually makes a pretty big
1:18:24
difference for me because I have in my
1:18:26
house uh a single Wi-Fi router provided
1:18:29
by my internet service provider. Um, I
1:18:31
haven't bought one yet myself, so it's
1:18:34
incredibly cheap. And it's in the and
1:18:35
it's in my office because I want to be
1:18:37
plugged in over Ethernet when I'm at my
1:18:38
computer, but the rest of the house
1:18:40
suffers because in my bedroom, I which
1:18:42
is on the opposite corner. Um, on my 16
1:18:46
Pro, I get pretty spotty connections.
1:18:47
Like it drops out all the time,
1:18:49
especially with like all the doors
1:18:50
closed. Um, but the 17E I think because
1:18:54
it has the N1 chip in it gets noticeably
1:18:56
better performance. It doesn't drop out.
1:18:58
I get a very solid Wi-Fi connection. So,
1:19:02
>> if you don't have a very good internet
1:19:03
setup at home, you just stick with
1:19:05
whatever you know Spectrum or AT&T gives
1:19:08
you. You'll probably notice a difference
1:19:10
in Wi-Fi reliability on the on the N1
1:19:13
chip,
1:19:15
which I I I don't imagine a lot of tech
1:19:17
bloggers talk about because a lot of
1:19:18
other people have uh much nicer Wi-Fi
1:19:20
setups like mesh networking. I don't
1:19:22
have that. I probably should, but
1:19:24
whatever.
1:19:25
>> Yeah. Yeah. this more more Apple silicon
1:19:28
excellence from from Apple, you know,
1:19:30
more chip excellence.
1:19:31
>> Uh inc kind of incredible what they're
1:19:33
doing. This only, of course, this took
1:19:34
him what, five, six years to make.
1:19:36
>> Yeah. Yeah.
1:19:37
>> Um
1:19:37
>> honestly, like the biggest thing that I
1:19:39
noticed having the the 17e is just the
1:19:41
weight. I I very quickly got used to the
1:19:44
iPhone Air's size and weight uh such
1:19:47
that like, you know, after two days with
1:19:49
it, I picked up my 16 Pro and I was
1:19:50
like, "Oh, what is this thing? It's like
1:19:52
I'm holding a brick in my hands." It's
1:19:54
taken me a bit longer with the 17e to
1:19:56
get to that feeling, but you know, after
1:19:58
I spent a week with it, it like I really
1:20:00
noticed the difference in weight. It
1:20:01
doesn't have that dramatic difference in
1:20:03
thickness, but the weight like I pick up
1:20:05
my 16 Pro and it's like, "Wow, I'm going
1:20:07
have to go back to this."
1:20:09
>> Yeah.
1:20:10
>> Feel very, very sorry for you.
1:20:13
>> I think that's a wrap. So, please give
1:20:15
us a fivestar rating uh or a review on
1:20:17
Apple Podcast or share it with someone
1:20:19
who you think uh would like to listen.
1:20:21
Text us uh on iMessage at cult
1:20:23
1:20:25
That's cult [email protected].
1:20:28
Send in your questions, comments,
1:20:29
feedback for the show. You can also send
1:20:31
an audio message or a short video for us
1:20:33
to play to. Um so you can find Christina
1:20:35
Warren on Macedon Blue Sky and GitHub at
1:20:38
Filmgirl and on Macre Weekly podcast
1:20:41
every Tuesday afternoon. That's an
1:20:42
awesome gig. Yeah, congratulations on
1:20:44
that.
1:20:44
>> Thank you so much. Thank you so much.
1:20:45
Yeah, I was I was really um um honored
1:20:47
that they uh they asked me to join them
1:20:49
and uh and it's it's been a lot of fun.
1:20:50
It's been a couple of weeks. And um
1:20:52
>> how do you how do you find the time?
1:20:53
You've got a full-time job.
1:20:54
>> I do. I do. Um GitHub was was wonderful.
1:20:57
They said, "Okay, as long as you can get
1:20:58
your other duties done, then you can you
1:21:00
can do the the show." Uh and like I
1:21:02
said, I'm on leave right now. So, uh we
1:21:04
we we will see like uh when when I when
1:21:06
I start back at at work, um I just have
1:21:09
a time blocked out of my schedule, but
1:21:11
I'm very lucky that I have an employer
1:21:12
who uh who is willing to basically say,
1:21:14
"Okay, as long as you have your other
1:21:16
tasks taken care of,
1:21:18
>> you can take a break. All right. Well,
1:21:21
thanks everybody for listening. Thanks
1:21:22
for watching. And we'll see you next
1:21:23
time. Have a great weekend, everyone.
1:21:25
>> Goodbye.
1:21:25
>> Bye.
1:21:26
>> See you.