#Apple event
1 messages · Page 3 of 1
The notch is weird but with how they are doing it, I'm ok with it
That market is quite profitable, but competitors might outperform and prepare tactics. Look at the past, bigger behemonths for their time died. Atari? Commodore? They lack a solid ground, Microsoft rescued them once 😄
Yeah same. I'm actually really stoked they are finally allowing the 14 inch and 16 inch to have feature parity
They died because they couldn't keep up. Right now, Apple are the only people pushing consumer level ARM, and are proving it to be successful. AMD and Intel will have to keep up. We already know it's the future with huge players in the server space making transitions to ARM and moving away from A&I to produce their own chips
it was a fucking meme
I want 100x more power and very advanced AI capabilities, neurochips. The rest if just Moore and limited digital technology. PCs are more modular, anyway
bring back Slot 1!
the new macpro could be basically a mac mini will more i/o and stuff to add inside
kinda like the power cube
@trim flare bring back trashcan!
Its wild to think
That they released the mac pro
and now 2 years later here we are where you can buy a fucking laptop with arguably better power lol.
does the m1 max outperform the gpu on the pro?
😦
Who owns ARM? That's the bottleneck. Everything can happen. Licensing an architecture from a third party is risky. Even PowerPC was a joint venture and died, and PowerPC was extremely promising too
it looks like it will, yes
is this a shitpost?
ARM licenses their cores. Apple makes their own cores
Im unsure on the highest versions though
God
But ARM was going to be bought by NVIDIA. Now the Chinese? That's not a solid ground for an architecture maker. I like ARM, despite it's extremely propietary nature
It's always been Chinese anyways, well at least in part split with the English company
But x86 is also only from Intel and AMD and VIA
and look at where that's led us
No. It's history. I want competitive architectures, but world tend to go thought spirals of power monopoly
There's a significant part in China i believe
and what is wrong with what apple is making?
I agree. VIA is practically dead
x86 will just be a small niche in the market
It's okay. But I miss powerful workstations outside x86 world. I would prefer ARM to be an ISO standard fastly developed by a mix of public and private research entities, but well
I'm bored to read that since the 80s. Unfortunately, never dies
It's more true by the day
And I don't like X86! It's just X86 still is competitive outside low power world
fooken 8086
x86 is looking like even the high power segment will be lost here soon
x86 loses a lot to how complicated the ISA is
Nothing is meant to be permanent, it happens. Just like corruption, mafias and corporations🤣
makes for a gigantic inefficient decoder that takes up significant area on the die
x86 is losing ground almost everywhere.
The PC laptop/desktop market is the only market not yet seriously affected.
Even Intel tried to kill X86 more than once! They even made ARM CPUs. Itanium. And those weird non-x86 cores
There's something to be said about "to early for its time"
What a time to be alive
The average consumer is 100% ready for a transition to ARM, considering how everything is a webapp and we have hardware translation layers now. Also things like Intel VBridge, while not finished yet, should provide similar performance to Rosetta
After the stagnation of intel
x86 never reached success outside desktops and laptops. Servers migth be lost, might not. Legacy compatibility is a key. Tons of supercomputers are still X86 too, etc
We got the nice boost from AMD
and Intel's system should also allow for ARM apps to run on x86
And then fuckin Apple is like "hold my beer, kid."
aren't a lot of supercomputers MIPS?
The server market would like to correct that statement
The entire cloud revolution was powered by x86
Wasn't literally everything in computing x86?
With the exception of some rare one off devices?
That's a very optimistic assumption. There's gamers, those that refuse to use X86 consoles. Many average users might need things useless thought webapps. Webapps sucks, but that's another thing
Yeah, I mean x86 consoles aren't exactly going anywhere in the next few years, but that's likely the last hold out until next gen
POWER has a small but significant market in supercomputing, banking, safety critical industries and the likes
I can absolutely promise you that Intel and AMD are looking at Apple and thinking twice right now
but otherwise it was x86 everywhere that wasn't mobile.
I would consider gaming outside of the average consumer. Enthusiast PC gaming is a drop in the total gaming market
It is a drop in bucket, all things considered
Enthusiat PC gaming spends 300X more money than average consumer too 😄
what architecture did Motorola have
That is true, gamers love to consooooooooom
the Motoroloa arch
Is the only name it has
The Motorola 68000 (sixty-eight-thousand; also called m68k, Motorola 68k, sixty-eight-kay) is a 16/32-bit complex instruction set computer (CISC) microprocessor, introduced in 1979 by Motorola Semiconductor Products Sector.
The design implements a 32-bit instruction set, with 32-bit registers and a 16-bit internal data bus. The address bus is 24...
TOP500?
Also, for most those apps that the average person has that wouldn't be available on an ARM pc... most of those things typically HAVE alternatives, that same group of people just doesn't want to search for the alternatives
Most apps that the average persons uses already are available on ARM on Windows
Browsers, office, etc. Already there.
indeed.
Yeah
and with the soon-to-be release Intel VBridge stuff it should not even matter
The issue the games and professional Applications like Adobe, video editors etc
But there's an effect that might spoil non-x86 systems, even more than legacy. Even those those average users might need special software even not available for ARM on Windows. ARM on Windows isn't so much of a success
Well, older ones lol
Then theres also the issue of niche applications
W11 is now native to ARM, full support
it's on the list of officially supported CPUs on W11's main page.
Ok so
Bunch of Qualcomm ARM stuff
Can you put an NVIDIA or RADEON? Where's the games list? Can it run the latest Call of Duty crap?
Can I do a VM of W11 on mac OS?
and run anything on windows that I used to run?
Does windows have that translation layer in place?
I mean... GPUs don't care about what CPU architecture you have. And under x86 emulation yes you can, they released 64 bit emulation modes
It used to be, not sure now
The only testing I've seen of it though was under a VM on an M1 mac, so not the best test target lol
We just don'th ave any fucking arm machines yet lol
6 x86, 2 PPC, 1 ARM and 1 custom Chinese ISA in the top 10 currently.
My biggest hesitation with buying the macbook pro atm is not being able to run windows software I need with in the VM because its on ARM
Emulation of completely different architectures is just a temporary solution until the software is native, not a goal. It happened in the PowerPC to X86 transition. It happened in the 68k to PPC transition
I mean of course.
Once mainstream ARM processors are released outside of Apple world is when we'll start seeing native builds of software
But all the missing pieces of the past are here now
Ironically, supercomputer architectures are more diverse than consumer ones 😛
Does W11 have a translation built in?
Yes
For macOS arm
What?
because they can be custom built for a specific job on an enormous budget
I.e, if I am on Mac OS and I VM W11, can I still run old x86 apps in the W11 VM?
Does that translation layer work a la rosetta?
You can, but that's irresepctive of it being a Mac
K
assuming you can get it running, with some config changes it is possible
because rn Windows/Apple are having a licensing fit about it
Yeah I could care less about that
If it gets my windows account banned, Ill make a new one
lol.
I may sound crazy, but the unstable ownership of ARM and not becoming a powerful ISO-ECMA standard joinly worldwide developed and researched migh e a big issue in the future. Despite it's shortcomings, RISC-V might be quite promising in 10-20 years if it becomes the idealistic True Open CPU Architecture
ARM started as a very humble architecture! And look now 🙂
and like all idealistic True Open projects, no one ever puts serious money into it
ARM has the momentum in spite of its ownership/licensing issues. RISC-V is still stuck in lack of momentum and tooling
and then its just a irrelevant side-note in history
I'm not saying it's possible but nothing is pointing to RISC-V taking over
ARM got designed to save transistors at maximum because the company didn't have the resources of big chip corps. Fairchild traitors etc
Sounds like they are popular.
meh who cares Apple owns their ARM license and can do whatever they want with it
Maybe is RISC-V, maybe it's other. But ARM started as a very humble architecture, slowly entering into the mobile world...
Can't licenses be revoked or be invalid if the company or the new owners want to?
no carl
No, that would be part of the argreement
Its similar to how a renter getting a new landlord works
Uhm
ARM started as a desktop architechure, anyway.
The previous argument is contractually kept between the parties.
Sure they could cut off new versions maybe? but who cares Apple is only Apple and so they can add whatever they need
The new company may try to revoke, but they would be sued in court
Also, ARMs product is licenses. Why on earth would they just start telling their customers "no"?
idk how perpetual ARM's license
and it'd be hilarious to watch
cause money
lol
lol
That is a good question
ARM has a lot of IP and talent
Apple M1 Max
More power for the most demanding workflows
Up to 4x faster graphics performance¹⁰
Over 400GB/s memory bandwidth
Supports four external displays YO WHAT
FOUR EXTERNAL DISPLAYS???
Too bad Nvidia has no talent or ability to make their own cores because of how poor of a company they are
Excuse me, but the situation of ARM architecture owners is quite hairy. ARM british vs ARM China...
https://en.wikipedia.org/wiki/Arm_Ltd.#Changes_of_ownership
Arm Ltd. (stylized as arm) is a British semiconductor and software design company based in Cambridge, England. Its primary business is in the design of ARM processors (CPUs), although it also designs other chips; software development tools under the DS-5, RealView and Keil brands; and systems and platforms, system-on-a-chip (SoC) infrastructure ...
Youre joking if you think the world will allow ARM to be exclusively owned by China lol
What are the most demanding workloads on an Apple ARM PC?
Anything a windows PC does?
I... what? lmao
music, video, photography, and ML/programming stuff @tough grail
I have an x86_64 phone
Of course not, but the friction between China and the rest is quite bad for the company
god was it shit
those are probably the most common workloads for those kinds of PCs
I mean I guess? But it largely doesn't affect the consumer
geopolitics definitely can affect the consumer and it has in the past
Oh of course, but what I mean is that the world isn't going to allow China to own ARM. It aint happening
ML is owned by GPGPU segment. Video? Well, Hollywood uses Linux internally. Photography? Maybe yes, but let's see. Music? Maybe yes.
It will, indirectly. That may affect how the architecture evolves
More than likely, China would close off its own "branch" of arm and develop that in-house without the rest of the world seeing it
Yes but in this form factor there's nothing like it
Also there's a ton of places that use Macs for Adobe and their suite
Adobe optimizes for MacOS which is unfortunate
they could perform the same but they just don't and often get updates/features earlier
things like video decoding and import from external media is way faster on Mac because of it
But ML is only really useful if you can integrate it in other products. Apple is a very walled garden. They might make the iCar, the iPhone, the iSkateboard, the iRobot, the i....
Man I wanted a 2TB drive, but with the 32 core GPU being an upgrade option, a 2TB drive is gonna be well into the 4K territory D:
But ML is only really useful if you can integrate it in other products. Apple is a very walled garden.
This is unrelated
You're making a connection where there isn't one
ML is only useful if you can integrate it in other products
like the billion iPhones currently being used
Yes but the Apple walled garden part is irrelevant

Alright I'm ready to order accessories that are about as much as the 14" macbook pro
Man of culture, ordering the cleaning cloth
It's related. Practicality is part of a feature. Who would use it outside Apple? Only Apple and consumer of Apple hardware. So ML in Apple is just for Apple stuff
Im proud of you
No.
I am ordering apple's value products
You can develop and train ML models that can be used on other machines
What you develop on is irrelevant
Spoilers: models aren't platform specific.
Don't forget the custom engraving
Apple. But iSurgery product? Uhm. ML for a device that isn't used by really advanced stuff? An engineer or a scientist isn't going to use Apple products for advanced research, for example
Why wouldnt they?
Tell that to all the engineers or scientists who use Apple products.
But your ML models need to be improved and optimized for other architecture. And that's why Nvidia has monpoly
I don't think you can get any of those products engraved
Er, no
JESUS FUCKING CHRIST LMAo
That is truly sad news.
3080 MOBILE LMAOOOOOO
I bet a lot of pros will come back to apple hardware
@tough grail NVIDIA has a generic tensor accelerator. Any model can be uploaded to it
what desktop card compares to a 3080 mobile?
Hey Boob, do you have the new Mini?
You just have to translate your model information into the format NVIDIA's cards would understand.
Are you sure? TensorFlow and such? Despite of that, final implementations even live in custom chips
I believe its a 2080, aint it?
no I only have the M1 air
And everyone uses CUDA and such?
Ah.
And that's why Radeon has competition issues, because of CUDA and such
I would highly recommend the macbook air
They have competition more than just CUDA 😛
And everything is bootstrapped upon bootstraps upon bootstraps. It's very easy to develop on a Mac something that could run on these technologies. Just running it at the highest performance is an issue
I've had mine for almost a year
I might get new Mini with LTE as streaming device at work
Yes cause CUDA is proprietary
how big is your phone?
I wanted Air but hard to justify the cost for the small amount I will use
But things need to be optimized at final stage at least
I think if I needed one ipad in the future, it would probably be the air
Bro what the fuck. 3080 mobile is on par with a 2080 desktop lmfao
Sure but the actual training and development of the model can be done without knowledge of its final implementation
Yes, I agree. And Apple loves proprietary technologies. Look at Metal, for example
I have 13PM but it's hard to dick around with it at work on a stand connected to my mega dongs for lossless audio
so that's what 3060?
They hate Vulkan, OpenCL etc
Apple made the world fastest mobile GPU
i don't understand their hate for Vulkan
Actually, a mobile 3080 is even better, once you go above 1080P
I understand it. If you don't understand the hate of open standard by Apple, you don't understand Apple at all

it looks closer to 3060ti
just in rough numbers
a 3060ti is about 4x faster than the 1050ti
Sure, but they barely put development work into Metal to begin with. It's not like they're actively trying to compete. They just dont want it.
21 hour battery life, power of a mid range desktop gpu
Apple were the first to embrace HTML5 over Flash, to be fair.
and Apple pushed for USB-C
Haters link shitty $6000 desktop laptop with maybe an hour battery life and 60 hz display
lol.
Apple improving their architecture and making competition in the mobile sector is very good news, but there's too much hype.
I wonder if they are partially responsible for the fucking mess that is USB now
was it ipad or phone that killed flash?
Apple love
do you not remember how much people complained about not being able to run sites they wanted to on the iphone
and flash became optional super fast once the iPhone became popular
so it is basically a 3060- 3060ti
like the m1 was 1050-1050ti
But a 3060 Desktop is like a 2070 desktop
Im not sure actually
I never gave a fuck about the 3060, because it was shit compared to the 3060ti
Actually, you might be right
and still won't put it on the iphone
oof
I honestly believe we will see a portless iphone before a usb-c iphone
remember all the terrible flash websites everywhere in the late 00s
I don't think Apple see C as an improvment to lightning in the phone.
I hate lightning
But I understand that apple probably just wants to apple watch the iphone
and go portless
I wonder if the Pro phones will get C and the base models will go portless - that would be silly
So its between a 2070 and 2080
TLDW
Go the time stamp, youll see the info
yeah maxed out on usage an hour bat life, its 96whr, and 4k 60hz, you forgot the 4k part
But yeah, 3060 Ti is a lot better
We're also trying to speculate on apple graphs based on different parts, based on different parts again
so there is a lot of room for error
Oh yeah absolutely
and we have no idea what ACTUAL performance will be like
I still fundamentally disagree with moving to a portless phone. Not for anything with usability, but for efficiency. Suddenly losing 20-60% charging efficiency on tens of millions of devices world wide (that are charged daily, mind you), would be a total step backwards.
anandtech did a tiny bit on the m1
Its just fun to speculate 😉
inductive charging is not ideal for sure
Its basically a 1080 Ti it seems
Or a 2070 Super, it seems
Which I will GLADLY take in a laptop
I wonder what thermals will be like
Sweet jesus.
yeh, a watt or two wasted as heat the world over really is a efficiency disaster
That just sounds like concern trolling, tbh.
Yeah Id imagine the 14 runs warmer
I mean I got a little less than that for $1600 a year ago
I mean it's not a watt or two and it adds up quite a bit.
Of all the other things to be upset about, slightly less efficient charging is so entirely irrelevant
like I predict most of the time for light work, the fans will never come on in the pros
And it's not slightly less efficient it's significantly inefficient
Sounds to me like that the 14 and 16 will have similar sort of functions a la air and 13 inch mac book pro.
without a case I believe inductive charging is what, maybe 75% efficient @hexed sage ?
my m1 has gotten to 99 C
yikes
I've seen numbers closer to 40-60% @trim flare
when discord and the browser crashed at the same time
for 15w the cooling must suck
in typical use cases, like video, I think like 75 max
I mean looking at that, it's fairly clear that this is targeting a 3070/3080 laptop
@upbeat vale At the current rate of computational power usage, we won't be able to sustain ourselves by i think it was 2050 or so
What's the event?
MSI GE76 Raider
[citation needed]
Reducing power usage especially at idle or low-use states should be a priority for all devices.
From the same panel that made the california pc power requirements.
fuckin lol.
do you know how much power is wasted worldwide on chargers sitting idle
it's a fucking insane amount
BRUH
it quite is
Ok seriously
Again [citation needed]
In that graph above, Apple is comparing the GPU performance and wattage to the MSI GE76 Raider, which HAS a 3080 in it
like the arguement to even a fraction of a watt per user is like saying "10 million people wonder what a single bottle means being thrown on the ground"
it really does add up
This is INSANE.
Comparing the future energy usage of computing to, say, the electric car revolution is a little hilarious
The thing is every little bit matters
The world is moving to electric everything, computing is not the straw to break that camels back
especially the loses in wireless phone charging.
I mean apple is making pretty efficient devices
Computing is a gigantic chunk of our global usage, actually.
Bitcoin alone uses enough power as entire countries
and that's a small number of users.
Indeed.
Bruh. Imagine having a laptop that uses 60W and competes with an i9 and 3080 laptop
What the actual fuck
what did they use to measure performance
Some shitty and vague graph I linked above
also interesting note is that the msi laptop has a much lower resolution screen
We will need to see what happens, but the graphs in the past have been really good indicators
Whats the resolution listed?
1920x1080, IPS, 360HZ, 17.3IN, 3ms g2g
So you mean to tell me
?
a 3080 at 1080p, is barely winning against native resolution damn near 4k 32core GPU?
What the fuck.
they probably took that into account
well, it could be userbenchmark level software or 3dmark, we dont know
Apple tends to be quite representative with its marketting graphs.
Because they know they get torn apart by nerds anyway.
They may be vague as fuck
Not to say they don't cherry pick, but its never completely unreasonable.
But they usually do well with their graphs
In terms of representing their product I mean
And how it actually performs to the competition
lpddr4x iirc
Interesting
but intel is going to be using ddr5 in a few weeks too
Yeh, the M1 used LPDDR4
Also the Max has twice the on-die SLC. Wonder what sort of effect that will have
" essentially it’s no longer an SoC with an integrated GPU but it’s a GPU with an SoC around it."
interesting take anandtech
https://images.anandtech.com/doci/17019/2021-10-18 19_22_58.jpg
So in this picture, they are comparing to a lenovo legion 5 82jw0012us, which has a 3050 Ti.
Er sorry thats for the 16 core GPU
I said that earlier
lol
So in the hand-waviest 32bit FLops metric, the Max is a desktop 2080 but in a 60w mobile part?
AMD advertises 26.8bn transistors for the Navi 21 GPU design at 520mm², Apple here has over double the transistors at a lower die size.
According to what has been released, yes
with about the same memory bandwidth too
Thats's still amazing.
Thats for 3080 mobile
M1Max 32 Core GPU offers 10.4 TFLOPs compute, 327 GTexels/s and 165 GPixels/s rates.
I am unsure exactly where those numbers are from, but thats the info we have atm on the 32 core
So by all respects, the 32 core should be on-par/better than a mobile 3080 lol.
On paper anyway
Almost half the maxmium compute of the 3080 Mobile
it ain't going to be better
Mighty fast, but not that fast
Yeah
Actually yes
thats true
So here's what they compared it to
for the 32 core GPU
And you can see the lines are just a touch below the 3080 there
Im trying to find what wattage that 3080 is at
From there you can just cross reference the two to get an idea of the performance
155 watts according to their 10UH review
Man that thing is ugly

Our test unit is a higher-end configuration with the Core i7-10870H CPU, 155 W GeForce RTX 3080 GPU, and 300 Hz 1080p display for approximately $2900 USD retail. Lesser SKUs with the Core i7-10750H, 130 W RTX 3060 or 140 W 3070, and 240 Hz or 144 Hz panels are also available.
That is per the 10UH
And the GPU is the same with the 11UH
So yeah, the line is exactly going across the 155W 3080
Also, interesting facet here. If you get a 3080 that uses less than 120 watts or so, you're looking at equal performance.
According to that graph
So yeah, that 32 Core is going to look like a monster compared to the rest lmao.
I wish I had the money to buy this
STS agress with that take https://twitter.com/stroughtonsmith/status/1450197888286576643
Good old apple credit card lmao
Give it better game support and I don’t need a desktop anymore
With what take?
OH AGREES
LMAO sorry
I misread. I'm sorry
I thought you said disagrees
😛
Sorry I had to shut you down there
So yeah, it looks like M1 Max 32 core GPU ~ 50 watts then
With the RTX 3080 mobile they used being at 155W
Is it?
50-60 yeah
In terms of performance, Apple is battling it out with the very best available in the market, comparing the performance of the M1 Max to that of a mobile GeForce RTX 3080, at 100W less power (60W vs 160W). Apple also includes a 100W TDP variant of the RTX 3080 for comparison, here, outperforming the Nvidia discrete GPU, while still using 40% less power.
Next important question: how the fuck?
I can see how they outplayed Intel, but outplaying Intel and nvidia at their specialities
in one product?
@upbeat vale oh my fucking god lol. The top comment on that anandtech link?
Yep, I saw it.
Are we all simping the MacBooks
Oh dear lord that's extremely sad to think
Whats amazing is that we still havent even dissected the CPU performance.
OH YEAH
I also just remembered
Honestly if you think about the fact that a MacBook Pro replaces a desktop
We are assuming M1-era power and efficiency cores. So it probably won't be suprising.
There's new "high powered" mode coming to macOS
It’s really cheap
Yeah I dont know how I can justify building a new PC now
Exactly
Unless they have thermally uncapped them and can clock them ⏫
I mean, windows aside of course
A a MacBook Air plus desktop is the same as a MacBook Pro
Like if you need windows for a certain app, thats different
And at some point it will get more gaming support
I can hope steg
I think this level of hardware is too good to ignore
We can see what sales are like, but I suspect some devs will start porting depending on what market share looks like
The real value is a Mac mini with this
I'll concede this point when Warzone comes to the Mac
The biggest issue, imo, is that Apple needs an easy way for users to port over
Anand talked about the 2014 MacBook Pro replacing his desktop
Look how expensive GPUs are though
While true, I think even the base 14 inch is still pretty nice
The only way to get that performance is a prebuilt
It is, with the current GPU market yeah
You cannot just go and buy a gpu right now
And who knows how long this will last for
The base performance is 1070 level?
Apple might have even gotten lucky with the timing there, given the price gauging goin gon
Hmm, will 32GB on the Max be quad channel?
12 gpu cores I think
This is an important question.
Unironically, I think this could help some of the price gauging issues
for GPUs
Typically is so little overlap between those two
But people may be thinking "well shit if I can get a laptop for the same cost as my desktop and more or less get the same power, might as well! 🤷♂️ "
How hard is it to get a laptop with a 3080?
According to that 10UH MSI one I linked earlier, apparently hard
Honestly looks fine
Its weird to have the notch in any case
But I dont hate it if Im honest
If these could mine at all
I don't think I hate it
People would buy them
but it is very wierd
It’s weird
That is so much notch just for a single camera
It makes sense
I mean.. like WHY?
But it’s huge
even if its a camera and an ambient light sensor
because nostril cam isn't appealing
I guess so
oh my bad, didnt see that
but yeah, totally gpu > cpu moment
which is something I like
I might actually have to do some work today
CE will likely get Monterey into Braveheart's before the end of the year, or shortly after the new year
Translation: we might not get approved to use Monterey until next year.
so ordering a new laptop now is probably a waste
Because they made some awesome laptop CPUs @calm sail
M1 is now up to 16/32 cores, with a stupid powerful iGPU
And the rest of the laptop looks great too, added a bunch of ports (finally), miniled display, adaptive 120hz, 32gb on-die RAM
And 65W total package, 30W for the CPU.

New 16-Inch MacBook Pro is Thicker and Heavier Than Previous Generation https://t.co/hp3XLsOCrZ by @SamiFathi_
485
That side profile isn't trendy and modern, for sure
You could almost call it a big middle finger to Ive-era laptop design
Its all ive ever wanted ❤️
Just got my apple credit card to pay for the 16 inch XD
who wants to be stuck emulating real programs all the time though?
To be fair, when the performance is so close to native, I doubt many would care
most apps that people use on Apple are native now
And they have hardware x86 translation which is a huge plus to translated performance
But also, things will get ported over
why not just use an ipad then?
i think it's within 5% at the best 15% at the worst
are you trolling
it's not just performance it's the stability and bug fixing issues that come with emulation
And that's where I'm out
you can get a keyboard for you ipad
haven't heard any issues of this sort
ok yeah you're trolling
Tbh, he probs isn't, but this is uh. wow
I am a OneDrive user with two accounts, having two instances of OneDrive running. It's been working great from Mojave on, up until now. I had two OneDrive icons in the menu bar, and could access either instance in finder.
Just wrapped up the 11.6 upgrade, and upon reboot, I can only run one...
doesn't work properly with onedrive
i've only seen it on the m1 computers
steam issues too
Interesting
I'd imagine that companies will likely do more and more native stuff as times go on
this article doesn't even answer the question
it says nothing about issues as far as i can tell
just says when optimized games may come out and how to run them
there were massive issues early on where it'd just crash loop
Damn the shipping dates are already well into November
Yikes
Might just wait for some stock at my local best buy or something
you see the M1 Pro now has 200GB/s memory bandwidth to feed them there 16 GPU cores? and the M1 Max is rocking a 512 bit wide memory bus.... in a god damn laptop!?
We did indeed 😉 haha
Not as fast as a regular GPU. Where you're expencting 500GB/s+
a 3070 is 450GB/s....
and the difference in transistor count from M1 Pro to M1 Max is 23.3 billion transistors....so thats 23.3 billion transistors in 16 more GPU cores, 256 more bits of memory bandwidth, and a couple extra encode/decode units and some other random little stuff....
3080/3080Ti/3090/A6000 die is 28.3 billion. 3070 die is 17.4 billion. so the difference between an M1 Pro and M1 Max is somewhere between a 3070 and 3080 in transistors....
I care less about the GPU here, until TVM supports metal/CoreML its pointless to me
but I am keen to know how the CPU scales power wise
yeah...
The M1 was comparable with a 5700u at load
@pallid veldt you doing machine learning?
ahhh
My other question was..... how much difference all that memory bandwidth will make for some database workloads.
Not sure it'd matter that much, having all of that cache will be nice though
Problem is, these are awful for databases because their IO is ass. Unless its timeseries, its just not useful
Even then, the software side of things makes that a non starter
I waited too long to get my order in
yeah me too
At this point, Ill just wait until stores have em
except im still waiting
What do you mean?
waiting for benchmarks
Ah
Yeah at this point I'm going to do the same
Since the shipping times are so long
Now its gonna be painful waiting the week for benchmarks to come out 😛
Im honestly shocked we dont have any sort of benchmarks or ANYTHNG right now aside from Apples own stuff
There's usually something
review units will probably arrive tomorrow
so expect results to start leaking out soon
I hope so
I cant wait a week for reviews 😛
@upbeat vale Whos your go to for reviews?
I'll read lots. I don't care too much for the medicore tech press.
Either give me Anandtech or give me more mainstream - Ars, The Verge, etc.
I'll read Joanna Stern as soon as she publishes
ditto John Gruber
Im excited lol
I wonder what Linus and Dave2D are going to say about the new MBP's lmao
yeh, see they go in the "medicore tech press" category
and IDGAF what they have to say
God Linus used to be good
No
Non-technical isn't necessarily a critism
Joanna Stern doesn't write technical reviews
but they are some of the best reviews out there
Vapid
Yeh, I agree
There's nothing of substance covered in the review
I dont need you tell me that the laptop has 400Gb/s bandwidth or whatever
But I also dont need you tell me that the particular color of blue they are using looks good
I think that sort of person is so scared by the internet into being "objective"
and become bland as a result
But its what Dave2D does
Like
"The panel looks fine and is mostly color accurate." is probably his most uttered sentence lmao
If everything looks fine, how am I supposed to know from your review which is better than the other?
We have tools to measure these things and you don't need to be super tech literate to understand them
Yes and no. "Its fine" is a perfectly good review conclusion.
Like, if its 2% less accurate than some other model is that actually useful information for most users?
and if you are a colourist who cares deeply about that you aren't reading a medicore tech youtubers opinions for your buying advice
Its not so much that i need the best panel
Its that I also dont want to buy a shitty color'd panel when the reviewer just says "its fine."
I need more context than that
You dont buy a product based off of "its fine."
lol
Well, not expensive ones anyway
I mean sure, for some of the sub 700$ laptops he reviews, fine, whatever. I just wish he was a bit more details and provided more context to how he formed his opinions
Yeh
But man I am really excited to see how this chip works. Guesses on geekbench score for multicore?
Lots.
Heaps even?
Do you think youre gonna love it?
Still don't know how I am going to acquire one yet
As in money?
I got an out-of-cycle upgrade to the M1 13"
Someone already posted one.. like... 11k or somethin?
Might be a challenge to also justify a 14"
Wait wot? can you find that please?
Benchmark results for a MacBookPro18,2 with an Apple M1 Max processor.
Also, how good is 11k?
I could just buy myself one, then I can use a proper UK keyboard not the stupid US one
(Me having paid attention about 5% of the time)
Im not aware they have hyper threading
11k is around 12 cores Xeons
Mac Pro (Late 2019)
Intel Xeon W-3235 @ 3.3 GHz (12 cores)
11876
lol.
In a laptop
what a time to be alive...
There we go
Yeah its also one result
Will be interesting see how the thermals do..
But my lord this is some wild performance
i think the thermals will be a lot better than the old intel ones
Given the beefier chasis and better fans + no dedicated GPU
The M1 13" is actually astonishing
Id be hard pressed to think itll be bad
passive is neat
coming from the Intel laptops the thermals are really amazing
so these Pro models should be similar
🤞
I haven't own a laptop with an aluminum/metal chassis, but you really notice it with this framework
How so? What's the difference
oh yeh, sweaty palms
Yeah
all the laptops I've had were plastic, so the heat transfer wasn't the same
The metal on your hands
Sweaty palm rests, mostly
Its annoying lmao
First glass trackpad for me, it's godlike
well, not the event, just the product
Yeah that's kinda my point
After seeing just how many improvements are coming to the Apple ecosystem so rapidly
Looking at Google i feel completely lackluster and left behind
Seems like it
Even the fucking Windows ecosystem is more cohesive than Google's.
AMD Ryzen 9 3900X
3.8 GHz (12 cores)
11730
Remember when this was the best desktop CPU you could buy for productivity? lol
Reasonably, that is
I have easy integration with Office 365, AD, multiple work accounts, email+calendar services, settings syncing, Your Phone, etc
TR 2000s were out
They were also like double/triple the price
they were best for productivity yeah
This in no way competes with TR
Right
Google's "ecosystem" doesn't mean shit when they keep shutting down products every month and spawning new inferior ones every two
TR 5000 when
rip google glasses
Yeeeeeaahhhh
This is already beating out desktop CPU's
Like, this score beats out intels top of the line consumer CPU
Intel has left the chat.
It's also a rapidly changing environment
Remember a lot of that is probably from the on-die memory which can easily cause synthetic benchmarks to have inflated scores
Excuse me while a 30W CPU beats a 220W desktop processor KEKW
Oh yeah of course
I'm meme'ing here
I just find it really amusing is all
Like, I went from a 6600u to a 1165g7, I was blown away that I could just double my per core and multicore
(roughly)
Yeah its crazy how much AMD and Apple have been changing the CPU/GPU space
I couldn't really care for desktop alderlake thou, mobile alderlake is where it's at
Energy efficiency #1
Do you think alderlake will be able to compete with that though?
With what apple just announced?
Uhhhhh
fuck
Thats a desktop chip
I google hk
fuck you google
Hahahahhaa
But stlll
thats the desktop version barely beating the laptop version though
Or was that not alder lake?
So who wants to finance my macbook, cries in empty wallet
Get a regular m1 macbook
ez
Ah
I would argue intel had decent integrated graphics
They improved them for a lot of the MacBooks
Apple really bullied them with CrystalLake
Wasn’t it a custom chip?
It turned up in a very small number of PC laptops
and the original NUC I think
but yes, it was an Apple-specified design
Now we have the hardware
We just need the software
It will be interesting to see what other companies can make for internal cpus
XE has really surprised me
I would think that with the amount of ram it had dualbooting windows should be an option
But maybe the new way to do other machines is via azure
What is that
I have not seen benchmarks
When intel made sandy bridge they vastly improved the igpu
Anand talked about it quite a bit
I played a bit of satisfactory on it, was quite surprised it was even playable
I liked the 8250u in the laptop I had
Until I used my a12x iPad Pro
Then it didn’t seem too impressive
It’s pretty insane what the m1 can do
It’s basically gaming laptop level performance from just a few years ago
And then you have the m1 pro and max
I wonder what 5nm amd will look like
And smaller Samsung
Which gen was this?
I'm honestly extremely excited to see whatll be like with the A15 cores. Jesus dude. The 32 core GPU is gonna be ACTUALLY a 3070 desktop or something lmfao
a melting chassis issue, that would be a good one
Not if the power consumption is low 😉
Also, that would require like what, 1000C or something?
Aluminium/Melting point
1,221°F
660.3°C
Ok so 660C, but still
lol
maybe
😮 POG
Which doesn’t even make the price crazy
Yeah hahahha
Thats actually a deal at that point
I wonder what future gpu prices will look like
Oh! I don't need the extra power. I wanted the other features and hoped for some longest battery life with less numbers of efficient cores, but that didn't happen. And given the massive price increase, it would be a waste. so I'll wait another year for this one to refresh
I'm not a fan of the design either. I think the one I have looks way better... and that notch :<
Im not a huge fan of the design either
But the notch really doesnt bother me as much as I thought it would
Oh yeah, i didn't look at the spec, is it 3:2 aspect ratio?
Skylake I think
Otherwise its a really funky aspect ratio
Er thats 6th right?
Yep
Interesting
never had a mac then
You cannot use nm measurement across manufacturers, as they don't agree on what to measure
This is incorrect right?
As in, the measurement is fine
But that's not the point of the measurement
As in, just because its lower nm doesnt mean its inherently better
There was no context applied to the quote
The nanometer number quoted on silicon manufacturing processes is a name
Not a measurement
Which is why neither Intel nor TMSC actually name their processes after a number of manometers anymore
“Intel 4” used to be the “7nm” Intel process.
The numbers don’t mean anything
Look at the table on this page https://www.anandtech.com/show/11859/the-anandtech-coffee-lake-review-8700k-and-8400-initial-numbers/2
The 14nm in 14nm++ doesn’t actually refer to any significant feature
So it’s not a measurement of transistor size any more?
Interesting
🍎
Yeah the nm in most processes really doesn't tell you a lot
it's so complicated it's difficult to say what exactly it means
apple moment
Benchmark results for a MacBookPro18,2 with an Apple M1 Max processor.
Heres another geek bench score
pretty tasty

to think you can get a 2080 in 40w now
Lol
Goes to show you don’t need to suck out 600w out of the wall for video games
*assuming games will run
ftfy
though I am now very interested in doing a hackintosh build
what now the software wont support intel hardware.
Qualcomm processor 
I've seen a few speculative estimates without any data that believe the m1 max is 3x faster than the m1 not 4x
But the only thing that will show the results is testing when these machines arrive to those people
