Mac Studio
- Thread starter Cmaier
- Start date
- Joined
- Jan 17, 2021
- Posts
- 2,057
Sonofabitch
Colstan
Site Champ
- Joined
- Nov 9, 2021
- Posts
- 822
The rumors have been all over the place. Unlike the PC side, you need a degree in Kremlinology to divine Apple's intentions. I wonder if this is the rumored 7K display, alleged by 9to5Mac, that will supposedly replace the XDR, which Ross doesn't believe it to be, or the monitor that sits between the Studio Display and XDR which was rumored to be $2,500. Regardless, unless you need a new monitor immediately, it might be wise to wait until after WWDC.
- Joined
- Sep 26, 2021
- Posts
- 6,318
- Main Camera
- Sony
Yep. It may be a $2500 display with promotion and miniLED, which makes for a very interesting product. In my case that would be too big a price jump for those features, but, then, I’m not a pro in an industry where I do video or graphics work.
Andropov
Site Champ
- Joined
- Nov 8, 2021
- Posts
- 713
It's not very secret by now.
Older iMacs and Apple displays could change between VESA and built-in stands. Bit of a shame they removed that.
Nycturne
Elite Member
- Joined
- Nov 12, 2021
- Posts
- 1,392
I’m not entirely sure where you got the 18K number for Nvidia. The 3090 claims 10496 CUDA cores, and the A6000 isn’t much bigger than that. https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3090/
@leman at tOP said:
While not getting there the same way, the math matches what AnandTech guessed at the time of the M1 release (since Apple does not share details): https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested/3
So it’s not 4K vs 18K as far as I can see, but ~8K vs ~10K.
DT
I am so Smart! S-M-R-T!
- Joined
- Aug 18, 2020
- Posts
- 6,435
- Solutions
- 1
- Main Camera
- iPhone
There a couple of fucknuts on MR that literally spent all day in that display thread, not contributing in any meaningful way, simply "liking/loving" every single post critical of the design, from morning, till evening.
Wow.
Don't get me wrong, I think that's a less than optimal design for the stand/mount, in an otherwise stellar looking product. One of the premium accessory manufacturers like Twelve South, should build a VESA compatible stand, that's matches the design.
Anyway, that's about all I have to say, and it took me a couple of minutes, one time vs. my __whole__freaking__day__.
Wow.
Don't get me wrong, I think that's a less than optimal design for the stand/mount, in an otherwise stellar looking product. One of the premium accessory manufacturers like Twelve South, should build a VESA compatible stand, that's matches the design.
Anyway, that's about all I have to say, and it took me a couple of minutes, one time vs. my __whole__freaking__day__.
Yoused
up
- Joined
- Aug 14, 2020
- Posts
- 6,840
- Solutions
- 1
It is possible that I was seeing some leaks/speculation on the 40xx series and conflated that with the 30xx models.
M1 Ultra's specs show it as having 32-core Neural Engine, but since it's composed of two M1 Max, we know it really has 64 ANE cores in four clusters of 16 per cluster. If they needed to do any work on load balancing or APIs, it's already done. That means if they were going to enable all the clusters, they could've done so already.
januarydrive7
Active member
- Joined
- Nov 8, 2021
- Posts
- 37
The interconnect at the bottom of the M1 Max wasn't exactly a secret, either (although it was less obvious, as Apple's die shots cut it off entirely).
D
Deleted member 215
Guest
$2500 might be a low estimate considering the 32" XDR display is $5000 and this one would have a 12 MP webcam and speakers. But maybe the price of the XDR display will be coming down soon...
Andropov
Site Champ
- Joined
- Nov 8, 2021
- Posts
- 713
Good point.
- Joined
- Sep 26, 2021
- Posts
- 6,318
- Main Camera
- Sony
At least one other person here did. Looking forward to your reviews!
dugbug
New member
- Joined
- Mar 12, 2022
- Posts
- 2
- Joined
- Sep 26, 2021
- Posts
- 6,318
- Main Camera
- Sony
Good! Random chess benchmarks are obviously the most important thing
Colstan
Site Champ
- Joined
- Nov 9, 2021
- Posts
- 822
Over at the other place there's open warfare about this issue. There isn't even agreement about exactly what Apple was trying to portray in its vague performance slides. Notice that Apple is using "relative performance" on the y-axis, instead of simply raw performance. I get the feeling that an engineer was trying to explain a performance/watt advantage in regards to the Ultra relative to the 3090, to someone in marketing, and then marketing ran with it, resulting in this thing. I'm not certain exactly what Apple is trying to communicate here. The GPUs inside the M1 SoC family are already impressive enough as is, particularly considering that it's a first-generation product. So why bring up the 3090 at all, and instead continue to compare them to the previous GPUs inside Intel Macs, which seems to work heavily in their favor?
On top of that, as you said, all of the competent reviewers that do in-depth analysis are no longer in tech journalism. That leaves us with short synthetic benchmarks like Geekbench which have little real-world utility, including compute results using the now defunct OpenCL, GFXBench off-screen tests which have similar issues, Cinebench in which Apple Silicon doesn't appear to be utilized to its full potential, and then of course gaming benchmarks which are considerably hindered because they are running under Rosetta 2. These benchmarks may be useful when doing a quick comparison between chips in the same family, but trying to divine anything meaningful when compared to a different platform is questionable, at best. Heck, at this point using a handheld stopwatch while running the latest beta of Baldur's Gate 3 would be more useful when evaluating the graphics performance of the M1 Ultra when compared to what has been presented thus far, because that's at least a real world game that is ARM native and optimized for Apple Silicon.
Unfortunately, the situation isn't going to get any better without someone to step up like Andrei or Ian, who both left Anandtech recently. I'm sure most of us remember how the tech press and PC partisans lambasted the A15 for being a warmed over A14, until Andrei proved otherwise, demonstrating that the A15 was in fact a substantial improvement over the previous generation. I'm going to defer to @Cmaier in his belief that Apple has actually been underselling the performance of Apple Silicon. However, when it comes to marketing the Ultra, Apple hasn't done itself any favors with these vague graphs. Maybe they're useful during a snazzy presentation for Tim Cook and his lieutenants to show off to the general public, but they have little utility for tech literate nerds who are attempting to judge actual performance, rather than "relative performance". At this point we'd have as much success consulting a witch doctor practicing haruspicy while reading chicken entrails.
