melgross

About

Username
melgross
Joined
Visits
127
Last Active
Roles
member
Points
10,978
Badges
2
Posts
33,723

Reactions

6.3KLike14Dislike735Informative

  • Apple Silicon Mac Pro could combine two M1 Ultra chips for speed

    melgrossmelgross
    melgross said:
    melgross said:
    Allow me to add that I don’t quite get the argument that it has to be M2 — I gather there is a technical reason for it, but I think that’s hard to say without knowing a lot more than we do. I find the idea that Apple would design this entire M1 line but not account for the Mac Pro to be absurd. 

    In terms of naming, I don’t think they will call it the Ultra Pro or Ultra+, they will all be Ultra, just with different core counts. Basically an Ultra is 2 or 4 Max fused together. 
    It’s pretty clear that they are finished with the M1. Will people please stop making new M1 chips up? They may use two or even four M1 Ultra chips. They may change their concept of introducing more powerful chips over the year, and have an M2 Ultra for the Mac Pro. We don’t know. But they won’t have a four chip Ultra. John made that pretty clear.
    John was introducing the UltraFusion process when he said that. That process may also allow them to connect two Ultras together, much like what you’re suggesting when you say they may use multiple M1 Ultras in the Mac Pro. It’s a plausible technical solution to the problem. That’s the whole point of the “chiplet” approach. I think you’re getting hung up on semantics, although I’ll concede that it was not a live event and everything said was carefully reviewed.

    It isn’t hard to imagine how John would introduce the idea, “UltraFusion not only allows us to fuse two M1 Max together and create the M1 Ultra, but it also allows us to connect two M1 Ultras together …”
    Look, he made it pretty clear that the Ultra was the last M1 chip. I don’t know why people insist that isn’t true. He didn’t say it had anything to do with Ultra /fusion, or anything else, just that the Ultra was the last. Earlier on, when they announced the Pro and Max versions, they could have said that too, and then popped out the Ultra with the UF connect, and acted as though it was just the same chip.

    but they didn’t. And like it or not, that means something. What would have wrong with not saying anything? It’s not semantics. Semantics is something that’s interpretable. A definitive statement is just that.

    the other thing thats] you guys are forgetting is that the cost of making these chips increases more than the added area because of increased defects and the risk of unusable chips. The greater percentage of wafer area a chip takes up, the more the cost. It’s a $1,000 upgrade to go from a 48 core Ultra to the 64 core version. And the cheapest Ultra costs more than twice what the Max version costs. So, even if they would do it, this new chip would likely cost at least three times as much. Maybe more. would that be worth it? I’m not so sure.
    The good thing is we’ll know soon enough. There’s no way this uncertainty is prolonged past WWDC. 

    But reading your last paragraph here makes me think you haven’t looked at the mock-up I’ve been referring to, in conjunction with the patent about this packaging tech Apple filed in January. It’s two Ultras stacked on top of each other (back-to-back)—doing so doesn’t change anything about the wafer layout for making the Max/Ultra. That’s why, no matter how it works, it can’t be considered a new chip. Because it’s not. The Max Tech video that someone pointed to is goofy YouTube sensationalism, but the reading of the patent seems accurate. 
    So, I don’t remember seeing the stacking that you’re referring to. But Apple’s interposer is part of the silicon and makes the two Max’s into an Ultra, one die chip. If this stacking isn’t doing that, but is instead just taking two Ultra’s and somehow connecting them together without turning them into one die, as with the Ultra itself, then that’s just two chips. It’s not another chip. I’ve been saying that Apple could do that in some way to offer more performance from the Mac Pro. Not double the performance, as in many tasks, as we’re seeing with the new ArsTechnica review that came up today, but pretty good. I don’t really care about mock ups. They’re just guesses, and rarely, until just before release, often don’t mean much. After all, they’re called MOCKups for a good reason.
    muthuk_vanalingamwatto_cobrafastasleep

  • Leaked M1 Ultra Mac Studio benchmarks prove it outclasses top Mac Pro

    melgrossmelgross
    Just to add this. I was a partner in a commercial lab for many years. At peak, we had 32 Macs in our graphics/photo/video department. For the 12 we used for current top flight work, I had bought Barco graphics/photo monitors. In the early 1990s, when I bought them, they were $16,000 each! Expensive. Today, they would go for $32,000, accounting for inflation. And that is what the top monitors for broadcast and other installations go for, though some are as high as $40,000.

    I just shake my head at people who just have no understanding why we MUST buy more expensive monitors for serious work that involves any graphics or particularly photo/video work. Just go to the sites of the graphics monitor makers, and the prices can be shocking. Check Eizo, or any other pro monitor maker. And by pro, I don’t mean “business” monitors. Those are usually just stripped down cheaper consumer units.

    so if a $300 monitor is all someone can afford, or is willing to spend, then that’s fine. Just don’t expect any professional quality work to be possible in any graphics/photo/video job. It won’t happen. You can’t properly color grade on a monitor like that, no matter how carefully you calibrate it, because the color gamut isn’t good enough for anything beyond sRGB. And even then, it may not be adequate, with blotchiness across the screen, uneven contrast, etc.

    Apple’s new monitor gets us an additional 100 nits above what the 27’’ iMac monitor gets, and also 100 bits above what the cheaply made LG 27” does at $1,300. That also a much simpler monitor, with none of the features of this new one. So while it’s not true HDR, it’s better than a lot of others at about the same price. 
    thtmuthuk_vanalingamfastasleepDetnatorlkrupptmaywatto_cobra

  • Apple's Mac Studio launches with new M1 Ultra chip in a compact package

    melgrossmelgross
    UrbaneLegend said:
    @melgross ;
    Clearly you didn't read my post carefully or the points flew over your head.

    Developers struggle to make effective use of threading in applications and even when they do there's a hug drop off in effectiveness after a few threads. I'd like to know what current workflow can keep 20 CPU cores busy on a consistent basis. I'm not talking about CPU rendering, it's 2022 most freelancers like me are GPU rendering.

    Apple has offloaded much of the work of the CPU onto fixed function units on the SOC like compression and decompression and also machine learning so there's even less need for large numbers of CPU cores than an equivalent PC would require. Apple is clearly marketing the Studio to creators, their whole presentation contained hyperbole from creatives and someone running office applications will be best served by the Mac mini or a Macbook. Let's not pretend there's a huge science and engineering market for Apple, there isn't.

    As a Davinci Resolve user I know just how reliant the app is on the GPU, even with beefy GPUs a complex effect chain can bring Resolve to its knees while not touching the CPU. Resolve is probably the best optimised app for the M1 and it will expose the relatively weak GPUs as the main performance limitation. 3D work is even more biased towards GPU performance with interactive viewport performance and rendering reliant on GPUs. Even in 2022 apps like Cinema4d are one thread wonders when interacting. Houdini and Blender are much better with threading but the law of diminishing returns kicks in very quickly double the CPU cores does not result in double the performance even when cooking sims.

    I bet there will be post production facilities waiting for the Mac Pro just to get double the GPU cores and have even more CPU cores doing absolutely nothing. Apple's boast that the Studio is faster than the Intel Mac Pro's 5700 is really no boast at all, not in the real world not at this price point. The 5700 is equivalent in compute performance to an nVidia 1080Ti so not exactly cutting edge.

    I've owned several Mac Pros in the past and used them for 3-4 years as I've been able to upgrade the GPU. In 3-4 years the GPUs in this Mac Studio will look silly compared to the state of art and as such the Studio will get old very quickly and having just 64 GPU cores is quite an Achilles heel.

    Some real world benchmarks of the new Blender 3.1 put the relative GPU performance into perspective. Apple did the development so no one can say this is down to a poor implementation.

    M1 - 198
    Pro - 363
    Max - 702

    3090 - 5566
    1080Ti - 839

    If we're charitable and double the Max score to 1400 that puts the Ultra at around an nVidia 2060. If we double the score again to give an indication of what a Mac Pro might perform like that puts it below a 2080 or 3080 laptop performance.

    This just isn't up to the level required for 3D artists to jump back on board Macs.
    Your points didn’t go over my head, most are just wrong. Video does use as many CPU cores as there are available. That why there as so many high core count CPU’s in the first place, other that for servers. There are video tasks that go through the encoders/decoders, and the rest of the chip as well. 

    As far as engineering goes, you’re wrong there as well. NASA for example is mostly Mac, and has been for many years. Drug companies use Macs for their R&D programs. There are number OSs other examples. As Apple’s performance has v]come back up, we’ve been seeing more CAD/CAM appear, particularly in architecture. I’ve been using Bently CAM for many years, and now use Fusion 360.
    muthuk_vanalingamfastasleepwatto_cobra

  • Hands on with Apple's new black and silver Magic Keyboard, Trackpad, and Mouse

    melgrossmelgross
    JinTech said:
    $199 For a keyboard? Yikes! Also shocked to see the mouse is packaged with plastic!
    You think that’s a lot, do you? The original Apple Pro keyboard cost $289 back in 1992, that’s $580 today. That was a great keyboard. I still have mine, though obviously I don’t use it. The IBM keyboard of the same era cost $329, and that’s $657 today. This isn’t expensive. But it’s one of the better ones out there.
    StrangeDayswatto_cobraAlex1N

  • Leaked M1 Ultra Mac Studio benchmarks prove it outclasses top Mac Pro

    melgrossmelgross
    keithw said:
    As impressive as this is, (no flames here- I'm an exclusive Mac user)  you have to get to Page 388 on the Geekbench 5 CPU results to get to this result of 24055.  The Ryzen Threadripper 3960X 3800 MHz (24 cores) from October 2020 already beat the Ultra.  The single core results are outstanding, and this much horsepower in this form factor with so little power draw is very impressive.  The ASi "Mac Pro" will have a pretty high hurdle to climb with the AMD EPYC (128 cores) coming in at 75539 GB5 multi-core.

    There are, of course, other factors to consider such as the unified memory, built in encoders/decoders, neural engine, etc.  All in all, a lot of bang for the buck.  All that said, my AMD Radeon RX 6900 XT eGPU (attached to my 5 year old iMac Pro) got a Metal GB5 score of 160166.  So they have another hill to climb in their ASi "Mac Pro" when it comes to graphics performance.
    A mistake that some people are making with the scores is that Apple isn’t making a chip that (yet) challenges the massive multi core chips we see in some places. And remember that most of those scores are from people testing the same chips in their computers. It’s not as though there are dozens upon dozens of these higher performing chips out there. There are a few chips in a lot of machines that people are testing over and over.
    tmaykeithwbulk001watto_cobra