There is not one M2 ultra in the top 100 wildlife scores. top score 4090 = No1 = 272537 number 100 = 194708 and no M2 ultra's in the wildlife hall of fame.
any links to these scores from the M2?
any links to these scores from the M2?
No ideas, I wasn't making any claims about Wildlife Extreme. Someone previously said the m2 ultra was equivalent to the 3090 in 3dMark at much lower power usage and I guess it would have to be a comparison in 3dMark wildlife or Wildlife Extreme.There is not one M2 ultra in the top 100 wildlife scores. top score 4090 = No1 = 272537 number 100 = 194708 and no M2 ultra's in the wildlife hall of fame.
any links to these scores from the M2?
No one has claimed 3dmark would beat or be close to the 4090. Claim was it would be approx 3090 which seems accurate.i cant find any M2 ultra scores in wildlife, and nothing in wildlife extreme. only thing in wild life extreme is an M1 Max with a score of 10017 so if the M2 ultra does 5 X that at 50.000 say, which i dout, the 4090 with a 13900K scores average 92.500 but over clocked can get 107.594 in the same test wild life extreme.
I just downloaded the IOS version of wild life extreme on IOS and looked at the IOS hall of fame, cant find any M2 listed. will keep an eye on that, as i dont think M2 ultra is going to be any where near a 4090.
No, it's running the iPad app, which renders at the same quality and resolution as the PC version.So the M2 ultra is running 3D mark in VM windows? and is on par with a 7900xt and 3090 i would like to see those results if you have a link? the CPU score would be interesting alone.
I find the scores of the 4090 suspicious. Ok it's fast, but is it really twice faster than the 3090?
4090 is a 450W part while the M2 ultra GPU portion is approx 70-80W.It is. 4090 had a process update over the 3090. So they bumped the power and made it more efficient.
That also means that cards like the 4080 and 4070 are also in 3090 territory - while being more mid level cards. So 3090 performance is the current gen mid level performance on the PC side.
4090 is a 450W part while the M2 ultra GPU portion is approx 70-80W.
For 3D raster performance it cannot match the 4090, but is more than 50% of the speed for 1/6 the power.
The 4070Ti is a 280W part and has the same performance as the m2 Ultra on this benchmark.
Another data point is gfxbench:
GFXBench - Unified cross-platform 3D graphics benchmark database
The first unified cross-platform 3D graphics benchmark database for comparing Android, iOS, Windows 8, Windows Phone 8 and Windows RT capable devices based on graphics processing power.gfxbench.com
The M2 Ultra matches the RTX4080 and is approx 65% of the 4090.
The 6900XT is slower in the 4k offscreen test, if you consider that the vast majority of the results indicate about 270-280 fps, and that outliers are probably overclocked cards or bogus results.It doesn’t match the ancient amd 6900xt.
The 6900XT is slower in the 4k offscreen test, if you consider that the vast majority of the results indicate about 270-280 fps, and that outliers are probably overclocked cards or bogus results.
That 336 fps result, which you probably refer to, would make the 6900XT faster than its successor (7900XT), which isn't faster than the M2 ultra either.
Two 6800 duos, that's 4 GPUs, right? They would represent 60 TFLOPs of single-precision compute power.Redshift 3.5.16 (macOS)
CPU: 24 threads, 0.00 GHz, 64.00 GB
GPU(s): [Apple M2 Ultra 48 GB 0.080ms]
Blocksize: 128
Time: 00h:05m:23s
changing the block size improves times a bit
Redshift 3.5.16 (macOS)
CPU: 24 threads, 0.00 GHz, 64.00 GB
GPU(s): [Apple M2 Ultra 48 GB 0.082ms]
Blocksize: 512
Time: 00h:04m:45s
My 7,1 with 2x6800 Duo will do the same test in 2:03s.
Two 6800 duos, that's 4 GPUs, right? They would represent 60 TFLOPs of single-precision compute power.
I'm not surprised they perform better than the M2 ultra (< 30 TFLOPS).
Looking at the Blender results on opendata.org, the M2 Ultra (76 cores) perform as well as the 7900XT (median score), and quite a bit better than the 6900 XT. Maybe Blender doesn't work well with AMD cards...
It's also about twice faster than the M1 Ultra (64 cores).
Except they don't function as multiple GPUs when running benchmarks. Feel free to look that up yourself.
Pro W6800X Quadruple MPXs versus other GPUs in Mac Pro 2019
real world speed test results for performance minded Macintosh usersbarefeats.com
I indeed have no idea what you're trying to prove, but I don't think it relates to my reading comprehension.I will repeat, since you lack reading comprehension:
I indeed have no idea what you're trying to prove, but I don't think it relates to my reading comprehension.
Your first link has no relevance to our discussion about redshift, as it shows how the card functions under boot camp.
The second videos shows two radeon 6800 duos performing much better than a single one under redshift. You think that redshift can take advantage of two separate cards but cannot use two GPUs on a single card?
Also look at the chart below. Is a single radeon pro W6800X GPU supposed to the faster than a radeon pro W6900X GPU?
The thing you said was that the duo cards "don't function as multiple GPUs when running benchmarks".Wow, you really are confused! Please quote exactly when and where I mentioned that a single W6800X is faster than a W6900X. I am waiting for the direct quote showing me saying something so stupid such as this... Come on.. Let's see it 🤦♂️