They had to use Cinebench R23 for this. Cinebench R23 is hand optimized for AVX and poorly translated to use NEON.I definitely don’t think the 4.3GHz ST will look great. And we know the MT clocks are at best 3.8GHz fwiw. Tops out at 80W platform power there so I mean in practice for ST I could see the power draw at 9-12W or something at 3.8GHz. Which isn’t too bad, the perf on GB6 is around 2770 or so at that point? I think AMD still draws more for similar.
Apple M2 Review
The new Apple M2 is being touted as a much faster SoC than its competitors, but we'll have to see that for ourselves as we test it...www.techspot.com
Here’s a review that took power from the wall minus idle (so display mostly and statics removed). M2 ST is about 8W.
You also notice MT power — and we saw this in the Mac Mini Andrei review too — is a bit higher than Apple’s claims probably because they do roughly Intel/AMD on their power figures when they say the M2 is 15W at top (or M3, 17W) — they are only referring to the CPU or SoC.
Also see the ST is a bit higher from even the AMD and Intel stuff. I’m not 100% sure what Phoenix looks like but I’d wager it peaks as badly as Rembrandt just with more performance.
I am aware of that, but it’s not going to affect the wattage itself substantially and you see the same totals elsewhere. I am only interested in the wattage here, because it will have overlap with ST workload dynamic power elsewhere where the code is fair, and this is what was available. Even Andrei’s M1 Mini Test ended up around 7W dynamic power ST. Point is, nearly everyone underestimates what their chips in any meaningful sense are consuming, and this is abetted by AMD/Intel and to a lesser extent Apple given how they qualify consumption.
For what's it's worth:
www.cgdirector.com
www.cgdirector.com
Yep. It would harm energy efficiency by virtue of being less performant for our outcome, but the *power* figures themselves are unlikely to change much, and I posted them more as proxies for what these chips look like under various workloads when stressed.For what's it's worth:
Cinebench R23 ST score: M2 is 28% below 14900K
Cinebench R24 ST score: M2 is 14% below 14900K.
For M1 it's -35% for R23 and -20% R24.
It's hard to say how that'd translate to power usage and efficiency, but I guess power would be very similar given that, even though translation is poor, NEON is used in R23.
Refs:
Cinebench R23 Scores [Updated Results]
Cinebench R23 is the newest instalment of the most popular CPU-Rendering Benchmark Cinebench. We have the Score Results for all modern Processors.Cinebench 2024 Scores [CPU & GPU] (Updated Results)
The new Version of Cinebench, 2024, benchmarks CPUs and GPUs in the popular Redshift render engine. We benchmarked all modern CPUs and GPUs and list them in our score overview.
Frequencies, power, and IPC are mostly disconnected. You can have a 6 ghz CPU that sips power and a 1 Ghz CPU that drinks it, both made on the same process.
NVIDIA and AMD disagree: GPUs are what are powering AI right now, mind you. They are massively parallel by design, low frequency, powerful, efficient for what they do, and turing complete.
And just like that it's a reminder of the real world. Plenty of AVX optimized open source software without NEON ports yet.
To be fair, Oryon only has a rough timeline of late Q2 / early Q3 ("mid-year 2024"). Arm v Qualcomm's trial won't start until September 2024, so perhaps just after Qualcomm's intended timeline.
The lawsuit is still heavily ongoing; you can follow along here:
Arm Ltd. v. Qualcomm Inc., 1:22-cv-01146 - CourtListener.com
Docket for Arm Ltd. v. Qualcomm Inc., 1:22-cv-01146 — Brought to you by Free Law Project, a non-profit dedicated to creating high quality open legal information.www.courtlistener.com
Interesting tidbit #1: AMD, Apple, Ampere, MediaTek, TSMC, NVIDIA, Cadence, Google, Synopsys, Intel, Cadence etc. are all involved in the trial now. Lots of people giving depositions / receiving subpoenas to testify in Court.
Interesting tidbit #2: So far with discovery & depositions, the judge is mostly siding with Arm and against Qualcomm.
The last update is October 25, so literally yesterday haha.
- ALAs from Arm: Judge says Qualcomm's motion is partly granted, partly denied. Can't see the details.
- Qualcomm tried to get a deposition from Masayoshi Son. Judge rules against Qualcomm here.
- Qualcomm wanted discovery of Arm's IPO. Judge rules against Qualcomm here.
- Qualcomm wanted docs from Antonio Viana at Arm. Judge rules against Qualcomm here.
- Qualcomm wanted Apple's & Ampere's specific ALAs. Judge rules against Qualcomm here.
That's significant, I think PHP is still used a lot. I wonder what the impact of AVX use is, I couldn't find any benchmark. The JIT supports AArch64, but doesn't emit NEON instructions it seems.
DX12 is a completely different beast from DX10/11 or DX7/8/9.
Nuh uh. If it were that simple, Intel's driver team wouldn't have had gone through so much trouble with ARC, trying to support older DirectX AAA titles. QC isn't targeting gaming so most likely, older DX games will not run or suffer from glitches that no one other than the community will have to fix.
So far QC's D3D11 driver sucks balls.
By then QC will likely have a Vulkan driver running on WoA with DXVK providing all the magic, leaving only x86 binary translation as the significant stumbling block.
Wow DXVK sounds amazing, converting DX9,10,11 to Vulkan.
They are in the process of adding D8VK to it soon also for DX8 support.
Not sure - it's possible that they will just leave that to devs to do much like some have been doing with DXVK-Native which linked it directly to the game code so that it produces Vulkan output without the OS/drivers coming into it.
Facebook
Twitter