Benchmark results v20000

[HugoIII]

Hey folks,

first I want to thank Serif for the great v2 upgrade - for me this is a real game changer (especially the publisher IPad App and the new Ipad optimized UI - I total love it and it changes my workflow completely in a very positive way. 
And to optimize it further yesterday the delivery guy brings me a nice new tool for my work. I want to share the benchmark results here if anyone else is considering to get a new IPad.

Now I am using a 12,9“ M2 Ipad 16GB Ram (it is my companion of a 15“ 2016mbp which is signifikant slower… in the affinity benchmarks)

I hope this is good Place for the results because @MikeTO wishes a new thread (he is the guy who brought all the results together in the past) and I think the beta photo forum part is not the right place for such a threat - If anybody else has a good Idea maybe the mods could consider to change the place.

And if you have other results to share with us this could be a very useful thread to get an overview wich System runs how well with the affinity suite.

Here is my result:

[MikeTO]

Apple MacBook Pro 14" M1 Pro (10-core CPU, 14-core GPU) with Ventura 13.0.1

[Dr. Joachim Neudert]

Apple Mac Studio M1 Ultra, 64 GB, 48 GPU Cores:

 

 

 

Edited December 3, 2022 by Dr. Joachim Neudert
Double graphics corrected

[Dr. Joachim Neudert]

Windows 10 AMD Ryzen 5950 64 GB RAM AMD RX5700 XT Graphics card

 

[StudioJason]

iPad Pro v.3 12.9”, 256G

 

[Tom61]

Intel i7-12700K, 64GB RAM, NVIDIA GeForce RTX 3060 Ti / Windows 11

Edited December 5, 2022 by Tom61

[iuli]

M1 iPad Pro 11” 256GB iPadOS 16.1.1

 

 

Out of curiosity I ran the benchmark for AP1 for iPad as well. The performances are about the same for all entries with the exception of Vector, Multi CPU which’ value is 500 higher in V2’s favour.

 

[Duskstalker]

AMD Ryzen 5900X, 2x16gb DDR4 3600 CL18, Intel Arc A770+A750 Multi GPU, Win10
For some reason the A750 is the main GPU in this dual GPU setup, even though the A770 is in the first pcie slot and is set as primary gpu in windows and in affinity. the a770 alone scores about 9000.

[pruus]

iPad mini 6

[tatanka]

iMac18,3 – Retina 5k, 27-inch, 2017. Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz

 

[baba_yu]

Intel® Core™ i7-13700K+RTX 3080

[Tom61]

Intel i7-12700K, 64GB RAM, NVIDIA GeForce RTX 3060 Ti / Windows 11 now with Multi-GPU


 

[Amber_Mtl]

iPad Pro 5th 2021 12.9" 256 GB/ IOS 16.1.1

 

[McD]

It’s a shame there isn’t better hardware identification built into these products generally (not just for benchmarks). I would have thought this would assist with fault reporting too.

 

[Duskstalker]

with the new 2.0.3 update the benchmark got changed. the scores for my A770 improved from 9000 to 13000, roughly. the cpu score decreased a little bit. can someone else retest their sytem to confirm this?

[Tom61]

Benchmark-Version 20300
Intel i7-12700K, 64GB RAM, NVIDIA GeForce RTX 3060 Ti / Windows 11 with Multi-GPU

[Erikyo]

I wanted to show the result of the i5 13500 and an old nvidia 960 which I must say make a good impression considering that the processor now costs 300 euros (the launch price) and the video card can be found for 100 euros from ebay. I have to say that the affinity team is putting a lot of effort into optimizing the code for metal/apple devices but they are not putting the same effort for openCL/other systems, because the gap is all here. In fact, if you check other benchmarks like 3dmark or cinebench the M1 pro gets similar scores to an Nvidia 1060 (which is an average gpu nowadays) but in the comparisons above I see a whole different situation...

[Dr. Joachim Neudert]

With the new program version: Combined Score 24 657 Points. Mac Studio Ultra, 64 GB, 48 GPU Cores:

[McD]

On 1/15/2023 at 10:00 AM, Erikyo said:

In fact, if you check other benchmarks like 3dmark or cinebench the M1 pro gets similar scores to an Nvidia 1060 (which is an average gpu nowadays) but in the comparisons above I see a whole different situation...

Cinebench is a CPU test, nothing to do with your GTX960. On 3D Mark (Wildlife), here’s an M1 Max beating out a 3080ti 

 

[Erikyo]

2 hours ago, McD said:

Cinebench is a CPU test, nothing to do with your GTX960. On 3D Mark (Wildlife), here’s an M1 Max beating out a 3080ti 

 

Please check before that guy channel... he's an apple fanboy, what do you expect from him? I'm talking about benchmarks with comparable results from everyone. Besides mine is a desktop not a notebook (which has other features including low power consumption) and I would like affinity apps to be optimised for these as well. I think it is a legitimate request.

Don't believe the hype - https://www.tomshardware.com/news/apple-m2-gpu-analysis

Edited February 4, 2023 by Erikyo

[McD]

6 hours ago, Erikyo said:

Please check before that guy channel... he's an apple fanboy, what do you expect from him? I'm talking about benchmarks with comparable results from everyone. Besides mine is a desktop not a notebook (which has other features including low power consumption) and I would like affinity apps to be optimised for these as well. I think it is a legitimate request.

Don't believe the hype - https://www.tomshardware.com/news/apple-m2-gpu-analysis

And Tom's hardware is a PC fanboy site with literally nothing good to say about Apple despite the figures. Look at how the M2 article focusses purely on hypothetical performance by compute (TFLOPs) when 2D & 3D rasterisation requires so much more. The M1 Ultra (64) outperforms the RTX3090 in rasterisation by ~10% with only half the compute (20 vs 40 TFLOPs).

Ad hominem zealotry aside, Cinebench is not a GPU benchmark and 3D Mark GPU benchmark you suggested doesn't back your argument.

3D Mark Wild Life Extreme on M1 Pro: 10352

3D Mark Wild Life Extreme on GTX1060: 8583 average

If there's another optimal benchmark we should be looking at by all means suggest it.

[Dr. Joachim Neudert]

Well, in fact I do own a powerful WIntel PC as well: AMD 16-Core 5950 CPU 5 GHz,with  64 GigaByte RAM and an AMD Radeon 5700 XT with 8 GB Video-RAM. But concerning the real-world Benchmark of Affinity Photo 2.0, it simply lies in the dust against my Mac Studio M1 Ultra. I have posted its benchmarks already some time ago, above in this thread. 

Raster (Einzel-GPU)

1092 - AMD Radeon RX 5700 XT

Raster (Einzel-GPU):
41291 - Apple M1 Ultra (Metal)

 

combined: 

Kombiniert (Eine GPU):
23868 - Apple M1 Ultra (Metal)

Kombiniert (Eine GPU)
1205 - AMD Radeon RX 5700 XT

 

The Mac Studio M1 Ultra with 48 GPU-Cores is faster by a factor of 20 to 40 to the AMD GPU (AMD graphics chips offer no hardware acceleration for Affinity photo), and factor 2 to 4 than any of the Intel-Nvidia-Benchmarks result that were posted here. A single GPU score of 41291 points has still to be beaten, a combined score of 23868 as well.

Best regards

Joachim

[v_kyr]

7 hours ago, McD said:

If there's another optimal benchmark we should be looking at by all means suggest it.

Well most of those benchmarks are just of synthetical nature (CPU/GPU limited) and thus don't tell much about a wholes system behavior under more daily application working and/or stress conditions. Thus real application benchmarks are in contrast here often overall much more meaningful than synthetical ones.

For application benchmarks the BAPCO suites are useful, but not all of them are crossplatform usable here. In order to check against the PC & Mac platforms here a cross benchmark (crossmark) would have to be used then.

[Erikyo]

9 hours ago, McD said:

Ad hominem zealotry aside, Cinebench is not a GPU benchmark and 3D Mark GPU benchmark you suggested doesn't back your argument.

3D Mark Wild Life Extreme on M1 Pro: 10352

3D Mark Wild Life Extreme on GTX1060: 8583 average

If there's another optimal benchmark we should be looking at by all means suggest it.

@McD Please note that only the latest version (R20 - R23 released after 2020) does not benchmark the GPU, but before that the R15 version did. Go look at the results and tell me if M1 has the same result of 1080 or something like that (and not the notebook version please).

Also what you say is right, the M1 pro is slightly superior to a (non ti) 1060, which is true, but the 1060 is an mid-low video card. Whereas in the benchmarks above, there is a very pronounced disparity as if Affinity was very optimised for apple M cpus and not so much for the others processors. Don't you think? 

[Twolane]

Huh. I guess my 12-year-old Dell running v1 doesn't rate.  (Not really surprised.) I haven't tried the v2 product on it - yet - but I might just to see if it will run and work adequately. You know, just to write posts in this thread. 😁