I’m also waiting to see!
Just got my new Apple MacMini G4
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 24 GB
GPU: Apple M4 16 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 07.31 fps 2X: 04.48 fps 4X: 01.65 fps
Iris 1X: 05.05 fps 2X: 02.84 fps 4X: 00.92 fps
Proteus 1X: 07.99 fps 2X: 05.20 fps 4X: 01.53 fps
Gaia 1X: 02.31 fps 2X: 01.71 fps 4X: 01.20 fps
Nyx 1X: 01.45 fps 2X: 01.09 fps
Nyx Fast 1X: 04.43 fps
Rhea 4X: 00.26 fps
4X Slowmo Apollo: 08.12 fps APFast: 09.58 fps Chronos: 02.45 fps CHFast: 03.96 fps
16X Slowmo Aion: 05.09 fps
1 Like
Thanks, Tobias.
Although I’ve always recommended setting memory at 10% for upscaling SD on M series Macs, I think it’s probably best to set it to 100%, or thereabouts, for HD sources. The reason for setting to 10% for SD was because it had been faster on previous M series as it made more use of the GPU. However, as the Neural Engine has become even faster, it could well be that the memory setting can be left set at 100% on the M4 for all resolution sources.
Perhaps you would like to experiment with the memory settings and source resolution and see the difference, if any, with the M4.
My M4 Pro Mac Mini is due to arrive in about an hour…
Thanks again.
Regards
Andy
Partial benchmark results for SD upscaling (can’t be bothered to wait for the models I never use) - 720x480 and 768x576
Max memory set to 10%:
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Pro 24 GB
GPU: Apple M4 Pro 16 GB
Processing Settings
device: 0 vram: 0.1 instances: 1
Input Resolution: 720x480
Benchmark Results
Artemis 1X: 43.99 fps 2X: 26.62 fps 4X: 09.64 fps
Iris 1X: 42.77 fps 2X: 18.30 fps 4X: 05.31 fps
Proteus 1X: 46.21 fps 2X: 28.74 fps 4X: 08.82 fps
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Pro 24 GB
GPU: Apple M4 Pro 16 GB
Processing Settings
device: 0 vram: 0.1 instances: 1
Input Resolution: 768x576
Benchmark Results
Artemis 1X: 42.68 fps 2X: 20.20 fps 4X: 07.31 fps
Iris 1X: 48.33 fps 2X: 12.31 fps 4X: 03.96 fps
Proteus 1X: 46.86 fps 2X: 20.68 fps 4X: 06.41 fps
Max memory set to 100%:
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Pro 24 GB
GPU: Apple M4 Pro 16 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 720x480
Benchmark Results
Artemis 1X: 39.97 fps 2X: 21.76 fps 4X: 05.76 fps
Iris 1X: 30.49 fps 2X: 16.54 fps 4X: 04.13 fps
Proteus 1X: 41.00 fps 2X: 28.46 fps 4X: 05.83 fps
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Pro 24 GB
GPU: Apple M4 Pro 16 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 768x576
Benchmark Results
Artemis 1X: 27.63 fps 2X: 17.06 fps 4X: 04.43 fps
Iris 1X: 20.15 fps 2X: 11.90 fps 4X: 02.58 fps
Proteus 1X: 27.80 fps 2X: 21.35 fps 4X: 04.40 fps
I should add that the speed for 720x480 will also apply up to 720x528. Anything above 528 will be at the slower 576 speed.
When the max memory is set to 100%, I can confirm that the GPU cores were not used, as seen in Activity Monitor. When set to 10%, the GPU cores were used, but not fully. Also, when using ASITOP, it shows that the Neural Engine IS still being used, but not fully, when the GPU is also used at the same time.
So, in a nutshell, despite in insanely fast Neural Engine in the M4 / M4 Pro, setting the memory to 10% to force the GPU into action in tandem with the Neural Engine still results in increased speed for SD upscaling. This is find surprising from the Geekbench AI benchmarks. It appears on there that the M4 Neural Engine was so fast, the GPU was almost redundant. Apparently not.
Andy
Hello.
All SYNTHETIC benchmarks are just for entertainment purposes ONLY (OC’ers bragging / taunting kiddie playground shenanigans) & they should NEVER… EVER… be appended to real-time performance; GeekBench, though, it’s the worst and GB has been caught plenty of times favouring Apple Silicon due to Apple & GB backroom deals. Just FYI.
Thank you, would you mind benchmarking under 1920x1080p? that is the benchmark matching your resolution on my M1 Pro, comparing with M4 Pro,
Topaz Video AI v5.3.6
System Information
OS: Mac v15.02
CPU: Apple M1 Pro 16 GB
GPU: Apple M1 Pro 10.667 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 768x576
Benchmark Results
Artemis 1X: 18.55 fps 2X: 10.81 fps 4X: 03.18 fps
Iris 1X: 15.51 fps 2X: 09.04 fps 4X: 01.74 fps
Proteus 1X: 16.83 fps 2X: 10.80 fps 4X: 02.99 fps
Gaia 1X: 05.98 fps 2X: 04.49 fps 4X: 03.24 fps
Nyx 1X: 02.51 fps 2X: 02.11 fps
Nyx Fast 1X: 04.57 fps
Rhea 4X: 01.37 fps
4X Slowmo Apollo: 20.07 fps APFast: 63.82 fps Chronos: 07.01 fps CHFast: 09.59 fps
16X Slowmo Aion: 22.66 fps
Thank you, wow, that’s barely only a bit better on Proteus model compared to my M1 Pro 14" base model and my old 9700k + 3080FE + 4x8GB DDR4-3200Mhz build, only slight 1-2fps more on Proteus x2 on both, and still more than half of my 4090 build on real performance; other than Proteus, they are still pretty low on other models. I will expect more performance increases on M4 Ultra on Mac Studio maybe in 2025.
1 Like
Is your M4 pro the 12 core CPU, 16 core GPU model or the 14 core CPU, 20 core GPU model?
Hi everyone,
I have just upgraded my pc. Here my Topaz benchmark with the new hardware. I have upgraded my cpu from AMD 5900x to Intel 14700K, my mainboard from asus Tuf x570 to asus Rog strix Z790 and my rams from 32GB 3000mhz cl16 to 32GB 7200mhz cl34. My GPU hasn’t changed and is 4080 Super.
The upgrade has brought in about %40 in fps. Honestly I was expecting more but this is not bad.
Topaz Video AI v5.3.6
System Information
OS: Windows v11.24
CPU: Intel(R) Core(TM) i7-14700K 31.747 GB
GPU: NVIDIA GeForce RTX 4080 SUPER 15.671 GB
Processing Settings
device: -2 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 30.56 fps 2X: 20.97 fps 4X: 06.10 fps
Iris 1X: 28.95 fps 2X: 19.27 fps 4X: 05.04 fps
Proteus 1X: 29.12 fps 2X: 20.11 fps 4X: 06.71 fps
Gaia 1X: 10.67 fps 2X: 07.35 fps 4X: 04.82 fps
Nyx 1X: 12.25 fps 2X: 10.48 fps
Nyx Fast 1X: 22.40 fps
Rhea 4X: 04.17 fps
4X Slowmo Apollo: 39.55 fps APFast: 87.93 fps Chronos: 23.09 fps CHFast: 34.41 fps
16X Slowmo Aion: 44.52 fps
If these are default clocks, not bad at all especially since Topaz in-house engine is not even close to FULLY utilizing any RTX (2000 series+) Tensor cores as they were designed for.
Thanks for asking - I was going to edit to add that detail but you got there first. It’s the 12 core CPU, 16 core GPU version. I wasn’t sure an extra £200 was worth it, given I can get a base spec model for £599. I may well do that at some point and use CLI scripts to use 2 in parallel.
Regards
Andy
Mac Mini with 12 Core CPU, 16 Core GPU:
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Pro 24 GB
GPU: Apple M4 Pro 16 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 09.58 fps 2X: 05.49 fps 4X: 01.91 fps
Iris 1X: 06.39 fps 2X: 03.51 fps 4X: 01.32 fps
Proteus 1X: 09.48 fps 2X: 06.63 fps 4X: 01.95 fps
Gaia 1X: 02.98 fps 2X: 02.18 fps 4X: 01.58 fps
Nyx 1X: 02.95 fps 2X: 02.82 fps
Nyx Fast 1X: 06.91 fps
Rhea 4X: 00.43 fps
4X Slowmo Apollo: 10.17 fps APFast: 17.58 fps Chronos: 03.36 fps CHFast: 05.45 fps
16X Slowmo Aion: 07.92 fps
4 Likes
Topaz Video AI v5.3.6
System Information
OS: Windows v11.23
CPU: Intel(R) Core(TM) i9-10900K CPU @ 3.70GHz 63.826 GB
GPU: NVIDIA GeForce RTX 3090 23.756 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 21.40 fps 2X: 11.47 fps 4X: 03.21 fps
Iris 1X: 20.56 fps 2X: 13.32 fps 4X: 03.41 fps
Proteus 1X: 20.80 fps 2X: 12.57 fps 4X: 03.50 fps
Gaia 1X: 07.49 fps 2X: 05.51 fps 4X: 03.02 fps
Nyx 1X: 09.79 fps 2X: 07.70 fps
Nyx Fast 1X: 18.18 fps
Rhea 4X: 02.99 fps
4X Slowmo Apollo: 27.14 fps APFast: 45.27 fps Chronos: 17.54 fps CHFast: 24.92 fps
16X Slowmo Aion: 23.25 fps
The 10 series is still impressive after so many generations later.
Super close benchmark to my 4090 build. Benchmark for 3080 below
Topaz Video AI v5.3.6
System Information
OS: Windows v11.23
CPU: Intel(R) Core(TM) i7-14700K 31.781 GB
GPU: NVIDIA GeForce RTX 4090 23.576 GB
GPU: Intel(R) UHD Graphics 770 0.125 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 42.14 fps 2X: 20.53 fps 4X: 05.29 fps
Iris 1X: 38.67 fps 2X: 26.14 fps 4X: 06.13 fps
Proteus 1X: 41.79 fps 2X: 23.40 fps 4X: 06.32 fps
Gaia 1X: 15.64 fps 2X: 10.85 fps 4X: 05.66 fps
Nyx 1X: 17.40 fps 2X: 14.67 fps
Nyx Fast 1X: 29.92 fps
Rhea 4X: 05.53 fps
4X Slowmo Apollo: 40.19 fps APFast: 81.41 fps Chronos: 29.89 fps CHFast: 35.12 fps
16X Slowmo Aion: 31.94 fps
Topaz Video AI v5.3.6
System Information
OS: Windows v11.24
CPU: Intel(R) Core(TM) i7-9700K CPU @ 3.60GHz 31.843 GB
GPU: NVIDIA GeForce RTX 3080 9.8174 GB
GPU: Intel(R) UHD Graphics 630 0.125 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 19.17 fps 2X: 11.11 fps 4X: 03.37 fps
Iris 1X: 18.96 fps 2X: 12.62 fps 4X: 03.26 fps
Proteus 1X: 18.69 fps 2X: 12.33 fps 4X: 03.35 fps
Gaia 1X: 06.54 fps 2X: 04.59 fps 4X: 02.93 fps
Nyx 1X: 08.36 fps 2X: 06.62 fps
Nyx Fast 1X: 15.62 fps
Rhea 4X: 02.60 fps
4X Slowmo Apollo: 22.05 fps APFast: 46.26 fps Chronos: 14.63 fps CHFast: 21.82 fps
16X Slowmo Aion: ... fps
Topaz Video AI v5.3.6
System Information
OS: Windows v11.23
CPU: AMD Ryzen 9 7950X3D 16-Core Processor 63.17 GB
GPU: NVIDIA GeForce RTX 4090 23.576 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 37.59 fps 2X: 17.33 fps 4X: 04.79 fps
Iris 1X: 35.76 fps 2X: 20.31 fps 4X: 05.45 fps
Proteus 1X: 37.07 fps 2X: 19.80 fps 4X: 05.66 fps
Gaia 1X: 14.91 fps 2X: 09.98 fps 4X: 04.64 fps
Nyx 1X: 16.58 fps 2X: 14.17 fps
Nyx Fast 1X: 29.10 fps
Rhea 4X: 04.79 fps
4X Slowmo Apollo: 39.10 fps APFast: 79.11 fps Chronos: 29.95 fps CHFast: 35.31 fps
16X Slowmo Aion: 30.67 fps
From the XEON8463 server I have rented, stronger single core and multi-core performance, as well as eight channel memory bandwidth, can indeed bring more advantages in super-resolution projects and increase the usage rate of 4090 by 100%
1 Like
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Max 128 GB
GPU: Apple M4 Max 96 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 14.44 fps 2X: 08.83 fps 4X: 02.88 fps
Iris 1X: 14.95 fps 2X: 05.22 fps 4X: 01.94 fps
Proteus 1X: 17.43 fps 2X: 09.78 fps 4X: 03.05 fps
Gaia 1X: 04.09 fps 2X: 03.03 fps 4X: 02.22 fps
Nyx 1X: 06.04 fps 2X: 05.85 fps
Nyx Fast 1X: 14.32 fps
Rhea 4X: 01.13 fps
4X Slowmo Apollo: 16.14 fps APFast: 36.75 fps Chronos: 05.39 fps CHFast: 08.81 fps
16X Slowmo Aion: 16.10 fps
7 Likes
Topaz Video AI v5.3.6
System Information
OS: Mac v15.01
CPU: Apple M4 Max 128 GB
GPU: Apple M4 Max 96 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 3840x2160
Benchmark Results
Artemis 1X: 03.18 fps 2X: 01.82 fps 4X: 00.66 fps
Iris 1X: 02.64 fps 2X: 01.32 fps 4X: 00.42 fps
Proteus 1X: 02.91 fps 2X: 01.99 fps 4X: 00.66 fps
Gaia 1X: 00.95 fps 2X: 00.66 fps 4X: 00.50 fps
Nyx 1X: 00.98 fps 2X: 00.93 fps
Nyx Fast 1X: 02.35 fps
Rhea 4X: 00.24 fps
4X Slowmo Apollo: 03.31 fps APFast: 08.98 fps Chronos: 01.33 fps CHFast: 01.99 fps
16X Slowmo Aion: 05.19 fps
Topaz Video AI v5.3.0
System Information
OS: Mac v14.03
CPU: Apple M3 Max 128 GB
GPU: Apple M3 Max 96 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 12.63 fps 2X: 07.74 fps 4X: 02.67 fps
Iris 1X: 08.20 fps 2X: 04.83 fps 4X: 01.68 fps
Proteus 1X: 12.28 fps 2X: 08.33 fps 4X: 02.37 fps
Gaia 1X: 03.29 fps 2X: 02.33 fps 4X: 01.69 fps
Nyx 1X: 05.39 fps 2X: 05.06 fps
Nyx Fast 1X: 12.34 fps
Rhea 4X: 00.82 fps
4X Slowmo Apollo: 11.04 fps APFast: 41.16 fps Chronos: 03.50 fps CHFast: 06.33 fps
16X Slowmo Aion: 13.43 fps
see post #2
Uh-huh!
So, the new Apple Silicon, M4 series, (as expected) is just an ant size difference in performance (margin of error… 
) compared to the M3 Max Silicon, eh?!
Who knew?! 
