GPU Benchmark! 4070 4080 4090 3080 3090 #2970
Replies: 39 comments 12 replies
-
|
2070 rtx mobile |
Beta Was this translation helpful? Give feedback.
-
|
Nvidia RTX 4090 FE 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:03<00:00, 6.17it/s] (edit: used default resolution first go around) |
Beta Was this translation helpful? Give feedback.
-
|
CPU vs GPU 1024x1024 2 steps Intel 13900KS 24,25 sec. https://files.catbox.moe/f4wb53.mp4 CPU.vs.GPU.test_2.mp4 |
Beta Was this translation helpful? Give feedback.
-
|
Not quite a benchmark, you Gorilla. |
Beta Was this translation helpful? Give feedback.
-
|
RTX 3090 |
Beta Was this translation helpful? Give feedback.
-
|
Just for laughs ... |
Beta Was this translation helpful? Give feedback.
-
|
My 1070 is my favorite, was my longtime friend still in good use, pretty good for what it was, 3060 is 2 faster with only half as many additional cores, but with the 4090 I have now, uncomparable. |
Beta Was this translation helpful? Give feedback.
-
|
RTX 3070 OC Drivers:~528.xx 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:08<00:00, 2.26it/s] |
Beta Was this translation helpful? Give feedback.
-
|
My poor old GPU 😢 |
Beta Was this translation helpful? Give feedback.
-
|
RTX 4080 |
Beta Was this translation helpful? Give feedback.
-
|
Would love to get some 3060 / 4060 data to better complete the regression... Then I'll post proposed for every 3000/4000 card.
|
Beta Was this translation helpful? Give feedback.
-
|
RTX 3090 driver:551.61 RTX 3090 driver:531.18 |
Beta Was this translation helpful? Give feedback.
-
|
RTX 4060ti 16GB |
Beta Was this translation helpful? Give feedback.
-
|
RTX 4070 12Gb got prompt |
Beta Was this translation helpful? Give feedback.
-
|
WSL2 (6.1.x kernel) 3060Ti Tesla P40 No xformers 3060ti Tesla P40 |
Beta Was this translation helpful? Give feedback.
-
|
4090 FE, Ryzen 7900X, Samsung 980 2TB M.2 got prompt
|
Beta Was this translation helpful? Give feedback.
-
|
RTX 3070 Laptop GPU
|
Beta Was this translation helpful? Give feedback.
-
|
Windows 10 I ran the test 3 times. |
Beta Was this translation helpful? Give feedback.
-
|
RTX4090 Power Limit 300W, 12600K, DDR4 3200 64GB RTX3090 Power Limit 300W, 5700X, DDR4 3200 64GB RTX3060 Power Limit 130W, 5700X3D, DDR4 3200 128GB |
Beta Was this translation helpful? Give feedback.
-
|
Wow you beat my 3.08 s. Nice! Win 11, 10 or Linux? May I ask, why do you need the 300W power limit and what version of the 4090 do you have? I mean I have Founder Edition with Ryzen 7900X and SF750 in a very small sffpc 10 liters Formd T1 case, air cooled, but I only have the CPU limited and the 4090 remains with stock settings. It never overheats and honestly generating in comfyui for hours on auto queue doesn't give the same load as benchamrk, I mean comfyui doesn't give a constant 100% load like a synthetic tests or other rendering, or gaming at high resolutions like 5120. |
Beta Was this translation helpful? Give feedback.
-
|
3070 laptop |
Beta Was this translation helpful? Give feedback.
-
|
Gigabyte RTX 4080 / i5-13600KF / 32GB RAM |
Beta Was this translation helpful? Give feedback.
-
|
4070 ti super, ram 64gb, Core i7 12700f, Windows 11, python 3.11.9
|
Beta Was this translation helpful? Give feedback.
-
|
I Believe this to be an accurate scrape of this thread until this comment #2970 (comment). I assumed non mobile GPU/CPU were desktop systems. Mines the very last few rows (RIP, lol). Hope this is helpful!
|
Beta Was this translation helpful? Give feedback.
-
|
Great Job !!! |
Beta Was this translation helpful? Give feedback.
-
|
removed outliers and reduced down to base model (combined Ti, SUPER, etc together) |
Beta Was this translation helpful? Give feedback.
-
|
Rtx 4090 16 GB mobile Dell |
Beta Was this translation helpful? Give feedback.
-
|
This is where the pain and sadness 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [02:28<00:00, 7.45s/it] |
Beta Was this translation helpful? Give feedback.
-
|
MSI GeForce RTX 3080 VENTUS 3X 10G / Ryzen 7 7800X3D / 32GB RAM 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:06<00:00, 3.00it/s] |
Beta Was this translation helpful? Give feedback.
-
|
Sapphire Nitro+ Radeon RX 7900 XTX Vapor-X with AMD Go Fast extension and some other tweaks, Linux 6.11.5, ROCM 6.2.0: |
Beta Was this translation helpful? Give feedback.
-
|
Update: I have it at 5.3it/s now with some very conservative undervolting and a few more tweaks. |
Beta Was this translation helpful? Give feedback.
-
|
T_T 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [04:37<00:00, 13.88s/it] |
Beta Was this translation helpful? Give feedback.
-
|
Hello all 😊, finally signed after a few months of following the discussions, now i can annoy everyone with questions lol results from my Acer TC1760 , i5-12400, 32gb DDR4 3200 ram, 1tb WD black sn770 nvme, and no laughing .. RTX 3050 6gb. Total VRAM 6144 MB, total RAM 32612 MB 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:25<00:00, 1.29s/it] |
Beta Was this translation helpful? Give feedback.
-
|
RTX 3090, NVIDIA-SMI 550.120, Driver Version: 550.120, CUDA Version: 12.4 The Model is already loaded. First run after cooling down the RTX to 33°C is always slower than the second run. After that it is getting slower until around 73°C. Temp: 33°C, with "Save Image" Prompt executed in 5.37 seconds Temp: 33°C, with "Preview Image" Prompt executed in 5.28 seconds Temp: 73°C, with "Save Image" Prompt executed in 5.46 seconds Temp: 73°C, with "Preview Image" Prompt executed in 5.39 seconds |
Beta Was this translation helpful? Give feedback.
-
|
We recently re-ran our ComfyUI-based flows and thought the results might be of interest to the local community. 🛠️📊 Here's just a screenshot:
If you're interested in other graphs, follow the link: Hardware Results Also one page down ( |
Beta Was this translation helpful? Give feedback.
-
|
Are you running with AMD Go Fast? https://github.com/Beinsezii/comfyui-amd-go-fast It's basically a free 2x. |
Beta Was this translation helpful? Give feedback.
-
|
No, I didn't even know about such a thing. Thanks for notifying me that it exists, I need to check it. In 3-4 weeks there will be new benchmarks, much more clearly grouped, with tests of token encoders when the prompt changes, and I will add some video model. |
Beta Was this translation helpful? Give feedback.
-
|
win11 wsl2 ubuntu 100%|███████████████████████████████████████████████████████████████████████████████████| 20/20 [00:04<00:00, 4.03it/s] |
Beta Was this translation helpful? Give feedback.
and others
-
Could we Do a benchmark for GPU´s ?
Need to get new hardware .
To keep it simple we could use the default load worklow on 1024*1024 with sdxl 1.0.
Seed 1
I know that ram comes in play when workflows get more complex , but for the start simple ist best.
Please post the second run. first run ist a little bit corupted by loading data.
theory pudget:
toms hardware 1111:
Beta Was this translation helpful? Give feedback.
All reactions