Factory Overclocked 2080 TIs?
asdf123
Posts: 167
in The Commons
Hi,
So I'm using dual ASUS Strix 2080 TIs in my 3900x build. The primary reason I'm using Strix is that, even with factory OC, they are the temp king of the 2080bTI lineup.
That said, I know Daz doesn't really benefit from OC GPUs. Ignoring the fact that I paid more for the OC, would there be a downside to underclocking the Strix back to 2080 TI base speeds?
My reasons for doing so would be... Even lower temps for the same performance as standard 2080 TIs, potentially more stability, potentially longer life for GPUs...
Thanks!
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2026 Daz Productions Inc. All Rights Reserved.
Comments
If you check the power usage when rendering iray, it is much lower then gaming or 3D benchmarking, even though your clock is still pinned.
From what I understand the lifespam of a processor (CPU or GPU) is primarily reduced by temperature and voltage, not frequency. If you are seriously interested in prolonging the life of your GPU, you might also consider undervolting or just improving air flow in your case.
I wouldn't bother though, by the time your GPU dies, it'll be time to upgrade anyway. Unless it dies soon, in which case you can claim on warranty, this is almost a reason to try and burn your card as much as possible.
You can always change your mind and go back to the previous settings. However clock speeds do impact render times, it is just a matter of how much the overclock is. The best way to find out is to test it. It wont hurt to downclock as like I said you can just revert it. You can test a scene with the base clock VS the factory clock and see how much it effects the render speed, as well as the temps.
Higher overclocks will make rendering faster, yes. It might not be a massive improvement, which is why some people don't bother. But it absolutely will be faster. Personally I don't suggest overclocking beyond the factory clock too much unless you know what you are doing. But this case, you are talking about downclocking from an existing factory clock, which is very easy to do.
Basically what I am saying is that if your cards are clocked higher than others, you should be a little faster than everybody else sporting 2080tis.
You can also try out the benchmark scenes made by some of us, and compare them directly to others. There are several people with two 2080tis.
Underclocking is undervolting. Doing so _should_ extend the life of the component.
This is true and I just checked, there is no option to undervolt anyway, at least not in my utility.
---
One last thing, you might be surprised how fast your GPU clocks with it's own boost because of the low power consumption when rendering Iray. I have a Gigabyte AORUS waterforce RTX2080ti (it has it's own AIO water cooler) which was rated at 1770 MHz, but automatically boosts to 2050 MHz when rendering Iray due to low power usage. The power target is set to a maximum of 110%, but only hits about 65% power usage when rendering, GPU only reaches 45 degrees celcius. I can get the GPU to 52 degrees celcius and 110% power usage with MSI Kombuster, the GPU clockspeed hits 1770 MHz under those conditions.
I have never bothered to find out what the most significant bottleneck is when rendering; GPU clockspeed, GPU memory speed or CPU clockspeed. My CPU is ancient (it's a 5930k from 2014) and appears to max out my GPU without issue, so I don't think CPU is significant.
The bottleneck changes during the render. At the very start of the render it is PCIE bandwidth/CPU power as the textures, geometries etc. must be loaded from disk into system RAM, which involves compression of textures based on your settings, and then transferred to VRAM.
After the render starts all that is really getting used are the streaming multiprocessors, the part of the GPU that handles CUDA. So they become the "bottleneck."