Nvidia RTX 3080 Gaming Performance at 1440p: CPU or Architecture Bottleneck?


Posts: 242   +331
The only people I REALLY recommend choose Intel over AMD are the competitive, low res, high refresh gamers where Intel's single threaded advantage shows up. That also may be about to change, we will see how Zen 3 plays out, recommendations will adapt as new information emerges. THAT is how it's done.

Thanks, many of these 'bottleneck' benchmarks are at 1080p where the frame rates are well into the 100s of frames per seconds. I have an intel machine and honestly, I wish I would have gone with AMD because about the only thing that Intel is really better at these days is gaming. And with Zen 3, that looks like that won't be the case anymore. When I built my latest PC I didn't do enough research, I just knew that my last AMD machine was not great at gaming compared to the Intel chips that released at the time. I should have done more research into Ryzen, but I didn't and that was my fault. Even the fact that I have to upgrade my motherboard to upgrade to 10th gen is annoying. AMD typically allows more generations to use the same motherboards. I believe the AM4 series motherboards will work with Zen 3, but I'm not 100% sure about that.

But you are correct, the difference between equivalent chips is typically only realize at 1080p, or maybe 1440p, but the framerates are usually over 100 fps already when you see these differences. The more you push your GPU, the less of a problem it usually is. The scary "bottleneck" is really meaningless unless you just have a really old processor or you bought something that doesn't have enough cores for gaming. A 10% bottleneck at 1080p is probably half that at 1440p, maybe even non-existent.


Posts: 155   +121
My 3440x1440p, 120Hz monitor begs to differ.

As do the Samsung G7 240hz 1440p monitors. Clearly Sir Alex Ice doesn't realize calling people dumb while asking a question fits the definition of irony quite well. 1440p @ high refresh rates is just fine for 3000 series cards and he clearly doesn't own one to talk of where they should or shouldn't be used.


Posts: 155   +121
It’d be interesting to see results but I think you’re probably right

It's true, swaggman nailed it exactly. At 1080p, CPU differences stand out noticeably with higher end GPUs, while at 1440p, those differences disappear to almost nothing, but still measurable slightly. At 4K those differences are gone and the GPU is the PRIMARY determiner of performance.

Every Core i7 from first generation to latest 10th generation tested at 1080p and 1440p using a RTX 3080.

Resolution is the key to making CPU difference disappear. The higher the res, the lower the CPU relevance.

Edit - For the record, I have owned 3 in this list. i7 920, i7 2600K, and i7 7700K. After seeing this and currently gaming on a 1440p@165hz monitor, I'm more interested in upgrading to 4K@144hz than I am worried in any way about my 7700K being paired with a 3090. At 1440p, these tests show a tiny delta of performance between 7700K and 10700K. Almost nill, to be exact.

To take it a step further, if you watch the 1440p portion of these tests, and extrapolate the 10% improvement of 3090 over 3080, then it begins looking like a 1440p@165hz monitor actually isn't so badly paired either as most of the time, newer games still push these GPUs below 165fps when things like ray tracing and AA are stacked on top of each other. It's actually a perfect pairing in efficiency in some ways...

Last edited:


Posts: 432   +210
Ampere seems to scale better with more gpu bound scenes that is why the 3090 is pared best at 4k and worst scaling at 1080p.
I went from a 3440 by1440p ips monitor to the cx 4k oled last summer at the time I had the 2080ti xc2 hybrid which scaled well at 1440p ultrawide but I knew that Ampere successor would be overkill for that monitor. Currently at 4k 120hz I believe I have another 1 to 2 generations of upgrading before upgrading my upgrading my 9900ks cpu.
Side note did you know that ddr5 ram kits being scalped 5x to 6x its MSRP currently guess people didn't see the benchmarks for the best ddr4 kits with extremely low timings but those are also in high demand usually but 6x! Craziness.