How much of a difference is my set up and this guys setup?

[noahthedestroyer5]

So I have a 2080 super and Ryzen 3700x. Im concerned because i saw this video of a guy with a 2080 Ti and Ryzen 3900x: Shouldnt my FPS be around his, since theres not much different between his and my equipment or am I wrong? My framerate seems to be kind of a lot lower because mine is usually in the 200s or 300s while his is consistently in 300s 99% of the time.

edit: i play on max settings 1440p like this video

 

[LIGISTX]

1 minute ago, noahthedestroyer5 said:

So I have a 2080 super and Ryzen 3700x. Im concerned because i saw this video of a guy with a 2080 Ti and Ryzen 3900x: Shouldnt my FPS be around his since theres not much different between his and my equipment or am I wrong? My framerate seems to be kind of a lot lower because mine is usually in the 200s or 300s while his is consistently in 300s 99% of the time.

 

CS is a horrible way to test things.... since it’s always running such high FPS. At that point, your mostly CPU bottlenecked. His 3900x is clocked a bit higher, and his GPU is notably faster. But.... being in the 200’s, your clearly fine. I would run a more normal game or benchmark to actually see how it does, not one where your up against the engines limits.

 

Also, remember, a 2080 ti is flat out faster than a 2080 super. Not by a huge margin, but it is noticeably faster. 

[zeusthemoose]

That sounds about right. The 2080ti is considerably faster than the 2080 super.

Edited August 25, 2020 by zeusthemoose

[Jurrunio]

He's running 3600MHz CL14 memory (i.e. low latency and high bandwidth), is yours that good?

[noahthedestroyer5]

1 minute ago, LIGISTX said:

CS is a horrible way to test things.... since it’s always running such high FPS. At that point, your mostly CPU bottlenecked. His 3900x is clocked a bit higher, and his GPU is notably faster. But.... being in the 200’s, your clearly fine. I would run a more normal game or benchmark to actually see how it does, not one where your up against the engines limits.

 

Also, remember, a 2080 ti is flat out faster than a 2080 super. Not by a huge margin, but it is noticeably faster. 

ah ok thank you. I appreciate the response

[noahthedestroyer5]

Just now, Jurrunio said:

He's running 3600MHz CL14 memory (i.e. low latency and high bandwidth), is yours that good?

Mine is Cl 16 dual channel 16 Gb 3600Mhz

[Jurrunio]

Just now, zeusthemoose said:

That sounds about right. The 2080ti is considerably faster than the 2080 super.

1 minute ago, LIGISTX said:

CS is a horrible way to test things.... since it’s always running such high FPS. At that point, your mostly CPU bottlenecked. His 3900x is clocked a bit higher, and his GPU is notably faster. But.... being in the 200’s, your clearly fine. I would run a more normal game or benchmark to actually see how it does, not one where your up against the engines limits.

 

Also, remember, a 2080 ti is flat out faster than a 2080 super. Not by a huge margin, but it is noticeably faster. 

With the 2080Ti usage around 30% in the section of game I've seen in the video, the GPU really doesnt matter here

[noahthedestroyer5]

1 minute ago, zeusthemoose said:

That sounds about right. The 2080ti is considerably faster than the 2080 super.

ok thank you

[Jurrunio]

1 minute ago, noahthedestroyer5 said:

Mine is Cl 16 dual channel 16 Gb 3600Mhz

If he tuned up all the subtimings plus proper CPU tuning then it's possible for it to pull out that lead

[Mateyyy]

5 minutes ago, noahthedestroyer5 said:

I have a 2080 super and Ryzen 3700x

5 minutes ago, noahthedestroyer5 said:

i saw this video of a guy with a 2080 Ti and Ryzen 3900x

He also seems to be running 3600 CL14 memory. Faster hardware should obviously result in better performance, even in a game like CS:GO which is a terrible benchmark.

Look up LTT's video of the 3700X and 3900X and you'll see that there's a substantial difference between them in CS:GO.

[Damien_X]

Based on these:

https://gpu.userbenchmark.com/Compare/Nvidia-RTX-2080S-Super-vs-Nvidia-RTX-2080-Ti/4050vs4027

https://cpu.userbenchmark.com/Compare/AMD-Ryzen-9-3900X-vs-AMD-Ryzen-7-3700X/4044vs4043

The difference shouldnt be big, so maybe there are other parameters like ram, overclocking on his side, settings in the game etc.

[zeusthemoose]

1 minute ago, Jurrunio said:

With the 2080Ti usage around 30% in the section of game I've seen in the video, the GPU really doesnt matter here

You sure? I was skipping around and the gpu was mostly 70-80% usage while the cpu was 13-16%.

[LIGISTX]

1 minute ago, Jurrunio said:

With the 2080Ti usage around 30% in the section of game I've seen in the video, the GPU really doesnt matter here

It is mostly CPU bound for sure. But still, even at lower % utilization it will likely pump out more FPS. But it’s mostly CPU bound...

 

To get “more” FPS here , high end intel is the only way. Not that I’m telling OP to change his platform up, just saying if you want blazing 300+ FPS... AMD is not the answer. 

[noahthedestroyer5]

1 minute ago, Mateyyy said:

He also seems to be running 3600 CL14 memory. Faster hardware should obviously result in better performance, even in a game like CS:GO which is a terrible benchmark.

Look up LTT's video of the 3700X and 3900X and you'll see that there's a substantial difference between them in CS:GO.

ah ok. Then its probably the memory or he has a faster gpu/cpu

[LIGISTX]

1 minute ago, Damien_X said:

Based on these:

https://gpu.userbenchmark.com/Compare/Nvidia-RTX-2080S-Super-vs-Nvidia-RTX-2080-Ti/4050vs4027

https://cpu.userbenchmark.com/Compare/AMD-Ryzen-9-3900X-vs-AMD-Ryzen-7-3700X/4044vs4043

The difference shouldnt be big, so maybe there are other parameters like ram, overclocking on his side, settings in the game etc.

Userbenchmark is terrible... don’t bother using that. 

 

2 minutes ago, zeusthemoose said:

You sure? I was skipping around and the gpu was mostly 70-80% usage while the cpu was 13-16%.

It’s still CPU bound. Just because it’s a low CPU % doesn’t mean it’s not a CPU bottleneck. Ryzen isn’t fast enough to feed the GPU with enough data per time, this is why for “very high FPS gaming” a 5+ghz 10900k for example will be “a lot better”. Is ~280 vs ~320-340 FPS (using educated guesses for FPS here) really matter? Probably not.... but if it does to you, Intel is the answer. For most people, that isn’t actually a good value proposition tho. 

[noahthedestroyer5]

4 minutes ago, Damien_X said:

Based on these:

https://gpu.userbenchmark.com/Compare/Nvidia-RTX-2080S-Super-vs-Nvidia-RTX-2080-Ti/4050vs4027

https://cpu.userbenchmark.com/Compare/AMD-Ryzen-9-3900X-vs-AMD-Ryzen-7-3700X/4044vs4043

The difference shouldnt be big, so maybe there are other parameters like ram, overclocking on his side, settings in the game etc.

i dont really trust user benchmark after what ive heard their bad practices

[Jurrunio]

5 minutes ago, zeusthemoose said:

You sure? I was skipping around and the gpu was mostly 70-80% usage while the cpu was 13-16%.

Ok it's 50-60% for the first two thirds of the gameplay (30-40% when killed), 70-80% at the last third of gameplay.

[noahthedestroyer5]

1 minute ago, LIGISTX said:

Userbenchmark is terrible... don’t bother using that. 

 

It’s still CPU bound. Just because it’s a low CPU % doesn’t mean it’s not a CPU bottleneck. Ryzen isn’t fast enough to feed the GPU with enough data per time, this is why for “very high FPS gaming” a 5+ghz 10900k for example will be “a lot better”. Is ~280 vs ~320-340 FPS (using educated guesses for FPS here) really matter? Probably not.... but if it does to you, Intel is the answer. For most people, that isn’t actually a good value proposition tho. 

so am I fine? i just wouldnt want bad fps to be in other games which is my concern.

[noahthedestroyer5]

Just now, Jurrunio said:

Ok it's 50-60% for the first two thirds of the gameplay (30-40% when killed), 70-80% at the last third of gameplay.

so it does play a good amount in FPS? because that makes sense since his GPU is better 

[Jurrunio]

Just now, noahthedestroyer5 said:

so it does play a good amount in FPS? because that makes sense since his GPU is better 

more like combination of many things. Basically when his GPU goes beyond 70%, your GPU will be hitting 100% and start limiting performance if you have the same CPU and memory.

[noahthedestroyer5]

1 minute ago, Jurrunio said:

more like combination of many things. Basically when his GPU goes beyond 70%, your GPU will be hitting 100% and start limiting performance if you have the same CPU and memory.

so hes doing better in performance because of his better GPU CPU and RAM. basically?

[Jurrunio]

Just now, noahthedestroyer5 said:

so hes doing better in performance because of his better GPU CPU and RAM. basically?

probably not CPU (I suspect a 3300x will beat the 3900x and 3700x in CSGO since it only has 1 CCX) as CS only cares about a few cores, but memory and RAM.

[noahthedestroyer5]

1 minute ago, Jurrunio said:

probably not CPU (I suspect a 3300x will beat the 3900x and 3700x in CSGO since it only has 1 CCX) as CS only cares about a few cores, but memory and RAM.

ah ok

[noahthedestroyer5]

2 minutes ago, Jurrunio said:

probably not CPU (I suspect a 3300x will beat the 3900x and 3700x in CSGO since it only has 1 CCX) as CS only cares about a few cores, but memory and RAM.

My ram cl is 16 and everything else is the same. How can Cl  14 make that big of a difference?

[straight_stewie]

10 minutes ago, noahthedestroyer5 said:

How can Cl  14 make that big of a difference?

It doesn't make that big of a difference.

What's going on here is that everything in your computer is a little bit slower than his, and that adds up to what you are seeing. You have slightly slower CPU, a slightly slower GPU, slightly slower RAM, and likely a slightly slower SSD. And while, individually those components might not make that big of a difference, taken altogether as a system the performance can be noticeably lower than a system that uses all top tier parts.

But in reality, you are getting more than acceptable performance. You have a brand new, high performance computer that is well balanced and that's working just fine. 

After all of our previous interactions, my advice to you would be that if you still do not want to get into overclocking, then what you should do is turn FPS counters off in your games, do not run anymore benchmarks, and just have fun playing.