10700K OC + RTX 3080 not giving more FPS on Warzone

[aeskt]

Hello! first of all, this is my PC:

MSI Z490 TOMAHAWK

INTEL I7 10700K + NOCTUA NH-D15s (140mm + 120mm noctua fans)

Corsair Vengeance PRO RGB 4x8 (32GB ) @ 3200 mhz X.M.P enabled 16 - 18 - 18 - 18 - 37 1.35V

RTX 3080 MSI VENTUS OC

Samsung 970 EVO M.2 NVMe 1 TB (main) -> this is where Warzone is installed

Samsung 850 EVO 500GB

WD HDD 1TB

SENTEY 750W 80 plus

 

What I want is to use a 1440p IPS monitor @ 165hz, but I'm ok with stables 144 fps and up.

 

On stock settings, only XMP enabled, I get 154 FPS AVG with LOW settings (competitive) at 1440p

 

I tried many configurations and my CPU seems stable on every of them, I don't know why, it passes all benchmarks, stress tests, I can play any game or whatever. The only thing that changes is the temperature between different onfigurations.

I found this configuration: CPU ratio: 51 AVX offset: -3 VCORE: 1.250 LLC: mode 3 and fixed mode to be the best one in terms of temperatures. Like 76 C on load (benchmarks/stress) and 58 C while gaming. But the AVG FPS are 140 FPS. FPSmonitor shows me that runs at 5000 - 5100 ghz while gaming Warzone.

 

I tried other config: CPU ratio: 51 AVX offset: 0 VCORE: 1.250 LLC: mode 2 and fixed mode. This one is warmer, and I get the same AVG FPS as stock settings. like 153 FPS AVG. Temperature while gaming is 60-64 C

 

So my question is: why I'm not getting more FPS with 51 of CPU ratio, good temperatures and RAM at 3200mhz?

 

[xg32]

hows the gpu usage, run geekbench 5 or cinebench r20/23 to completely rule out cpu, also make sure your cpu isn't power limited by the motherboard.

[aeskt]

1 minute ago, xg32 said:

hows the gpu usage, run geekbench 5 or cinebench r20/23 to completely rule out cpu, also make sure your cpu isn't power limited by the motherboard.

r20  5243 pts @ 5000/5100 ghz 80 C

r23  13576 pts @ 5000/5100 ghz 81 C

 

GPU usage while gaming is 98% on 1440p and CPU usage 40%

 

"make sure your cpu isn't power limited by the motherboard" How can I see that on the tomahawk?

[xg32]

if the gpu usage is already 98% the cpu is likely not the problem, overclocking ram would only net you 3-5% more

cinebench scores look fine. you can see how much power the cpu's using with hwinfo (though power limit is unlikely to be the problem with those scores)

If the gpu is already drawing 300w+ then something else funky is going on. iirc it's suppose to average 170-180fps, unless you are on an ultrawide.

[Avanta8]

31 minutes ago, aeskt said:

GPU usage while gaming is 98% on 1440p and CPU usage 40%

You're completely GPU bottlenecked.

[aeskt]

4 minutes ago, Avanta8 said:

You're completely GPU bottlenecked.

So you mean even using a 5800x, 10900K OC, 59xx or whatever I wont get more FPS? is this rtx 3080 model way worse than others?

[Avanta8]

3 minutes ago, aeskt said:

So you mean even using a 5800x, 10900K OC, 59xx or whatever I wont get more FPS? is this rtx 3080 model way worse than others?

No, almost all 3080s will pretty much perform with 5% of each other. You're probably just playing with max settings. A 3080 should be able to do 180+ fps at 1440p Warzone with reasonable settings.

[aeskt]

6 minutes ago, Avanta8 said:

No, almost all 3080s will pretty much perform with 5% of each other. You're probably just playing with max settings. A 3080 should be able to do 180+ fps at 1440p Warzone with reasonable settings.

All is in low settings, only textures are in normal, without tesselation, without AA.

I have this doubt: does the virtual resolution 1440p (I don't have a native 1440p monitor yet) impact on performance worse than native?

[SkilledRebuilds]

Can you tried tightening ram timings or upclock ram speed at similar timings?

People say 3-5% but it depends on the 'to and from'

 

If you had 3200CL16/17 and went to 3200CL14/15 or 4000CL16 (tweaked further still) I'm sure it helps a decent chunk on this particular games highIPC Low Latency needs when the CPU subsystem is used quite high loads (50%+) the ram speed or reduced timing actually come into effect.

 

You could likely gain more than 3-5% using different ramkits/settings there...

OCing the same ram, yhen yeah 3-5% gains.

[Avanta8]

7 minutes ago, aeskt said:

All is in low settings, only textures are in normal, without tesselation, without AA.

I have this doubt: does the virtual resolution 1440p (I don't have a native 1440p monitor yet) impact on performance worse than native?

Maybe slightly, but not really. Are you playing like 130% render resolution?

[aeskt]

14 minutes ago, Avanta8 said:

Maybe slightly, but not really. Are you playing like 130% render resolution?

mm I don't know. I have DSR Factors all of them checked and DSR smoothness 33%, That's all. I put 1440p resolution inside the game with 107 FOV.

[Mister Woof]

If your gpu us 95%+ it's not your cpu or RAM.

 

Keep in mind other reviews aren't testing exactly the same place in the game you are in

 

It's not apples to apples if only that changes

 

[Avanta8]

Just now, aeskt said:

mm I don't now. I have DSR Factors all of them checked and DSR smoothness 33%, That's all. I put 1440p resolution inside the game with 107 FOV.

Using DSR will have no performance difference compared to native. Can you run this and post a link to your result: https://www.userbenchmark.com/

[iCoNrant]

2 hours ago, aeskt said:

Hello! first of all, this is my PC:

MSI Z490 TOMAHAWK

INTEL I7 10700K + NOCTUA NH-D15s (140mm + 120mm noctua fans)

Corsair Vengeance PRO RGB 4x8 (32GB ) @ 3200 mhz X.M.P enabled 16 - 18 - 18 - 18 - 37 1.35V

RTX 3080 MSI VENTUS OC

Samsung 970 EVO M.2 NVMe 1 TB (main) -> this is where Warzone is installed

Samsung 850 EVO 500GB

WD HDD 1TB

SENTEY 750W 80 plus

 

What I want is to use a 1440p IPS monitor @ 165hz, but I'm ok with stables 144 fps and up.

 

On stock settings, only XMP enabled, I get 154 FPS AVG with LOW settings (competitive) at 1440p

 

I tried many configurations and my CPU seems stable on every of them, I don't know why, it passes all benchmarks, stress tests, I can play any game or whatever. The only thing that changes is the temperature between different onfigurations.

I found this configuration: CPU ratio: 51 AVX offset: -3 VCORE: 1.250 LLC: mode 3 and fixed mode to be the best one in terms of temperatures. Like 76 C on load (benchmarks/stress) and 58 C while gaming. But the AVG FPS are 140 FPS. FPSmonitor shows me that runs at 5000 - 5100 ghz while gaming Warzone.

 

I tried other config: CPU ratio: 51 AVX offset: 0 VCORE: 1.250 LLC: mode 2 and fixed mode. This one is warmer, and I get the same AVG FPS as stock settings. like 153 FPS AVG. Temperature while gaming is 60-64 C

 

So my question is: why I'm not getting more FPS with 51 of CPU ratio, good temperatures and RAM at 3200mhz?

 

I remember this occur a lot actually in Warzone. Is probably Warzone problem not yours

[Mister Woof]

Online stuff is near impossible to benchmark accurately from system to system from moment to moment

[Mister Woof]

What are low competitive settings btw?

 

I set everything to low and getting over 200fps in the gulag area at native 1440p on a 3080.

 

But 150-160 seems about normal for outside areas regardless of settings 

[aeskt]

1 hour ago, SkilledRebuilds said:

Can you tried tightening ram timings or upclock ram speed at similar timings?

People say 3-5% but it depends on the 'to and from'

 

If you had 3200CL16/17 and went to 3200CL14/15 or 4000CL16 (tweaked further still) I'm sure it helps a decent chunk on this particular games highIPC Low Latency needs when the CPU subsystem is used quite high loads (50%+) the ram speed or reduced timing actually come into effect.

 

You could likely gain more than 3-5% using different ramkits/settings there...

OCing the same ram, yhen yeah 3-5% gains.

 

Sure, I can't pay for more parts anymore. I spent a lot of money haha so I will keep this corsairs for now.

 

1 hour ago, Mister Woof said:

If your gpu us 95%+ it's not your cpu or RAM.

 

Keep in mind other reviews aren't testing exactly the same place in the game you are in

 

It's not apples to apples if only that changes

 

 

Yes I know, and that is why I tried to copy benchmarks from YouTube. I go to downtown and start from there.

 

1 hour ago, Avanta8 said:

Using DSR will have no performance difference compared to native. Can you run this and post a link to your result: https://www.userbenchmark.com/

 

Here is the result: https://www.userbenchmark.com/UserRun/40375926

From there: GPU "Performing below potential (65^th percentile)" 

I tried to overclock with afterburner, I got errors with core clock at +110 and memory clock +700. So my GPU does not have margin to overclock I think, so bad

 

53 minutes ago, Mister Woof said:

Online stuff is near impossible to benchmark accurately from system to system from moment to moment

Well, I've watched TONS of benchmarks on youtube and I'm starting to think that they are sh**... all of them always have a lot of FPS and users in comments including myself always are asking how the hell got a lot more FPS than us. I really don't know...

 

I must say that I had a Ryzen 5800x + this corsairs sticks + this GPU. I have returned that crap because of BSODs and temperatures and moved again to intel and I'm happy now. But, I could play for a few hours with that 5800x, I got the same FPS.. 154/156 fps average in downtown.

 

I attach HWiNFO64 screenshots after playing warzone a few hours 

 

 

[Mark Kaine]

2 hours ago, Avanta8 said:

3080 should be able to do 180+ fps at 1440p Warzone with reasonable settings.

it definitely should cause im getting like 170 with almost max settings (just particles on low and raytracing off) 

 

4 minutes ago, aeskt said:

I've watched TONS of benchmarks on youtube and I'm starting to think that they are sh**...

Well of course they are, there's a huge audience for those, just like there was for alien sightings in the 90s/early 00s... as long there's audience for fake, they will produce fake... 

[aeskt]

2 minutes ago, Mark Kaine said:

it definitely should cause im getting like 170 with almost max settings (just particles on low and raytracing off) 

 

 

With that setup in your profile? I can't believe it so... I got the same FPS with two different CPUs and MOBOs. Can you share a video, please? showing settings and fps

[Mister Woof]

Here's mine: 

https://www.userbenchmark.com/UserRun/40327071

 

I was running around the stadium on max settings and getting around 150-180fps

 

It bounced below 120fps a few times and bounced above to 220fps a few times.

 

It doesn't seem to actually matter what settings I use low or high, FPS stays the same, so I presume it is either CPU or network bottleneck.

 

My 3080 is +10% power limit, +75 core, and +200mhz memory.

 

Also, this game is confusing to play.

 

you young whipper snappers and your battle royale games

[iCoNrant]

13 minutes ago, aeskt said:

 

Sure, I can't pay for more parts anymore. I spent a lot of money haha so I will keep this corsairs for now.

 

 

Yes I know, and that is why I tried to copy benchmarks from YouTube. I go to downtown and start from there.

 

 

Here is the result: https://www.userbenchmark.com/UserRun/40375926

From there: GPU "Performing below potential (65^th percentile)" 

I tried to overclock with afterburner, I got errors with core clock at +110 and memory clock +700. So my GPU does not have margin to overclock I think, so bad

 

Well, I've watched TONS of benchmarks on youtube and I'm starting to think that they are sh**... all of them always have a lot of FPS and users in comments including myself always are asking how the hell got a lot more FPS than us. I really don't know..

If you said benchmark charts, just watch with a grain of salt, they are always useless. They just reinstalled clean windows 10 without anything on the task tray what so ever and they do benchmark. 

In real world, people used computer to do like multitask, open chrome, streaming, triple monitor etc and fps on benchmark chart can't showed it at all. 

[Mister Woof]

13 minutes ago, iCoNrant said:

If you said benchmark charts, just watch with a grain of salt, they are always useless. They just reinstalled clean windows 10 without anything on the task tray what so ever and they do benchmark. 

In real world, people used computer to do like multitask, open chrome, streaming, triple monitor etc and fps on benchmark chart can't showed it at all. 

Then you've got the ones like TheSpyHood which is just some dude replaying the same gameplay videos over and over again with made up numbers on top.

[iCoNrant]

6 minutes ago, Mister Woof said:

Then you've got the ones like TheSpyHood which is just some dude replaying the same gameplay videos over and over again with made up numbers on top.

Thats funny really funny

[Guest]

One way you can get more performance is by disable the HT and increase Cache to 4.7ghz+. 

[aeskt]

6 minutes ago, ShrimpBrime said:

One way you can get more performance is by disable the HT and increase Cache to 4.7ghz+. 

Do you know where is that HT option on MSI boards? and cache is ring ratio?