Question bottleneck

[wraith790]

I have a ryzen 9 3900x.
And a X570 Pro motherboard
32GB 3200 speed ram.
I want to get the RTX 4080.

Will I bottleneck too hard if i put the 4080 in?
I am planning on getting more upgrades but right now id prefer only upgrading one component at a time.

3440x1440 res if that matters.

 

[rgd1101]

stop worrying about bottleneck if you going to upgrade 1 component at a time

 

[wraith790]

stop worrying about bottleneck if you going to upgrade 1 component at a time

but will it still perform up to standard or will it bottleneck? would still be nice to know since i dont know how long between upgrades

 

[rgd1101]

so don't upgrade

 

[rgd1101]

can't have it both way. upgrade to increase 1 components performance and at the same time no bottleneck.

 

[Eximo]

He's not wrong. If you are going to buy a shiny new GPU, of course there are CPUs that will deliver better output from said GPU depending on the game and resolution.

When you are purposely mixing old and new, don't worry about it.

 

[geofelt]

There is always some sort of limiting factor. commonly miscalled a "bottleneck" and usually either cpu or gpu.
Try this simple test:
Run YOUR games, but lower your resolution and eye candy.
This makes the graphics card loaf a bit.
If your FPS increases, it indicates that your cpu is strong enough to drive a better graphics configuration.
If your FPS stays the same, you are likely more cpu limited.

 

[Ralston18]

@wraith790

I will add the suggestion to begin by using available system tools to observe system performance.

Task Manager, Resource Monitor, and Process Explorer (Microsoft, free).

Reference - Process Explorer:

https://learn.microsoft.com/en-us/sysinternals/downloads/process-explorer

Use all three tools but only one tool at a time. Focus on understanding more about your system and its' components.

Learn what system resources are being used, to what extent (%), and what is using any given resource.

Will likely vary with circumstances and whatever apps etc. are being run at any given time.

Keep notes. Then, as mentioned, do some testing and keep notes. Measure/observe before some change and then do so again after the change.

Determine if performance actually improves or degrades. Or is even noticeable from an end user viewpoint regardless of how resource %'s change.

Be methodical and change only one thing (hardware, software, configuration) at a time allowing time between changes.

The results are likely to prove helpful in prioritizing component upgrades.

Or demonstrate that all is well and that some upgrade may not prove worth the cost and time.

Just my thoughts on the matter.

 

[sizzling]

Even the very best systems can 'bottleneck' in certain circumstances. I think you really have 2 different questions to ask.

1) Can your cpu deliver satisfactory fps in the games you want to play? Satisfactory depends on what you want, 100, 150, 200? Check out game and cpu reviews for this information.

2) Can the gpu deliver the fps you want at 3440x1440 at the game settings you want?

Someone wanting 200fps at medium to high settings has a much higher cpu requirement than someone aiming for 100fps at max settings. If aiming for 200 fps your cpu could do with upgrading, if aiming for 100 fps your cpu should do very well in the vast majority of games.

 

[USAFRet]

I think this question stems from a misconception of what "bottleneck" actually is and what happens.
(this is very very common)

A bigger 'bottleneck' does not reduce overall performance.

If you put in a better performing part, the overall system performance does not go down. Period.
It goes up.

Replace your current GPU with a 4080?
What happens?
OMG!! Its a bigger bottleneck!! Help me!

What really happens?
You get better performance.
You get the same current framerate from the CPU, but you can boost the graphics level to ultimate CountTheNoseHairs.


When swapping in a better part, it is VERY common to get both a bigger bottleneck (however you measure that) and better performance.

 

[Karadjgne]

I dislike the word bottleneck as it generally assumes there is something wrong, when in reality there's only limitations.
Cpu is fps.
Gpu is eye-candy.

The cpu is going to send the max fps it can to the gpu, no matter what that gpu actually is. It'll send the same 500fps in CSGO at 720p on a GT1030 as it would to a 4090 in CSGO at 4k. The fps on screen is dependent on the gpu, detail settings and resolution. You'd get all 500fps from the 1030 at low, 250fos at ultra. On the 4090 you'd be closer to 250fps on low, 150 at ultra.

Change games to something like the new Harry Potter and those fps will go in the toilet.

So what's the real bottleneck? The cpu, the gpu or the monitor or the game itself.

You are worrying about maybe slipping on the ice outside during the winter, when right now it's summertime and you are in the Bahamas.

 

[wraith790]

@wraith790

I will add the suggestion to begin by using available system tools to observe system performance.

Task Manager, Resource Monitor, and Process Explorer (Microsoft, free).

Reference - Process Explorer:

https://learn.microsoft.com/en-us/sysinternals/downloads/process-explorer

Use all three tools but only one tool at a time. Focus on understanding more about your system and its' components.

Learn what system resources are being used, to what extent (%), and what is using any given resource.

Will likely vary with circumstances and whatever apps etc. are being run at any given time.

Keep notes. Then, as mentioned, do some testing and keep notes. Measure/observe before some change and then do so again after the change.

Determine if performance actually improves or degrades. Or is even noticeable from an end user viewpoint regardless of how resource %'s change.

Be methodical and change only one thing (hardware, software, configuration) at a time allowing time between changes.

The results are likely to prove helpful in prioritizing component upgrades.

Or demonstrate that all is well and that some upgrade may not prove worth the cost and time.

Just my thoughts on the matter.

Thank you for your in depth response to my question