Question New cpu no improvement

LiveIsCruel

Reputable
May 25, 2017
29
2
4,535
0
Hello, I'm looking for a help as I have upgraded my cpu from ryzen 5 2600 to ryzen 5 5600x(I'll put specs below) and there is no difference in performance at all. I turned on PBO without any oc. Please help me as I have no clue at all.

Bios is updated as all the drivers. Gpu when gaming at 80c and cpu barely 65 cooled with corsair.

Motherboard: Asus b450m-a
Memory: 2x8gb 2660mhz
Cpu: Ryzen 5 5600x
Gpu: Nvidia RTX 2060 6gb

PS. please don't mind my nick as I don't mistake somehow when I created account 🤣
 
What games?
What resolution?
What game settings?

If the gpu isn’t strong enough to utilise the extra cpu performance you won’t see any gains. You will only see a gain if the 2600 was the component limiting performance.
 

LiveIsCruel

Reputable
May 25, 2017
29
2
4,535
0
What games?
What resolution?
What game settings?

If the gpu isn’t strong enough to utilise the extra cpu performance you won’t see any gains. You will only see a gain if the 2600 was the component limiting performance.
Games like COD WARZONE, EFT, COD BO, all games tbh
I play 1080 in all games
Cod Warzone competitive low/medium is at about 110-90fps, EFT 60-70fps
 

Andrewbandrew05

Prominent
Jun 30, 2019
221
16
595
2
Run 3dmark and see if your cpu score is in line with what you should expect from your processor you are getting increased performance, just a different component is the limiting factor. Id assume that either a your ram has become the limiting factor or noe your gpu is. If I were you id do some research on bottleneck percentage for your cpu and GPU and see what they say. Compare that with your old cpu and GPU and see what it's like. If assume youve gained a few frames just not that many.
 

Andrewbandrew05

Prominent
Jun 30, 2019
221
16
595
2
Regardless of the issue ryzen cpus should be paired with fast ram. It’s mostly common knowledge
True it does help, question is does it help enough in this case to warrant it? While gaming is becoming more and more multi threaded, I believe that it is still primarily influenced by single threaded performance which faster ram will not help with.
 
Regardless of the issue ryzen cpus should be paired with fast ram. It’s mostly common knowledge, but to you it may be a random suggestion
It helps where cpu single core performance is the limiting factor. That has been ruled by upgrading the cpu to one with a higher IPC and seeing no gain. Yes RAM of 3200-3600mhz with low latency will offer a higher potential FPS but is assumes there are no other limiting factors.
 

Andrewbandrew05

Prominent
Jun 30, 2019
221
16
595
2
Awful suggestion. Do not do that. Bottleneck calculators are meaningless garbage
I'm not saying they're perfect. I'm saying treat them like user benchmark. Not good for being incredibly accurate, but decent enough for a quick comparison. The idea is if there is a hypothetical little to no bottleneck with older cpu then there should be little to no expected gains. I highly recommend running 3dmark and comparing scores tho. That will let you know for sure if your cpu perf increased. (Cinebench would be just as goof if not better for this.)
 

Andrewbandrew05

Prominent
Jun 30, 2019
221
16
595
2
Also about the ram, you could try ocing it a tad. Try to push it to 3600 if possible if not then just as high as it'll go. Before doing this however, familiarize yourself with the location of your cmos reset jumpers so that you can quickly reset the cmos if needed. (I panicked the first time I tried, ripped my cpu out bending the pins and forcing a purchase of a new one.)

If I may ask, what about this post lead you to downvote it? It doesn't really matter I'm just curious.
 
Last edited:
Reactions: LiveIsCruel

hotaru.hino

Commendable
Sep 1, 2020
1,605
497
1,640
36
Looking at several Ryzen memory scaling articles, at least for Zen 2, while you can show that memory speed can have a somewhat appreciable impact to performance, appreciable bumps are few and far between. If anything, I'm calling that the Ryzen 2600 was good enough for the RTX 2060 and the previous setup was GPU limited to begin with.

I'm not saying they're perfect. I'm saying treat them like user benchmark. Not good for being incredibly accurate, but decent enough for a quick comparison. The idea is if there is a hypothetical little to no bottleneck with older cpu then there should be little to no expected gains. I highly recommend running 3dmark and comparing scores tho. That will let you know for sure if your cpu perf increased. (Cinebench would be just as goof if not better for this.)
I don't like Userbenchmark, or really comparing scores against user-submitted ones, because they weigh heavily towards people who have overclocked or tuned their systems. So they can cause people think their parts are underperforming when in fact, they're within the expected performance.
 

Karadjgne

Titan
Ambassador
A frame takes a certain amount of time to populate, regardless of cpu power. Game code has to be deciphered, objects placed, files dl'd from ram-storage, Ai, computations for damage or movement etc. The amount of times a cpu can finish that ore-render is cpu fps.

All that data gets sent to the gpu, which adds color, shading, follows cpu instructions and paints the picture on screen. That too takes a certain amount of time to finish rendering the frame. That's gpu fps output and that's what you'll see on the bean counters.

So matters very little if a 2600 can supply 300fps and a 5600x can supply 500fps, if the gpu can only output 150fps.

Its only when the cpu struggles, 2600 supplys 60fps and the 5600x can supply 100fps and the gpu could output 150fps, that you'd see the performance difference.

In your case, you see little improvement because the gpu is the limiting factor, regardless of detail settings, it simply cannot render as many frames as either cpu could supply.

Move upto a 3080 and you'd have a gpu capable of 400fps output, so you'd see the 300fps of the 2600, but see the leap to 400fps from the 5600x as now only that is exceeding the 400fps possible.
 

Andrewbandrew05

Prominent
Jun 30, 2019
221
16
595
2
Looking at several Ryzen memory scaling articles, at least for Zen 2, while you can show that memory speed can have a somewhat appreciable impact to performance, appreciable bumps are few and far between. If anything, I'm calling that the Ryzen 2600 was good enough for the RTX 2060 and the previous setup was GPU limited to begin with.


I don't like Userbenchmark, or really comparing scores against user-submitted ones, because they weigh heavily towards people who have overclocked or tuned their systems. So they can cause people think their parts are underperforming when in fact, they're within the expected performance.
I agree with your idea of what happened. I do still think userbench and benchmark scores can be helpful for a very rough and basic idea of performance. Mainly its just easier than digging through benchmark articles. The articles are better when looking for benchs for an individual component but its pretty hard to calculate a bottleneck for two cpus and one gpu. The one thing i like userbench for is 'hey I wonder approximately how good my gpu/ cpu is compared to this other one?' If I actually want accurate info I use benchmarks from PCWorld Toms etc
 

LiveIsCruel

Reputable
May 25, 2017
29
2
4,535
0
Also about the ram, you could try ocing it a tad. Try to push it to 3600 if possible if not then just as high as it'll go. Before doing this however, familiarize yourself with the location of your cmos reset jumpers so that you can quickly reset the cmos if needed. (I panicked the first time I tried, ripped my cpu out bending the pins and forcing a purchase of a new one.)
I know I can push my ram to 3200mhz as it is currently oced but i don't really know how to deal with timings tho.
 

LiveIsCruel

Reputable
May 25, 2017
29
2
4,535
0
Looking at several Ryzen memory scaling articles, at least for Zen 2, while you can show that memory speed can have a somewhat appreciable impact to performance, appreciable bumps are few and far between. If anything, I'm calling that the Ryzen 2600 was good enough for the RTX 2060 and the previous setup was GPU limited to begin with.


I don't like Userbenchmark, or really comparing scores against user-submitted ones, because they weigh heavily towards people who have overclocked or tuned their systems. So they can cause people think their parts are underperforming when in fact, they're within the expected performance.
So the best way is to upgrade GPU right? What GPU will be a good shot for 5600x?
 

ASK THE COMMUNITY

TRENDING THREADS