• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

5900x Low TimeSpy Score

PureJaz

PureJaz

PureJaz

Associate
Joined
26 Sep 2012
Posts
205
Hi All,

Just installed my 5900x and wanted to see how it compares to my 3900x.

Ran TimeSpy and it only got 10,000 which seems incredibly low. I think my 3900x actually beat this.

During the benchmark the CPU did not hit 70 degrees (aio cooled).

Anyone got any ideas as to what I should look into to see what maybe causing the low score?

Many thanks
 
10K CPU score? that would be low considering my 5800X gets 12K+

If so I would make sure all the cores are showing as available and in use task manager and ryzen master etc
 
I had a low timespy cpu core too with my 1950x. Score was normal after going from 2 to 4 memory modules. I'd look at memory first but not sure the 5900x would benefit from more modules as that cpu doesn't support quad channel?
 
All cores show up in task manager/ryzen master.

I have 2 memory modules. Fairly high end 2x16 gig 3200 cl14s. In slots A2 and B2.

I will check to see if they are being picked up properly and maybe ill try an oc on them as they oc very well.

I suspected it could be a memory issue.
 
UPDATE

So figured out that my memory was being restricted to 2133mhz. Changed this too 3200 mhz.

When running TimeSpy my score increased to just over 11,000. So its still slower than normal but getting there.

This time though the CPU didn't go above 60c. This makes me feel something is causing the cpu to not use its full potential. Any ideas?
 
Good afternoon,

I am actually facing the same problem albeit only having one stick of 16gb of ram (3200 cl 16 trident z royal) until tomorrow (I had to cannibalize my old system until my new ram gets here).
My score is currently around 11000+- 80, everything stock. Using PBO i can go up to 11300.

My scores in R20 or even time spy extreme are actually normal compared to online reviewers and article.

Have you ever figured out the issue?
 
PureJaz said:
UPDATE

So figured out that my memory was being restricted to 2133mhz. Changed this too 3200 mhz.

When running TimeSpy my score increased to just over 11,000. So its still slower than normal but getting there.

This time though the CPU didn't go above 60c. This makes me feel something is causing the cpu to not use its full potential. Any ideas?
Click to expand...

Is that 11k your Time Spy Score or your CPU score?

What are your PBO settings?
 
Thibault Deneve said:
In my case that is cpu only, I believe it is fair to imagine the OP is probably in the same situation
Click to expand...

I just tested mine and got 13532 on the CPU score. Looking at your specs, the only difference I can see between us is your RAM.

I'm new to these modern AMD chips. Does having 3200Mhz RAM mean your FLCK is at 1600Mhz? I know that the FLCK speed has a big impact on CPU performance when the load is CPU limited. Beyond that, a 1:1 FLCK to DRAM frequency is also important. My RAM is 3600Mhz, which means my FLCK is at 1800Mhz. Could that be the difference?
 
Thanks. I will test tomorrow night ( US east coast time) when I receive my new kit. I'll post my findings on here (hopefully good news haha).
I am brand new to AMD as well that is is I am looking for help a little more easily than I would for intel where I feel a bit more confident.

If I get anything around 12500 at basic xmp i'll be happy
 
@Thibault Deneve you've not answered 2 people asking if you mean CPU Score for this? And you're not reading the overall score with GPU?

I get 17,892 combined Time Spy score... breaks down to 13,428 CPU and 19,008 GPU - with a 5900X and 3080 FTW3 Ultra for comparison.
 
Sparx said:
@Thibault Deneve you've not answered 2 people asking if you mean CPU Score for this? And you're not reading the overall score with GPU?

I get 17,892 combined Time Spy score... breaks down to 13,428 CPU and 19,008 GPU - with a 5900X and 3080 FTW3 Ultra for comparison.
Click to expand...


I did 3 post above actually. It is CPU only.

My gpu was around 21000 which was well within spec, that's why I didn't bother with it nor the combined.
 
Sparx said:
@Thibault Deneve you've not answered 2 people asking if you mean CPU Score for this? And you're not reading the overall score with GPU?

I get 17,892 combined Time Spy score... breaks down to 13,428 CPU and 19,008 GPU - with a 5900X and 3080 FTW3 Ultra for comparison.
Click to expand...

I think you're getting the chap that started this post confused with Thibault! I agree though, if the overall score for PureJaz is 10k then there may well be no problem at all. My overall score is 10.3k with a 5900X and 1080Ti.
 
Sorry yes I did! Should’ve said PureJaz :)

Running on too little sleep and not reading properly! Time to catch up on it...
 
I get 14850 in time spy cpu test with my 5900x, no OC and on Air cooling so 11k does seem low,
what m/Board, bios, drivers, and on HWiNFO64 what do each chip boost to max?
 
Just ran a test with the 5900x

deJPkMZ.jpg
 
SimonV6 said:
I get 14850 in time spy cpu test with my 5900x, no OC and on Air cooling so 11k does seem low,
what m/Board, bios, drivers, and on HWiNFO64 what do each chip boost to max?
Click to expand...


Dayum!
I have a Crosshair VIII hero wifi with bios 3003 from 2 weeks ago, clean windows install from Tuesday on a new hard drive, nvidia driver are up to date.I will have accurate temps and all when i rerun the test tomorrow after swapping the ram but if my memory serves me well, the chip will boost north of 4.9 to even 5.0 for a few quick bursts, but most of the benchmarks hovers around 4.580-4.600 (all at stock no PBO or anything). temps don't reach above 65 (and more like 61 but don't quite quote me on that, I would have to rebench to check because I'm pretty sure it was actually in the mid 50s). I don't go above 77 on the hottest core during asus realbench stresstest that's for sure (and that was before i activated my oc mode on the RTX 3090 which has more aggressive fan curves and makes for a lower overall temp). I downloaded the latest chipset drivers from AMD directly. I also made sure the power setting is on performance and all core/ threads are enabled in task manager.


I really hope that running a single dimm is the culprit and that it will be fixed tomorrow when I get the new kit.

my R20 score is in the 8360-8420 (i got a 89xx with pbo enabled) (similar to both Jayztwocents, LTT, and a couple of other online reviews) and my time spy extreme is 7360-7400 which is also similar to reviewers that run it (all of this completely at stock settings asides from that one try with pbo enabled, I don't use ryzen master)

I take it both of you run your ram at 3600 or above?

Thanks y'all!
 
Hey everyone,
sorry for the delay USPS has been overwhelmed with mail so my package was late by a day.

Good news. with my new ram kit (exact specs are in my signature) i just benched a 13768 with stock values on the cpu and the ram at its DOCP speed of 3200. I pretty much think that the issue is resolved in my end. Thank you very much for everyone's input. I will fine tune later to see what else i can get.

All in all I think it was an interesting experiment to run as well and shows the strength of single vs dual channel. Btw I first ran the ram at stock value (2133 or 2400 I imagine) score was 11383. so we can see how much ryzen benefit from higher speed.
 
You must log in or register to reply here.
Back
Top Bottom