Soldato
Strange but yeah, it's wise not to fret over an almost undetectable difference that may or may not be realised by it being being set any higher.
-
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.
5800X Boost set to 5050Mhz all cores with AMD Curve Optimizer
- Thread starter gerardfraser
- Start date
5800x user here using an b450 tomahawk MAX, i go into motherboard settings- AMD overclocking advanced options etc, change PBO and curvature settings, press F10 to save my settings and it keeps saying that no settings have been changed, any advice?
Have you checked to see if it saves the settings anyway? MSI's BIOS is a bit weird about the AMD overclocking menu settings in my experience. It doesn't seem to save them properly in overclocking profiles either. When I was using a 3100 as a temporary CPU in my X570 Ace I had some trouble with overclocking, as it would never load the settings back properly if they were part of the universal AMD stuff, even though it showed them as changing. I'd have to go toggle them to a different state and then back to get them to set properly.
Associate
- Joined
- 5 Apr 2004
- Posts
- 1,197
Interestingly I've found increasing pbo to +200 or 100mhz seems to slightly lower cinebench scores, whilst using a curve optimiser of between-25 and -15. Why do you think that is?
For multithreaded Cinebench, increasing clocks on 5800X essentially does nothing, because clock ceiling is never hit in the first place.
just a random fluctuation
Associate
- Joined
- 5 Apr 2004
- Posts
- 1,197
R23:
Multi Core 15705
Single Core 1592
-25 in Curve optimizer
Ryzen 5800x
Multi Core 15705
Single Core 1592
-25 in Curve optimizer
Ryzen 5800x
Been tweaking the curve on mine and have found that I have two cores that can do -15, three that can do -25 and three that can do -30. That's with a +150MHz boost offset which allows it to hit 5GHz under lighter loads. OCCT's single-core test seems pretty good for finding errors fast when you go too far, and be set to cycle through cores automatically. Only been adjusting in +/-5 increments, so could probably get more granular, but this seems good enough. 16107 from a quick R23 run, which is ~900 more than it scored at stock.
I've just started teaking my curve optimiser and set all to -10. In OCCT It was showing errors on physical core 2 so adjusated it back to -5 but when adjusting I noticxed that I had set something incorrect from the factory wattage. Set this correctly at 142W now and re-ran OCCT wiith all cores at -10 again. This time it is showing errors on physical core 1 only?
Associate
- Joined
- 1 Nov 2010
- Posts
- 234
My x570 tomahawk did that on an older bios, the settings did stick but would reset now and again, on the latest bios it all works perfectly. Are you on the latest bios? it might just be that they are a bit slower releasing stuff on B450 so might need to wait for a new bios.
Soldato
On my MSI mobo PBO settings show up in two sections of the BIOS and only when doing from one of them(the OC\ section I think?) does it show prompt that settings have been changed.
Soldato
I think there's something wrong with my 5800X, I'm seeing in loads of reviews that in Cinebench R20 people are getting 6100 in MT and around 620 in ST, Mine gets 5500 in multi and 540 in ST, Even with the curve optimiser boosting the clocks to 4.80+GHz on all cores during an R20 run it doesn't get any better.
Last edited:
540 is way too low. Can't be running at 4.8GHz and scoring so low at same time
Is it only Cinebench or other benchmarks too?
CpuZ single core should be ~650 I think
How are your temps?
Soldato
Figured it out, Corsair iCUE and Razer Synapse together gimp the performance.
Soldato
What settings on these 5800x are a guaranteed minor reduction in temps and boost in performance without the need for extended testing? - in otherwords, like OEM+
5% reduction in limits so:
PPT: 135
EDC: 133
TDC: 90
Scalar at auto with no +boost and -1 or -2 negative with the curve optimizer?
5% reduction in limits so:
PPT: 135
EDC: 133
TDC: 90
Scalar at auto with no +boost and -1 or -2 negative with the curve optimizer?
hello, was wondering how my score of my 5800x is compared to you guys, im at 1667, 16091 (though i actually reached 16300 a really cold day at boot) but 16050-16180 is like the avg score.
im running nzxt 280mm x63 rgb, with liquid metal, and did +150 boost, wiith ppt settings at 139, 160, 200 and co at -21 2 best cores, -27 2 second best cores, -30 4 rest cores
+ i did negative offset of the cpu voltage, at 0.0625v, allday stable, getting 13100-13200 timespy score
im running nzxt 280mm x63 rgb, with liquid metal, and did +150 boost, wiith ppt settings at 139, 160, 200 and co at -21 2 best cores, -27 2 second best cores, -30 4 rest cores
+ i did negative offset of the cpu voltage, at 0.0625v, allday stable, getting 13100-13200 timespy score
Last edited:
- Joined
- 27 Apr 2014
- Posts
- 857
Well here is some cinebench scores from the 5800X from 2 years ago,I think I did these test at 5175Mhz on single thread.
Well here is some cinebench scores from the 5800X from 2 years ago,I think I did these test at 5175Mhz on single thread.
How did u obtain that High clock speed. Cant get more Then +200 clockspeed so Max is 5050mhz. If i Enter 300 the value automaticly changed to 200. At +150 @5ghz it hits Those 1667 and 16000+ multi. Higher Then scores goes Down. At r20 i Got it to 656 singlecore and 6200 multi
AMD capped the boost override increase to +200, on very early bios you could do +500.