• 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.

AMD Ryzen 7 9800X3D

Just got down to having a bit of a play around this morning and punched in +200 on the core and CO negative 30, had 30 mins of stress testing running and it seems to be ok so far.

What’s everyone else seeing in regards to OC potential on these chips?
 
matty007 said:
In general or for me personally? Isn’t it typical to see a slightly higher latency when raising FCLK past 2133?

I have lowered the VSOC from 1.27v to 1.18v and the GFLOP figures in Linpack are consistent. Testing with P95 Large FFT’s now. Currently at 1 hour with no errors.
Click to expand...
No, latency in Aida should drop with higher FCLK, bandwidth goes up too.
 
LtMatt said:
No, latency in Aida should drop with higher FCLK, bandwidth goes up too.
Click to expand...

Umm. I’m getting conflicting reports about this online. General consensus seems to be that outside of 3:2 ratio (In the case of 6400, 2133mhz) that latency goes up slightly, but general bandwidth and performance increases.

All stability testing seems to point to my FCLK being stable, but I’ll keep an eye on it.
 
matty007 said:
Umm. I’m getting conflicting reports about this online. General consensus seems to be that outside of 3:2 ratio (In the case of 6400, 2133mhz) that latency goes up slightly, but general bandwidth and performance increases.

All stability testing seems to point to my FCLK being stable, but I’ll keep an eye on it.
Click to expand...
Okay, good luck. :)
 
matty007 said:
Umm. I’m getting conflicting reports about this online. General consensus seems to be that outside of 3:2 ratio (In the case of 6400, 2133mhz) that latency goes up slightly, but general bandwidth and performance increases.

All stability testing seems to point to my FCLK being stable, but I’ll keep an eye on it.
Click to expand...
not sure if you noticed it but your cache latency actually went down, flck will affect cache latency more than the actual latency right?

edit: although could just be run variance

edit 2: i noticed increasing to 2200 flck my end improved latency and cache latency but my timings are far from great, just using asus enhanced timings as i cant be arsed to faff around with it, also running 6400mhzcl32
 
Last edited:
henson0115 said:
not sure if you noticed it but your cache latency actually went down, flck will affect cache latency more than the actual latency right?

edit: although could just be run variance

edit 2: i noticed increasing to 2200 flck my end improved latency and cache latency but my timings are far from great, just using asus enhanced timings as i cant be arsed to faff around with it, also running 6400mhzcl32
Click to expand...

Yes. Performance overall improved also.

The difference in latency was so small that it could just be run variance as you said, or even background tasks at that specific moment. Truth be told, the latency doesn’t really bother me, especially if it’s so small.

There is a bug in AIDA with the 9800X3D and the latency figures reported are about 6ms higher than in actual reality.

Raw FPS and performance is what matters, and 2200 showed clear gains.

I’m just finishing up stressing with Prime95 Large FFT’s, which is extremely hard on the IMC and FCLK. Once it’s passed that, I’m completely done tinkering and optimising until the next release.
 
Noxia said:
Thanks, will have a looksie. Didn't do it on a restart earlier, tbh may have been whenever I change a bios setting.
Click to expand...

If Memory Context Restore isn’t enabled, the RAM will be retrained every time the system reboots. It can result in VERY long post times.

After I updated the bios and installed my 9800X3D, I completely forgot to enable MCR before tightening the RAM. It literally took over 20 minutes to post lol
 
matty007 said:
Yes. Performance overall improved also.

The difference in latency was so small that it could just be run variance as you said, or even background tasks at that specific moment. Truth be told, the latency doesn’t really bother me, especially if it’s so small.

There is a bug in AIDA with the 9800X3D and the latency figures reported are about 6ms higher than in actual reality.

Raw FPS and performance is what matters, and 2200 showed clear gains.

I’m just finishing up stressing with Prime95 Large FFT’s, which is extremely hard on the IMC and FCLK. Once it’s passed that, I’m completely done tinkering and optimising until the next release.
Click to expand...
didnt know that about the bug, thanks for the headsup. makes me feel a bit better about my 72oddns at crap timings :D
 
FCLK and IMC have just been cooked in Prime95 Large FFT for two hours. No errors, and 0 background WHEA errors. 2200 is okay.

That's me done with tinkering now. Happy gaming!.

Cw1ciHj.png
 
Last edited:
parisv said:
here's my tweaked ram settings.

https://imgur.com/a/6McZWKS

Any idea why is CPU VDDIO 0.87?
Click to expand...

Your latency is sky high. It’s about 10ns higher than it should be.

Your TRFC is double what it should be, and
Trefi is only 15,000 when it could run at 65535

Tras and trc are at least 5x higher than they should be too.

Although X3D chips aren’t RAM sensitive, you’re totally gimping your system running timings this bad.
 
Joxeon said:
My first time playing with DDR5 so just banged in some numbers for my ram to get the ball rolling as I don't have an expo profile, I'd imagine there is quite a bit of headroom once I dial it in fully.

Screenshot-910.png
Click to expand...
Those are some bogus aida numbers - looks too out of date I think - single CCD chips cannot hit those read speeds for example :)
 
matty007 said:
Your latency is sky high. It’s about 10ns higher than it should be.

Your TRFC is double what it should be, and
Trefi is only 15,000 when it could run at 65535

Tras and trc are at least 5x higher than they should be too.

Although X3D chips aren’t RAM sensitive, you’re totally gimping your system running timings this bad.
Click to expand...
I haven't played with timings at all I just set expo and some voltages that's it.

I'm using this ram: https://www.gskill.com/product/165/390/1723430443/F5-8000J3848H16GX2-TZ5NR

DDR5-8000 CL38-48-48-128 1.45V

So recommendations welcome!
 
parisv said:
I haven't played with timings at all I just set expo and some voltages that's it.

I'm using this ram: https://www.gskill.com/product/165/390/1723430443/F5-8000J3848H16GX2-TZ5NR

DDR5-8000 CL38-48-48-128 1.45V

So recommendations welcome!
Click to expand...


Well, even easy YouTube Buildzoid (Actually Hardware Overclocking) timings would be vastly superior to those god awful timings. Look up his easy Hynix timings. Given you have an 8000 kit, it’s guaranteed A-Die.

However, I can provide you with some tighter timings myself that ‘should’ run, but I can’t guarantee it. All this assumes that you’re comfortable doing timings in the BIOS.

My suggestions just off the top of my head for A-Die would be -

Set trfc to around 500 and trefi to 65535 asap. Those are killing your latency.

Then TRAS to 52. Set trc to 100. Tfaw to 32. Twtrs to 4. Twtrl to 16. Twr to 48.

RDRDSCL and WRWRSCL both to 8.

Trp to 16. RDWR to 20. WRRD to 6

Put 1.45v into VDD and VDDQ
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom