• 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

I am using 32GB RAm Kingston Fury KF564C32-16 6400 xmp sticks. and since the 7700 not able to run it on 6400.. Asus tuf x670 e plus
I tried on the 9800x3d and applied DOCP 1 and enabled 6400 mhz from the dropdown box. And it boots but eventually my gpu will crash and display will stay on blackscreen.
With nlvdmm errors in the logbook when i reboot.
It is really annoying i can't seem to run these sticks at advertized speeds.

Whilst memory overclocks are never guaranteed, that does seem a little strange.

Are you using EXPO profiles? I would try using slack timings at 6400mhz and pumping more voltage through. You probably just need more voltage.

Whilst I could never get my Trident Z5’s fully test stable with the 7800X3D at 6400 (hoping that will change with the 9800X3D), I suspect that was because I was trying to stay fairly tight.

Up the voltage a bit and go loose, and go from there.
 
Yes but own it, don’t try to convince yourself you are future proofing :)

I agree that future proofing is an odd term, and I suspect it means different things to different people.

For me though, I see it as a thing I do every 2-3 years now. It's unlikely I'll buy another processor in this generation unless the 9800x3d turns out to be lame duck, and I skipped the RTX4000's knowing I'd treat myself to a 5080 ot 5090 (looks like 5090 due to lack of VRAM on the 5080).

I used to think I'd buy at the price/performance curve point before it went vertical, but bitcoin has crapped all over that idea in the graphics card arena, and there isn't that much movement in the mobo/ram/cpu arean either.
 
Whilst memory overclocks are never guaranteed, that does seem a little strange.

Are you using EXPO profiles? I would try using slack timings at 6400mhz and pumping more voltage through. You probably just need more voltage.

Whilst I could never get my Trident Z5’s fully test stable with the 7800X3D at 6400 (hoping that will change with the 9800X3D), I suspect that was because I was trying to stay fairly tight.

Up the voltage a bit and go loose, and go from there.
I am using the XMP profile 1 yes. https://www.kingston.com/datasheets/KF564C32RS-16.pdf
As is seen here it should use 1.4V. But yeah. no go on both cpu's it seems...
Yes i hope you will better results then me...
You mean up the voltage even over 1.4V?
How much and how would i even do that? IS IT DRAM VDD AND VDDQ setting? i can only increase to max 143500mv unless i disable high dram voltage mode. also in extended memory menu i see 2 options again that seem to control voltage called memory 'something" forgot. not sure if they are one and the same but...
 
Last edited:
I am using the XMP profile 1 yes. https://www.kingston.com/datasheets/KF564C32RS-16.pdf
As is seen here it should use 1.4V. But yeah. no go on both cpu's it seems...
Yes i hope you will better results then me...
You mean up the voltage even over 1.4V?
How much and how would i even do that? IS IT DRAM VDD AND VDDQ setting? i can only increase to max 143500mv unless i disable high dram voltage mode. also in extended memory menu i see 2 options again that seem to control voltage called memory 'something" forgot. not sure if they are one and the same but...

It’s M-Die so it’s actually fine for well over 1.5v, but depends on cooling etc of course.

I believe the XMP 1 profile for that ram is CL32, so quite decent at 6400. You could try upping the voltage to 1.45v.

If it doesn’t achieve stability at that, I’d personally shoot for 6200CL30, or even 6000Cl28. Should be fully achievable.
 
Last edited:
Last edited:
I posted this on the other 9800X3D thread (taken from the OCUK site)..

oRlK9mx.jpg
 
I don’t care how often Steve or any other tech bro tell me “real world” doesn’t count. I’m not wasting money on a new CPU if it isn’t giving me real world performance improvement TODAY.

Edit: before someone tries to tell me “you obviously don’t get it”. I do get it, lower settings shows CPU overhead. But for real world it’s far better to get on an upgrade cycle every 3 or preferably 4 generations. That way the benefits are actually realised rather n real world gaming.

Well yea, duh. If someone has a 7800X3D already upgrading to a 9800X3D is probably not worth it outside of very specific cases.

Real world is a stupid way to test because what is real world to you is different for me so there is no objective benchmark to test.

The other important point is that most people will target a desired frame rate and then tune the settings until they hit it (or the settings max out). That means if your desire is a 4K 120fps experience you need a GPU that can do it at an IQ you are happy with and a CPU that can sustain that throughput. No amount of tuning settings will get a CPU that can only manage 60fps upto 120fps regardless of the resolution.

As for when you should upgrade, the answer is easy, when you are playing games and it feels like it could do with an upgrade because a game you want to play is not playing well.

I built a PC for playing some older games so it was based around a 2200G, then I got into SWTOR and a few newer games so I got a GPU that could handle them. Then I moved to playing a lot of Civ and Stellaris because they had some new content and I upgraded to a 5800X3D to handle those games and to give the platform more legs so it could support another GPU upgrade if I felt the need for one. Right now my system is entirely unbalanced because it has a 4K screen with a 6600XT and a 5800X3D. Not really a problem though as I do most of my gaming on a steam deck due to a change in circumstances. At some point I will balance this rig out and give it 32GB of ram and a 4080s/4090 tier GPU when that performance level is more affordable and it will only be after that where I bother with doing a full rebuild rather than piecemeal updates.
 
It’s M-Die so it’s actually fine for well over 1.5v, but depends on cooling etc of course.

I believe the XMP 1 profile for that ram is CL32, so quite decent at 6400. You could try upping the voltage to 1.45v.

If it doesn’t achieve stability at that, I’d personally shoot for 6200CL30, or even 6000Cl28. Should be fully achievable.
I set mem to 1.45V and did the extreme 1 test on testmem 5. after 15 minutes peak temp was 57.5 degrees. and found "1" error... So what now? do the 6200 thing ig? but keep the voltage at 1.45V?
 
They will most likely be packed based on order time. First in, first out...

If an order came in at 13:59 snailmail, I'd expect it to be boxed and shipped before a 14:09 next-day.
I'm hoping that's how they do it. I ordered mine dead on 2pm. If people that ordered theirs after me get theirs shipped first I would be a bit miffed.
 
I'm hoping that's how they do it. I ordered mine dead on 2pm. If people that ordered theirs after me get theirs shipped first I would be a bit miffed.

They might well get shipped out before yours depending on when the various companies do their collections.

However I wouldn't worry about it, it'll just be a difference of hours at most and ultimately when you get yours will depend on the shipping method itself not the dispatch time.

Really your just overthinking it :)
 
Last edited:
I'm hoping that's how they do it. I ordered mine dead on 2pm. If people that ordered theirs after me get theirs shipped first I would be a bit miffed.
not how the real world works my place orders print of the guys pick the orders in-between the printer and the warehouse they could end up in any order

now if the system is completely digital then it would be first come first served
 
Back
Top Bottom