cancel
Showing results for 
Search instead for 
Did you mean: 

X670 resource

Shamino
Moderator

ill use this thread to collect some new test bioses for the boards, maybe also to explain some less understood options

to disable cores ccd go here and choose ccd xx bit map down core.
each ones stand for an enabled core
best to disable from the back, ie:
110000
instead of 0011000
after selection press downcore apply changes or discard if made mistake

ocpak/octools

FAQ:
7950x not boosting pass 5.5G -> check that CStates is not disabled
Detailed Explanation on CState Boot Limiter


Test BIOSes:

new:
X3D OC Preset for those MB with asynch BCLK Support: (for simple slight perf boost for X3D)
97792

DOCP/EXPO Tweaked: (for simple timings tightening)
97793

strixe-e 1515 

strixe-f 1515 

strix e a 1515 

crosshair hero 1515 

crosshair gene 1515 

crosshair extreme 1515 

creator 670 1515

creator b650 1515

strix 650E I

strix 670 itx

 

 

for crosshair and strix e-e:

explanation of segment2 Loadline:

dualseg.jpg

customize a heterogenous loadline for a dual segment workload range.

example above shows loadline=L6 when current is in range of 0~40A, and Level4 when current is above 40A.

 

 

 


Adds for x3d

dynamic ccd priority switch with core flex, os / driver agnostic so win10 win11 ok

97403

97404

Algo as follows:
If condition reached and ccd0 specified, then check current mem/cache activity > threshold and hysteresis reached, if fulfilled then switch
If condition reached and ccd1 specified, then check current mem/cache activity <=threshold and hysteresis reached,, if fulfilled then switch
Default hysteresis =4

Can combine multiple algos for ccd priority so combinations are wide

works on non x3d too but of course senseless on it. detailed explanation here.

505,613 Views
2,305 REPLIES 2,305

NP.

I don't think this will be something that could be fixed. My guess is that this is caused by how young these platforms and the ddr5 controllers are. It happens every generation, you can even look at it from just the Asus ROG Crosshair VIII Extreme for example. The 5000 and 3000 CPUs and the 5000 and 4000 APUs all could go as far as 5000MHz on DDR4 but the 2000 CPUs topped out at 3600 MHz which is almost certainly related to advancements in the memory controller and the fact they split from one large die to the chiplet design we have now. And if you look at it just as a whole, DDR4 released in 2014 and the JEDEC speed was 1600 MHz but as it evolved the RAM pushed more base speeds and it ended with the JEDEC standard reaching ddr4 3200 so that forced the memory controllers and cpus to improve to keep up. 

For the 3d modelling, the best recommendation I can give you if you're still just learning or you're still young in the business is avoid using the super high resolution texture files where possible (like avoid the 8K stuff and even 4K for smaller or harder to see objects). The textures are what gets you cause after the GPUs VRAM gets filled (which is inevitable unless you have like multiple RTX a6000s with 48gb of VRAM each or even just 1 if it's smaller modelling stuff) it'll spill into RAM and too many of those high res textures will make it harder to build more complex scenes because the renders will just error out due to a lack of graphics memory available (graphics memory is VRAM plus a portion of RAM thats allocated as shared graphics memory).

BoostedSTIG
Level 7

Hello everyone,

I've got an X670E-F Strix board with a 7950X3D, on the newest BIOS 1003. I seem to be stable using XPO 6000CL30 2x16GB, and now am able to use MCR with Power Down Enabled with no issues. Cool.

But what I am wondering about here is what does everyone's idle power usage look like? My system seems to be functioning 100% normal, c states, core parking, etc., but I've got my PC plugged into a UPS that has a draw display, it just seems like this machine is drawing way more power than my previous machines by not a small margin.

For clarity, my 2 previous setups were all using the same case/fans/PSU/storage disks/AIO/etc. Before, with both my 5950X and 5800X3D, I was seeing power draw reporting around the 90W range out of the wall at idle, and with this machine I'm sitting at like 125-130W no matter what.

Also, to be clear, I don't really care about it from a financial standpoint, I'm just wondering if this is similar to what others are seeing and if this is considered normal. 

Thanks

kamyk155
Level 10

Hello guys again. After few days of fighting with USB disconnect problems in windows, 3 new bioses and a lot of testing I found the culprit - corsair iCUE. Final test was to uninstall it - and for third day now without issue.

Unfortunately - another problem appeared at the start of the computer:

"The VGA card is not supported by UEFI driver.
CSM (Compability Support Module) settings have been changed
For further adjustments, press [F1] to enter BIOS setup or F2 to change settings and continue or something like that.....

My card is RTX4080 Palit Gaming PRO.

 

I have one BIG question - why in the new bioses those two fast boot options are again on auto or disabled and we need to turn them on again ? I'm talking about POWER DOWN and MEMORY CONTEX RESTORE.

And one more thing - for now computer is stable with EXPO and PBO EXHANCED 80*C.

I can even tweak memory to 6200MHz just by change it one click from 6000:

ram.jpg

The VGA card thing could just be a firmware issue, I see from the screenshot you're using the beta BIOS 1003 maybe downgrade to an official release like 0925 and see if that changes anything.

As for the fast boot options, I think that the BIOS update probably changed something in the firmware related to these boot options so they simply reverted to the default.

lolzdetroit
Level 8

ok after more testing, i realized i have an intel xmp ready set of ram!!! no matter what i do in bios the ram never actually leaves 3200 mhz!!!!

OK so finally figured somethings out. ok so because it was branded "intel xmp ready" all of my monitoring systems are for some reason pulling that info. i went into task manager and the ram settings actually are taking my changes. im successfully running both sticks at 6200 mhz, i down scaled them just  a little. thanks again for all the help

 

That's good to hear

Squidbean
Level 9

I'm posting this because I Imagine some of you might find this interesting! 

So with the official release of the beta bios 1003 that gives support for high density memory (the new 24 and 48GB ddr5 sticks), I went and got the new Corsair vengence 192GB kit (4 sticks of 48GB) with XMP (not EXPO) 

and holy **bleep**...  So for context, because I was an extremely early adopter the options for EXPO memory was limited when I got my original 128GB (4 sticks) of Corsair Dominator, they were technically 2 kits of 64GB making 128GB which, should be fine as they will be running dual channel anyway, and not quad channel like on Threadripper. So like many others with 4 sticks, plenty of memory issues, some bios updates eventually made at least using the EXPO profile II stable, as long as you ran the official 3600mhz for 4 sticks (even though the modules individually were 5600mhz). So it worked, just how it all worked. 

Now to my surprise.. I have installed the new 4 stick kit, 192GB 5200mhz XMP. 
XMP profile II runs and after testing at 3600mhz, as that made the most sense, everything looked good. However! to my surprise after a "why not" I dropped it to its default 5200mhz and boom! it works 🙂 

4 sticks, 192GB ram at 5200mhz with XMP profile.. boy is this nice!  

for reference I have the Asus Crosshair X670E Extreme motherboard and a 7950x (only OC is curve optimizer at -30 on all cores) 
Not sure if I can post images here so I made a quick imgur link with a cinebench score and the cpuZ

CBr23 = 39422 

Anyway! I am so so happily surprised that this is even a thing with these ram modules. 

"Corsair Vengeance 192GB (4x48GB) 5200MHz CL38 DDR5"

This is great. How stable is it? Have you run MemTest86, for example?

 

I bought the same 192GB kit and the DOCP profile worked! I did get a couple of errors in MemTest86, but these were easily resolved by upping the RAM voltage from 1.25v to 1.26v.

192GB running at a very stable 5200Mhz. Very, very cool!