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.

42,909 Views
2,100 REPLIES 2,100

install 1303

already appeared

Same here - X670 Extreme BIOS 1302 - yesterday some "expert" recommended putting an offset on soc - which i did. Before i had soc on 1.2V Manual - normally it was 1.35, so i put an offset -0,15. Result: System Bootet once. After a reboot it was stuck an i heard the power suply lcikcing, i thought it died. I had to pull the powercord from the wall and reset. 

Without "expert" offset it works fine on 1202. And Soc was not "variable" with offset -0,15. It was nailed at 1.2 in Aida, Cinebench, Prime, Karhu btw. 


@pz123456 wrote:

Before i had soc on 1.2V Manual - normally it was 1.35, so i put an offset -0,15. Result: System Bootet once. After a reboot it was stuck an i heard the power suply lcikcing, i thought it died. I had to pull the powercord from the wall and reset.


It's not the case that every system can use the same offsets and settings. And no, the SOC voltage is certainly not a static thing. On my X670-P (BIOS 1412 BETA currently) the SOC voltage ranged from 1.18 to 1.20V within a few seconds after starting HWINFO64. I am currently running Offset -0.05V. But I could confirm that an extreme offset of -0.15 also would not allow me to start my system.

Yesterday I was on offet -0.075V but found the system to spontaneously reboot once. So I increased slightly to -0.05V.

Different BIOS might require different offset, so you might have to find out after each upgrade where the limits are. Also I see today BIOS 1413 (1303 for some E-Boards) seems to have be released where the change notes say "SoC voltage for Ryzen 7000X3D series limited to a maximum of 1.30V to protect the CPU and motherboard."

Interestingly only X3D CPUs are referenced here and they might have introduced a hard-cap to 1.3V maximum so under no circumstances it's going to spike above 1.5V eventually frying something. Perhaps this is just an intermediate solution until the cause of the spikes is found (might be a bug in AGESA).

So SOC voltage is dynamic and this is why an offset mode is actually provided. Setting a fixed limit might apply higher voltage in some situations than actually required. Or cause the limit to be too low in situations higher voltages are actually required.


@SkyBeam wrote:

Setting a fixed limit might apply higher voltage in some situations than actually required. Or cause the limit to be too low in situations higher voltages are actually required.

Might. Period.

Telling people to use offsets is the best way to make their board past beyond 1.30 VSOC.

Fixing 1.25 VSOC  with auto or lvl 1 LLC, is the safest way to go and I can tell it doesn't involve instability on my rig.


Telling people to use offsets is the best way to make their board past beyond 1.30 VSOC.

Fixing 1.25 VSOC  with auto or lvl 1 LLC, is the safest way to go and I can tell it doesn't involve instability on my rig.


Unless you know the exact code used for voltage regulation this is just an assumption

You're essentially asking people to permanently overvolt the SOC to 1.25V just to avoid situations where it "might" exceed safe voltages under situations which are not yet clear. Also it's not yet clear if permanently fixed voltage prevents voltage to spike (bugs might cause to ignore fixed voltages under some circumstances too).

So my SOC on "Auto" using EXPO-Tweaked and MC voltage of 1.37V is running at SOC 1.176V currently (set to AUTO, offset -0.05) in BIOS 1413.

Also there is just a BIOS released which seems to cap SOC voltage to 1.30V under any circumstances. So it might be time to revise the advise to set a fixed voltage very close to the absolute maximum.

But sure this is just an advise from my experience and my advise might be as good as yours and everyone is free to chose any settings but has to live with the consequences. Having said that the only safe advise until further notice might be to use Auto/JEDEC timing and not using EXPO.


@SkyBeam wrote:


this is just an assumption

 "might"  might might might  might


I'm not the one making asumptions and advising people to go the funky undetermined offset way 🤣

Buildzoid confirmed 1.25 VSOC works, I can confirm it works, and there's no stupid "might" in this sentence.

Most people, and it seems it's the same for you, actually don't know what will happen on F10 using offset roulette, so don't tell people to use it.


@SkyBeam wrote:

Telling people to use offsets is the best way to make their board past beyond 1.30 VSOC.

Fixing 1.25 VSOC  with auto or lvl 1 LLC, is the safest way to go and I can tell it doesn't involve instability on my rig.


Unless you know the exact code used for voltage regulation this is just an assumption

You're essentially asking people to permanently overvolt the SOC to 1.25V just to avoid situations where it "might" exceed safe voltages under situations which are not yet clear. Also it's not yet clear if permanently fixed voltage prevents voltage to spike (bugs might cause to ignore fixed voltages under some circumstances too).

So my SOC on "Auto" using EXPO-Tweaked and MC voltage of 1.37V is running at SOC 1.176V currently (set to AUTO, offset -0.05) in BIOS 1413.

Also there is just a BIOS released which seems to cap SOC voltage to 1.30V under any circumstances. So it might be time to revise the advise to set a fixed voltage very close to the absolute maximum.

But sure this is just an advise from my experience and my advise might be as good as yours and everyone is free to chose any settings but has to live with the consequences. Having said that the only safe advise until further notice might be to use Auto/JEDEC timing and not using EXPO.


No more VSOC offset in 1303, wonder why 😂



No more VSOC offset in 1303, wonder why 😂


Well. No 1303 for my X670-P and BIOS 1413 released yesterday does still have it and I am happily using it with -0.06V offset resulting in 1.17 to 1.20V being applied on my system. But if I were ASUS I would now be very careful and even remove options which can be misused and are not fool-proof. Unfortunately this will lead to some options or value ranges being removed which are required for some extreme configurations (at risk of damaging hardware).

Perhaps mainboard manufacturers should think about locking down the settings to allow only safe options and ranges until a disclaimer has been signed. Still allowing people who know what they do or willing to risk damages can make full use of the potential even if it's unsafe.

Soon we will have ComboAM5Pi 1.0.0.7 BIOS dropping in with additional fixes.

I had this issue the first day I got my motherboard, X670E Gene and 7950x3D, picked up the motherboard last weekend, BIOS 1202.

After dialling in manual ram timings and voltages, never experienced it again.

This is the protection mechanism introduced for x3d when cpu temp over 95c. Was the processor running very hot ?