10-04-2022 06:30 AM - edited 10-18-2023 07:20 AM
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)
DOCP/EXPO Tweaked: (for simple timings tightening)
strixe-e 1515
for crosshair and strix e-e:
explanation of segment2 Loadline:
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
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.
Solved! Go to Solution.
04-17-2023 11:37 AM - edited 04-17-2023 11:38 AM
Had to go back to 1004. I was getting too many hard lock ups (black screen, fans ramp to 100, and can't reset unless I turn off the PSU) with the 1101 bios. Also, under 1101 I was getting more frequent USB, NIC, and Audio dropouts and straight up disappearances. At least with the 1004 I can turn off the Thunderbolt/USB4 and cut down on those dropouts. They don't go away but they become MUCH less frequent. With 1101 turning on or turning off Thunderbolt/USB4 seemed to make no difference. I just dropped out a LOT more.
04-17-2023 12:24 PM
18:40 mark.
"My biggest fear, actually, is the Asus motherboard. There have been SOOOO many reports of Crosshair Hero and Crosshair Extreme motherboards just failing. Bad motherboard traces. Apparently bad memory traces have been discovered. Which is kinda, I think, on the Extreme specifically. Which I think is exactly what happened to my Asus Hero. Ummm. Its been frustrating. And considering the fact that the problems didn't show themselves until a week later has me very scared for. Am I gonna last a month? Two months? Six months? If it starts to become problematic then what I'll more than likely do is keep the 7950x3d and switch to another brand."
Asus?
04-17-2023 04:26 PM
Still having USB PCIe expansion card initialization issues. Coin flip if it's going to show up in Windows when I get logged in. Manually setting PCIe generation per slot doesn't help.
04-18-2023 07:46 AM
do the core flex algos have any impact on chips w a single CCX? (7700,78000x3D,etc)?
04-19-2023 04:55 AM
Maybe @Shamino could share if all these stability problems come from AGESA or just ASUS BIOS? It's unaceptable that these problems are still not resolved after all these months since release. I do get that ASUS BIOS dev team might be understaffed or some other corporate bs, not blaming the devs, but this is beyond ridiculous.
04-19-2023 11:07 AM
i had to go back to 0805. 1101 had better boottimes with MCR / PDE on enabled. but this setting causes errors on Karhu with Cache enabled bettween 400 and 1000% ( i canceld after 1000%). QVL Ram G.SKill 6000C36 on D.O.C.P II
Same settings with 0805 - no errors. I have Cosshair Extreme. This BIOS thing is unbelievable - i built a PC for X-Thousand of Euros and i have to deal with Bluescreens, Errors, Bootloops, E00, and so on
04-19-2023 11:10 AM
1101 also bootet often my 4090 to PCIE-SPeed 3.0 or it got no sensor Data so the PCIE-ran on 16x1.1 ......... with 0805 no problems.
Do we know when new Agesa 1.0.07 will come?
04-19-2023 01:11 PM
That is not a bug, it is a feature, NVIDIA runs at PCE4 1.1 on IDLE but as soon as you kick some GPU workload will revert back to PCI full 4.0, check below GPUZ on my 4090 IDLE vs GPU worload (game in the back running)
IDLE
SOMETIMES IDLE with a bit of browser is even switched to PCIE 4 2.0
Justo to conserve power i guess makes sense.
04-19-2023 01:18 PM
i
i know about the "?" and that pcie goes to 1.1 in idle and 4.0 under load.
My prob looks like this (picture) and when i do a TimeSpy i have like 5fps because the PCIE Speed stays at 16x1.1
04-19-2023 01:27 PM - edited 04-19-2023 01:40 PM
Ok trying to help , since you are aware and already excluded this scenario, just a few questions to try and maybe find the root cause:
1 - Did you update the chipset drivers with latest?
2 - Are GPU drivers well installed (maybe uninstall , full reboot, then reinstall, full reboot again before any test)
3- Did you tried to actually kick in other software , games etc, to check is the same scenario?
4- If using windows 11 did you set "Hardware-accelerated GPU scheduling" On? should be!
5 - In Bios is REBAR on? should be!
6 - If nothing of this works (or already tried all) - suggest boot to bios , load all defaults, set just your EXPO profile , set CPU VSOC to 1.25v nothing more! and check again!
7 - All failing either revert to previous BIOS working or reflash BIOS with full reset settings!
Hope it helps! If not well ... tried---
Clarification, i am running bios 1004, not latest 1101 or whatever is the latest now , because i also had a lot of issues lately with BIOS updates , se my posts before!
AMD/ASUS X670 boards have been a hit and miss lately full of issues , don't know who is in fault here and don't care ASUS is the vendor needs to step up and solve all this! Otherwise my only final advice is just keep you most stable BIOS until good news starts to happen!