-
-
Notifications
You must be signed in to change notification settings - Fork 17
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
1.2.5 (nor previous versions) load past "initializing" splash screen. AGESA 1.2.0.0+ #24
Comments
Thanks for reporting. Will have to compile a debug version that outputs everything in a file, so we can see where it stops first. |
I'll be happy to pilot that and send you the log(s) |
Attached version has trace enabled for the dll. |
Wow, hi Ivan!
First of all I just want to say I think it's so cool you took the time to
compile this and send it to me for testing.
Unfortunately I can't help you now, because the program ran fine upon start
and then I tried the normal release of ZenTimings 1.2.5 and that, too, ran
fine.
Since my original issue, for different reasons, I reinstalled the latest
AMD chipset driver and did a clean reinstall of my graphics drivers.
Probably the former fixed it. Do you still want the log output?
Pete
…On Thu, Jul 8, 2021 at 3:41 PM Ivan Rusanov ***@***.***> wrote:
Attached version has trace enabled for the dll.
It will be logged to trace.log
ZenTimings_v1.2.5-trace-debug.zip
<https://github.com/irusanov/ZenTimings/files/6786857/ZenTimings_v1.2.5-trace-debug.zip>
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#24 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AT6I5FRPFKZYCW6XTL54NKLTWX5PBANCNFSM47OWFQFA>
.
|
Just want to add, that before when it was crashing out, I believe I tried reinstalling chipset drivers and it had no effect. This time, though, there was a newer version on AMD's site. Updated chipset driver may solve? |
Who knows. Haven't installed chipset drivers from a long time on my machine, but sometimes OS updates cause issues, drivers probably too. |
I tried running Zen Timing for the first time and also had the CPU not supported error. I ran this debug version and here is the trace log:
My system is a Dell G3 3590 with an i7-9750H cpu. Thanks! |
Its a program for ryzen cpus not intel stuff. |
Hi, I have the same problem with the launch of Zentimings v1.2.5, like #24 Lufusol. Configuration: Launch log:
|
Hello, Looks like same problem is present for ASUS G713RM laptop with AMD Ryzen 7 6800H running on Windows 11: Log:
Thank you for your work. |
The 6800H is not (fully) supported in 1.2.5. If you feel adventurous, attached is a debug version of 1.2.6 which is not really finished yet, but might work on Rembrandt. Unfortunately I don't have a Rembrandt or Renoir. |
For 1.2.6 version result is the same:
It would be nice to have an option to lunch debug version with traces enabled e.g. by adding parameter to the shortcut like:
Or any other way to easy enabling of trace option for cases when ZenTimings couldn't lunch GUI properly without compiling separate version. |
The plan is to add automatic debug/trace log in case of a crash. |
I have recompiled the DLL, extract it from the archive and replace in ZenTimings folder. Should have the detection fixed. Hopefully the rest works then. |
New message appears:
|
Yeah, sorry. I did that at work, but the source was outdated. |
Yep, but at least I know that my CPU should qualifeid for it :)
And HWinfo report: |
No chanegs with this revision:
|
It seems the CPUID is not a problem anymore, but something else fails. I don't have a proper debug logging in the app. ZenTimings_v1.2.6.619-debug.zip Here's a build of the new ZenStates which only loads the DLL and has no UI elements yet. If that loads then the DLL is at least half-workling. Thanks for the testing, if none of the above has any positive effect I will have to implement that debug report functionality and then try again. |
No changes here.
Looks like it works:
No problem. You are welcome. |
Can you copy the ZenStates-Core.dll from the ZenStates zip and replace in the ZenTimings directory? |
Coping ZenStates-Core.dll from ZenStates-2.0-20220513.zip to ZenTimings_v1.2.6.619-debug.zip leads to:
|
Thanks for trying, something is not initialized correctly in the app and I'm not handling the error correctly. The DLL seems to be working. Last try for now, added some more checks in the code |
Result is the same:
|
@lHolyDiverl Sorry to bother you again, but do you remember what is the last message displayed in the loading screen? |
Hi @irusanov,
Same as here:
Or maybe you are reffering to some other version? |
Alright, that would probably help in tackling where it fails. Thanks! If you're willing to test, I might temporary disable that part when I get back home from work and see how it goes. Btw, you might want to try the DRAM Calculator, which has timings only and no other complications, so it might start without an error. PS: I think the problem is that it detects 2 DIMM modules (or 1 if your laptop has just one DIMM) and twice the channels and that's where it fails. |
Yes, it only reads common timings. I just need to fix the channel detection and ZenTimings will start, too. We can try with disabled first to verify, then will try to fix the initialization. |
Attached are 2 debug versions of the app. "1" has the channels info disabled and "2" has the whole memory modules info disabled. Let me know if any of these launch successfully. ZenTimings_v1.2.6.622-debug-1.zip PS: If it works, I'll be glad if you can attach a debug report. |
First one ZenTimings_v1.2.6.622-debug-1.zip is working from the box. Here is debug report for you: |
@lHolyDiverl Thank you very much! I have the base now to get the full support. |
@irusanov You are welcome. Happy to help :) |
Oh and one small addition. When ZenTimings_v1.2.6.622-debug-1.zip starts it hang for a while on: With following warning:
Than there
This is a test build but still maybe there are some other hidden issues since some power data table is present in debug report. Here is application screenshot: |
The power data in the debug log seems to be gibberish, I will have a new build at the end of the day which would hopefully fix this problem. |
Hopefully this one works a little better. |
Both versions ZenTimings_v1.2.6.622-debug-1.zip and ZenTimings_v1.2.6.623-debug.zip are running fine when NOT in advanced\debug mode except 623 see the amount of available RAM and stick to choose at the bottom: If debug mode is lunched both versions ZenTimings_v1.2.6.622-debug-1.zip and ZenTimings_v1.2.6.623-debug.zip are stuck at With warning:
and exception for Memory controller:
Screenshots for advanced mode to compare: And some debug report from ZenTimings_v1.2.6.623-debug.zip: |
Well, it's a progress. Not sure why the power table does not work. |
SMUDebugTool_v1.3.3_20220518.zip Here's the debug tool. Close all other applications, unzip the tool and run the exe. You should see something like this, if it starts correctly.Select SMU tab, then with the RSMU selected from the dropdown, click on "Send" to test for correct response. Command ID = 1 is a test command, result should be the value in Arguments field + 1.Now click "Reset". For each item in the Mailbox dropdown, click "Monitor" - it will open a new window. Leave them open.Launch HWinfo with monitoring enabled (as in the screenshot). After a while, close hwinfo and make a screenshot of the windows, like so.Please don't enter random commands and arguments and click Send - this could be dangerous. Scan button usually works, but it might also lock the system, requiring a hard reset, so it's also not advisable. PMTable button is safe, but it will probably display incorrect values in the table (like in the debug report). Thanks for the cooperation, hopefully this will help! |
ZenStates-Core.dll.zip |
Same result with this DLL as it was with original ZenTimings_v1.2.6.623-debug.zip for both Advanced Mode turned off and on |
I tried SMUDebugTool_v1.3.3_20220518.zip: SMU tab: Argument 0: Argument 1: Monitor part I have basic configuration for HwInfo and run it as (After starting monitor): Results are: P.S. I had Chrome running in background at this run. I believe it shoudn't affect the results. |
Thanks. Can you do the same Monitor, but with the other Mailbox opened as a SMU Monitor window? PS: Another request. Can you scroll to the very top of the SMU Monitor windows before taking the screenshots? |
@lHolyDiverl Thank you in advance! |
@irusanov |
@lHolyDiverl I still don't get it why it doesn't work :/ Here's a test dll with the address hardcoded (assuming it is shown on the screenshot): |
Result are pretty the same. ZenTimings_v1.2.6.623-debug.zip with ZenStates-Core-test-rembrandt.zip provide following result: With warning:
And exception:
When And screenshot of the app: Maybe you would need another configuration for testing to determine what is wrong. |
The error you get is because AMD seems to have removed the command to get memory controller config :/ One more build, but I expect similar results, although a different last error. If that doesn't work, then I give up for now. Won't be able to fix it unless I have access to such a system. At least it shows main timings and as a last resort I can turn off AdvancedMode for Rembrandt until I figure out what's wrong. |
ZenTimings_v1.2.6.633-debug.zip running fine with Advanced Mode turned off: And same for Advanced Mode turned on:
With warining:
With app window: As for Rembrandt you can also add a warning message that "Advanced Mode for Rembrandt is not available yet" or somethin like it and maybe dissable Advanced Mode option in settings. People could also modified settings.xml but it will be their own choice. Good luck in your "battle" against Rembrandt :) |
CPU Ry5 3600
Mobo AsRock B550M SL
BIOS 1.9, (AGESA 1.2.0.0) 2.0, 2.1 (AGESA 1.2.0.2)
OS Win 10 Pro x64 21H1
RAM OLOy WarHawks DDR4-3600 2x8GB (b-die)
Same issue as some others, only the update prompt on first run and "initializing" splash screen on subsequent runs are shown, program exits before ever making it to the GUI. :( Previous versions back to 1.2.2 more or less the same (1.2.2 hangs on "initializing" for longer)
The text was updated successfully, but these errors were encountered: