problem detection temp crossfire sapphire 7950 dual x

vince360

Member
good day. as the title I wanted to report a bug that I've noticed on all versions of hwinfo64. in active practice with crossfire if I open hwinfo64, I'm busting the sensors of one of the vga that goes from 0 degrees to its real temp, so the fans start and stop continually resulting in increased temp. if I disable crossfire or a vga connection at a time no problem ... you can solve the problem with a future update? thank you very much

p.s. I guess the problem is the new vga bios that has a different and still not properly detected by hwinfo64 .. the version of the new card is: 11196-19-20g and the old is: 11196-16-20g
 
The problem is the AMD ULPS feature, which completely disables the secondary GPU if not used. In such disabled state it's not possible to query that GPU for any parameters, otherwise a total system crash can occur. So HWiNFO is behaving correctly. This has been discussed several times here. To mitigate this problem you can disable ULPS (using registry or the TRiXX tool).
 
Martin said:
The problem is the AMD ULPS feature, which completely disables the secondary GPU if not used. In such disabled state it's not possible to query that GPU for any parameters, otherwise a total system crash can occur. So HWiNFO is behaving correctly. This has been discussed several times here. To mitigate this problem you can disable ULPS (using registry or the TRiXX tool).

i have disabled it by trixx. the problem is not crossfire but is the new vga bios on the new card. i confused... if i connect only the new card have the same problem... can fix this please?
 
If it's a VBIOS problem, then I'm not sure if that can be fixed in an application.
Nevertheless, please attach the HWiNFO Debug File and I'll check it.
 
Martin said:
If it's a VBIOS problem, then I'm not sure if that can be fixed in an application.
Nevertheless, please attach the HWiNFO Debug File and I'll check it.

Thank you for your help. here the required files. I enclose also a picture where you see the problem clearly. thanks in advance
[attachment=775]
[attachment=776]
[attachment=777]
 

Attachments

  • hwinfolog.CSV
    17.8 KB · Views: 4
  • HWiNFO64.DBG
    842.2 KB · Views: 4
  • Immagine2.jpg
    Immagine2.jpg
    414.4 KB · Views: 6
Does it mean, that only when you have HWiNFO running, then all the other tools (TRiXX, MSI AB) sometimes report 0 C GPU temperature and if you close HWiNFO, they report the correct temperature always again ?
 
Martin said:
Does it mean, that only when you have HWiNFO running, then all the other tools (TRiXX, MSI AB) sometimes report 0 C GPU temperature and if you close HWiNFO, they report the correct temperature always again ?
yes if i start msi ab and/or trixx all work fine. when i start hwinfo msi and trix showing 0° on the primary card. if i close hwinfo msi and trixx show the correct temp...
 
Please try to disable those two GPU VRM CHL8228 sensors in HWiNFO (right-click on sensor and Disable Monitoring) and let me know if that helped.
 
Martin said:
Please try to disable those two GPU VRM CHL8228 sensors in HWiNFO (right-click on sensor and Disable Monitoring) and let me know if that helped.

i don't have chl8228 on my hw info but i try to disable the sensor on the screen and now all work fine. tnx for the help. if is possible fix in the new hwinfo update, if not, no problem ;) really tnx [attachment=782]
 

Attachments

  • Immagine.jpg
    Immagine.jpg
    387.3 KB · Views: 4
I think this is a conflict between the applications when they run concurrently.
My only advice is not to run them all at the same time.
 
if I disable that sensor as screen does not give me any problem.. in this case i can also keep them active both at the same time right?
 
Back
Top