VOGONS


HWiNFO support of vintage hardware

Topic actions

Reply 382 of 393, by repaxan

User metadata
Rank Newbie
Rank
Newbie

On my ECS NForce4-A754, the NB chipset fan is incorrectly reported as the CPU fan.

Attachments

  • Filename
    NFORCE4-A754.rar
    File size
    43.08 KiB
    Downloads
    5 downloads
    File license
    Public domain
  • 20200613_123131.jpg
    Filename
    20200613_123131.jpg
    File size
    24.66 KiB
    Views
    251 views
    File license
    Public domain

Reply 384 of 393, by repaxan

User metadata
Rank Newbie
Rank
Newbie

Asus P6T Deluxe with wrong voltage reporting.

I also noticed that when minimized and re-maximized sometimes this graphical glitch would happen.

Attachments

  • hwinfo.png
    Filename
    hwinfo.png
    File size
    187.22 KiB
    Views
    205 views
    File license
    Public domain
  • Filename
    p6tdeluxe.rar
    File size
    128.84 KiB
    Downloads
    3 downloads
    File license
    Public domain

Reply 385 of 393, by Mumak

User metadata
Rank Member
Rank
Member
repaxan wrote on 2020-06-19, 04:20:

Asus P6T Deluxe with wrong voltage reporting.

I also noticed that when minimized and re-maximized sometimes this graphical glitch would happen.

Thanks, voltage values will be fixed in the next build.
But I have no idea why that glitch happens, haven't seen that yet.

Reply 386 of 393, by vorob

User metadata
Rank Member
Rank
Member

Hi, I'm again with my laptop Fujitsu Siemens Amilo Xi 1546. This time it's not about HWINFO (though it still reads weird 97C temperature from gpu sensor that doesn't exist)... I'm here to ask maybe you know how to solve my problem with faking temperature sensor, making it think that machine is over 90C? With that info computer force fan to ran on maximum speed, which is essential to save GPU from dying. I know that this thread is not about such weird stuff, but maybe you deep knowledge will guide me?

I've explained this in my own thread [REQ] Control laptop's fan via fake cpu temperature

ps. for some reason HWINFO doesn't read this sensor which is fooled via NHC app, but HWMonitor does see it.

Reply 387 of 393, by Mumak

User metadata
Rank Member
Rank
Member
vorob wrote on 2020-06-22, 20:20:

Hi, I'm again with my laptop Fujitsu Siemens Amilo Xi 1546. This time it's not about HWINFO (though it still reads weird 97C temperature from gpu sensor that doesn't exist)... I'm here to ask maybe you know how to solve my problem with faking temperature sensor, making it think that machine is over 90C? With that info computer force fan to ran on maximum speed, which is essential to save GPU from dying. I know that this thread is not about such weird stuff, but maybe you deep knowledge will guide me?

I've explained this in my own thread [REQ] Control laptop's fan via fake cpu temperature

ps. for some reason HWINFO doesn't read this sensor which is fooled via NHC app, but HWMonitor does see it.

I assume you refer to the THRM sensor shown by HWMonitor. It's quite a simple fake via ACPI - the tool creates an ACPI Thermal Zone object.

Reply 388 of 393, by vorob

User metadata
Rank Member
Rank
Member
Mumak wrote on 2020-06-22, 20:44:

I assume you refer to the THRM sensor shown by HWMonitor. It's quite a simple fake via ACPI - the tool creates an ACPI Thermal Zone object.

Yep, that's it. Any easy way to fake its values? Other then this horrible ancient app I'm using?

Reply 389 of 393, by Mumak

User metadata
Rank Member
Rank
Member
vorob wrote on 2020-06-23, 11:10:
Mumak wrote on 2020-06-22, 20:44:

I assume you refer to the THRM sensor shown by HWMonitor. It's quite a simple fake via ACPI - the tool creates an ACPI Thermal Zone object.

Yep, that's it. Any easy way to fake its values? Other then this horrible ancient app I'm using?

I'm not aware of.. perhaps some other tool that can inject custom ACPI methods/tables, but I have never used one so can't recommend..

Reply 390 of 393, by dottoss

User metadata
Rank Member
Rank
Member

Hi Mumak, on my LGA 775 build with an q6600 there are some core temperature discrepancy between aida64, speedfan and HWiNFO. The two other is consistent while HWiNFO is reporting higher values. Attached you will find the debug file and below some screenshots.

Capture.PNG
Filename
Capture.PNG
File size
244.72 KiB
Views
103 views
File license
Public domain

Attachments

  • Filename
    HWiNFO64.zip
    File size
    45.21 KiB
    Downloads
    4 downloads
    File license
    Public domain

Reply 391 of 393, by Mumak

User metadata
Rank Member
Rank
Member
dottoss wrote on 2020-06-28, 10:43:

Hi Mumak, on my LGA 775 build with an q6600 there are some core temperature discrepancy between aida64, speedfan and HWiNFO. The two other is consistent while HWiNFO is reporting higher values. Attached you will find the debug file and below some screenshots.

Capture.PNG

This is most probably due to a different Tj,max assumption in different tools, but none of them knows this for sure. It's not possible to detect the correct value on older CPUs.
HWiNFO believes Tj,max=100 C for this CPU (which should be correct according to the specifications), others most probably think it's 90 C.
If you're sure it's not correct, you can change this by right-clicking on the core temperature in HWiNFO -> Adjust Tj,max.

Reply 392 of 393, by red-ray

User metadata
Rank Oldbie
Rank
Oldbie
Mumak wrote on 2020-06-28, 13:21:

HWiNFO believes Tj,max=100 C for this CPU (which should be correct according to the specifications), others most probably think it's 90 C.

AFAIK TjMax depends on the stepping and [B3] is 90°C whereas [G0] is 100°C. Given this I expect Speedfan + AIDA64 did not get updated for the [G0] stepping.

file.php?id=86950

Attachments

  • Kentsfield.png
    Filename
    Kentsfield.png
    File size
    108.24 KiB
    Views
    88 views
    File license
    Public domain

Reply 393 of 393, by dottoss

User metadata
Rank Member
Rank
Member
Mumak wrote on 2020-06-28, 13:21:
This is most probably due to a different Tj,max assumption in different tools, but none of them knows this for sure. It's not po […]
Show full quote
dottoss wrote on 2020-06-28, 10:43:

Hi Mumak, on my LGA 775 build with an q6600 there are some core temperature discrepancy between aida64, speedfan and HWiNFO. The two other is consistent while HWiNFO is reporting higher values. Attached you will find the debug file and below some screenshots.

Capture.PNG

This is most probably due to a different Tj,max assumption in different tools, but none of them knows this for sure. It's not possible to detect the correct value on older CPUs.
HWiNFO believes Tj,max=100 C for this CPU (which should be correct according to the specifications), others most probably think it's 90 C.
If you're sure it's not correct, you can change this by right-clicking on the core temperature in HWiNFO -> Adjust Tj,max.

Cool, then HWiNFO is correct, which is good. That's the one more important for me 😀

Thanks