Anomalous sensor readings in HWiNFO64

lesovers

Member
Asus Prime X470 Pro Bios update 5220 (AGESA 1.0.0.3 ABBA dated 24/9/2019) has the same issue.
Chipset Temperature, Tsensor 1 Temperature and CPU OPT fan RPM are all incorrect;


HWiNF064 v6.12-3930 Asus X470 Bios 5220 Bugs.jpg

Les
 

lesovers

Member
Here is a shot of the bios monitoring screen. CPU temperature should be at the top however the screen was refreshing when I took the photo;

IMG_0392.JPG
 

Martin

HWiNFO Author
Staff member
The BIOS uses a different access method, which applications cannot use.
This problem is BIOS related, so it looks like ASUS is going to fix this in another BIOS update.
 

KotOr

Member
Hello. i want report maybe bad reading of CPU PLL voltage. HWinfo shows ( CPU PLLs OC ) 1.2-.1216V in BIOS i have Two PLL voltage 1. PLL termination voltage (if i change it nothing happens) 2. Eventual PLL termination voltage. if i change it to 1.1V HWinfo shows still 1.216 but on my motherboard Maximus Z390 APEX XI i have direct Voltage measurment points and on default they read 1.25V and if i change 2. PLL to 1.1V they shows around 1.1V. So that ( PLLs OC ) is something different or maybe just not accurate readings?
 

Attachments

  • HWiNFO64.DBG
    1.2 MB · Views: 1

Martin

HWiNFO Author
Staff member
Hello. i want report maybe bad reading of CPU PLL voltage. HWinfo shows ( CPU PLLs OC ) 1.2-.1216V in BIOS i have Two PLL voltage 1. PLL termination voltage (if i change it nothing happens) 2. Eventual PLL termination voltage. if i change it to 1.1V HWinfo shows still 1.216 but on my motherboard Maximus Z390 APEX XI i have direct Voltage measurment points and on default they read 1.25V and if i change 2. PLL to 1.1V they shows around 1.1V. So that ( PLLs OC ) is something different or maybe just not accurate readings?
It might be something different. Does the BIOS under monitored values show a PLL value that changes accordingly?
 

dokbenway

Member
New BIOS out today:

PRIME X470-PRO BIOS 5406
Update AM4 combo PI 1.0.0.4 patch B


Unfortunately it doesn't seem to fix the incorrect temp and CPU OPT fan readings.

Maybe next time...
 

KotOr

Member
Hello. i'm having weird situation here. i was using HWi 6.12 than update to 6.14 and i'm sure i chose disable monitoring Asus EC as recommended but still i have VRM and PCH temps but now after reinstall OS 6.14 doesnt show this temps. Even if i manualy enable monitoring Asus EC i only get PCH temps. VRM is disable so i try manualy enable them but HW info crash everytime. so i try download 6.12 enable Asus EC and try 6.15. and i read on difrent topic that you don´t recommend monitoring this if Asus AI suite is installed (i dont have it) so shoudl i disable it or use 6.12 version. now on 6.15 look like its working
 

Martin

HWiNFO Author
Staff member
There should be no issues monitoring the VRM sensor via ASUS EC. I always recommend to use the latest version.
 

A.M.

Member
Hi, guys!

What ASUS e-mail/whatever do you recommend to write to? I've purchased the board and faced the same problem (expectedly).
 

KotOr

Member
Hello. i notice on Maximus IX Apex Z270 that my CPU_OPT Fan showing weird RPM i tried latest full version HWinfo and beta and same. So its HWinfo problem or motherboard/fan ? That fan can go max 1500RPM. Thank you
 

Attachments

  • Maximus IX Fan.PNG
    Maximus IX Fan.PNG
    13.7 KB · Views: 9

Martin

HWiNFO Author
Staff member
Looks like some glitch during EC access/read. Are you perhaps running any other monitoring tool along with HWiNFO?
 

A.M.

Member
Hi, Martin.

I have ASUS Prime X470-Pro and faced this problem. Done some investigations and found some stranges.

Davidinlv82 has the same board, BIOS 5216, earlier supplied dbg file with
Code:
ASUS WMI GETINFO[00] = 0, 0, 1, 3,  = CPU Core Voltage
ASUS WMI GETINFO[01] = 0, 3, 1, 3,  = +12V Voltage
ASUS WMI GETINFO[02] = 0, 3, 1, 3,  = +5V Voltage
ASUS WMI GETINFO[03] = 0, 3, 1, 3,  = 3VSB Voltage
ASUS WMI GETINFO[04] = 1, 0, 1, 0,  = CPU Temperature
ASUS WMI GETINFO[05] = 1, 3, 1, 0,  = Motherboard Temperature
ASUS WMI GETINFO[06] = 2, 3, 1, 1,  = CPU Fan
ASUS WMI GETINFO[07] = 2, 3, 1, 1,  = Chassis Fan 1
ASUS WMI GETINFO[08] = 2, 3, 1, 1,  = Chassis Fan 2
ASUS WMI GETINFO[09] = 2, 3, 1, 1,  = Chassis Fan 3
ASUS WMI GETINFO[10] = 2, 3, 1, 1,  = AIO Pump
ASUS WMI GETINFO[11] = 2, 3, 1, 1,  = Water Pump
ASUS WMI GETINFO[12] = 1, 4, 2, 0,  = Chipset Temperature
ASUS WMI GETINFO[13] = 1, 5, 2, 0,  = Tsensor 1 Temperature
ASUS WMI GETINFO[14] = 2, 3, 2, 1,  = CPU OPT
and
Code:
ASUS WMI BUF:
      00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F   0123456789ABCDEF
------------------------------------------------------------------------
0000: 41 53 55 53 53 45 4E 53 4F 52 30 31 94 74 15 00   ASUSSENSOR01”tя
0010: C8 9E B6 00 9A 42 4B 00 C0 8F 32 00 2C 00 00 00   Ић¶яљBKяАЏ2я,яяя
0020: 1C 00 00 00 39 04 00 00 94 11 00 00 AE 05 00 00   яяя9яя”яя®яя
0030: 48 07 00 00 99 14 00 00 00 00 00 00 52 00 5F 00   Hяя™яяяяяяRя_я
0040: 53 00 45 00 54 00 54 00                           SяEяTяTя

As you can see, erroneous values are the last 12 bytes (3 groups of 4 bytes) whose give those large numbers.

Later lesovers wrote that BIOS 5220 has the same problem with the same values (the only difference is Celsius vs Fahrenheit).

But I do have BIOS 5220 too and my dbg file shows last 12 bytes as zeros.
Code:
ASUS WMI BUF:
      00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F   0123456789ABCDEF
------------------------------------------------------------------------
0000: 41 53 55 53 53 45 4E 53 4F 52 30 31 94 74 15 00   ASUSSENSOR01”tя
0010: 30 9D B9 00 2A 96 4E 00 10 3A 33 00 3C 00 00 00   0ќ№я*–Nя:3я<яяя
0020: 24 00 00 00 B0 02 00 00 F2 02 00 00 D8 02 00 00   $яяя°яятяяШяя
0030: B0 02 00 00 00 00 00 00 00 00 00 00 00 00 00 00   °яяяяяяяяяяяяяя
0040: 00 00 00 00 00 00 00 00                           яяяяяяяя

Anyway my zeros and those millions are all wrong.

But the main thing I want to ask: as you can see ASUS WMI reports SIO source for the first 12 values (source = 1, 3rd digit in GETINFO), and EC source for the last 3 (source = 2).
I found linux driver project for ASUS WMI and as I could understand (I may be wrong) from its source code, it compares source of particular sensor and source of buffer and decodes value if the sources are equal. I mean maybe there is another ASUS WMI buffer for EC source values? Which contains possibly trash for the first 12 sensors but does have proper values for the last three.
 

Martin

HWiNFO Author
Staff member
No, there's no other buffer. The BIOS provides only one and it's solely responsible for filling it with proper values.
I'm not familiar with the Linux project, all I know is that it was me with elmor (when he worked at ASUS) who designed this interface to mitigate an issue with concurrent access to the SIO chip, which could result in corruption of the SIO registers. And it worked well at the beginning, but later BIOS updates brought bugs here.
 
Top