General Comments and Z97 Board Specific

[parsec]

Hi Martin, I haven't posted in the forum for a while, I hope you are well. If you remember me, we worked out a few details for reporting a couple things in HWiNFO64 when used with my ASRock Z87 Extreme 6 board. I have an ASRock Z97 Extreme 6 board now, and a few things have changed relative to the Z87 board. I also have some basic questions.

General comments: I really appreciate the display of UEFI booting status (which has changed over time) since I use UEFI booting Windows 8 installations. It is accurate, when I change from UEFI to Legacy booting, the display in System Summary becomes gray rather than green.

Currently (v4.40-2240) there is a green "UEFI" display next to the "BIOS Version" field, which I assume just means UEFI firmware was detected, correct?

The green "UEFI Boot" next to the "Operating System" heading indicates UEFI booting, which will appear gray when Legacy booting, correct?

I noticed that msinfo32 does not correctly identify UEFI booting. Given a UEFI booting Windows 8 installation, if I configure my board's UEFI/BIOS for Legacy booting (CSM set to Enabled), msinfo32 still shows BIOS Mode as UEFI, which it does when CSM is disabled. HWiNFO64 is able to detect a Legacy or UEFI boot correctly. Apparently msinfo32 is simply checking the partitions on the boot device to identify UEFI booting, which is not enough. I'm not surprised that you are doing it correctly.

I've been using the portable version of HWiNFO64 for a while now, and don't have the persistent driver installed on at least one of my PCs. What is the value of using the driver and installed version versus the portable version?

I've noticed in the forum users mentioning a fan speed control option in HWiNFO. Not that I really need it, but I can't find it, where is it?

Regarding the ASRock Z97 Extreme 6 board and HWiNFO64: This board uses the Nuvoton NCT6791D sensor chip, according to HWiNFO64. I noticed a previous version of HWiNFO64 reported the Vccin (CPU Input Voltage) as 1/2 the actual value, you've fixed that at least one version ago, thanks, IMO the value displayed is correct.

My ASRock Z87 Extreme 6 board's sensor provided the Z87 PCH temperature (via Auxiliary in HWiNFO64), but I don't get an Auxiliary value with the Z97 board, or anything in its place that could be the Z97 temperature.

FYI, various voltages detected by HWiNFO64 are valid values and map to these voltages:

VTT and VIN5 both are the CPU Digital IO Voltage. (If I change this value in the UEFI or Windows utility, I see the change in HWiNFO in both values.)

VIN11 is the CPU Analog IO Voltage. (Verified the same way as above.)

VIN13 is the System Agent Voltage. (Verified as above.)

VIN6 is unknown or bad data, varies between ~0.008V and ~0.840V.

VIN12 is unknown or bad data, it changes between 0.000 - 1.000V.

VIN14 is unknown or bad data, a constant 0.216V.

I appreciate the Total DRAM Power readings, which change with CPU usage, OC, etc. I noticed a strange thing about it. As I over clocked my memory from 1600, to 1866, and to 2000, I saw an increase in DRAM power, which makes sense IMO. When I OC'd to 2133, the DRAM power dropped to a much lower value, even lower than at 1600. The memory voltage was the same for all these speeds. I know you would need to see the log data to possibly explain this, which I have not saved yet.

I like the network chip detection and download/upload speeds. Two boards I have with dual NIC chips (one two Intel chips, the other one Intel and one Realtek chip) are detected and displayed fine, even when teamed.

I had the chance to test a Plextor M6e SSD, which is an M.2 interface (PCIe) SSD, and comes with a PCIe adapter card for use in a PCIe slot. I used it both ways on the Z97 board, and HWiNFO64 detected and displayed its information fine.

That's more than enough, as usual HWiNFO64 is the best, thanks as always!

 

[Martin]

Hi Martin, I haven't posted in the forum for a while, I hope you are well. If you remember me, we worked out a few details for reporting a couple things in HWiNFO64 when used with my ASRock Z87 Extreme 6 board. I have an ASRock Z97 Extreme 6 board now, and a few things have changed relative to the Z87 board. I also have some basic questions.

General comments: I really appreciate the display of UEFI booting status (which has changed over time) since I use UEFI booting Windows 8 installations. It is accurate, when I change from UEFI to Legacy booting, the display in System Summary becomes gray rather than green.

Currently (v4.40-2240) there is a green "UEFI" display next to the "BIOS Version" field, which I assume just means UEFI firmware was detected, correct?

Yes, that means the BIOS should be UEFI capable.

The green "UEFI Boot" next to the "Operating System" heading indicates UEFI booting, which will appear gray when Legacy booting, correct?

Correct - means that the OS has booted in UEFI mode.

I noticed that msinfo32 does not correctly identify UEFI booting. Given a UEFI booting Windows 8 installation, if I configure my board's UEFI/BIOS for Legacy booting (CSM set to Enabled), msinfo32 still shows BIOS Mode as UEFI, which it does when CSM is disabled. HWiNFO64 is able to detect a Legacy or UEFI boot correctly. Apparently msinfo32 is simply checking the partitions on the boot device to identify UEFI booting, which is not enough. I'm not surprised that you are doing it correctly.

I've been using the portable version of HWiNFO64 for a while now, and don't have the persistent driver installed on at least one of my PCs. What is the value of using the driver and installed version versus the portable version?

When Persistent driver is enabled, then after first run of HWiNFO the driver remains installed in system and is not uninstalled after closing HWiNFO. That might speed up next starts. Also it might allow to run HWiNFO without Administrator rights, though I haven't tested this option extensively and it might not work properly in some cases.

I've noticed in the forum users mentioning a fan speed control option in HWiNFO. Not that I really need it, but I can't find it, where is it?

This option is available only on very few specific machines. Almost all DELL Alienware, some HP notebooks (like 8740W) and many DELL systems are supported. This option is not widely published, since it was never my intention to implement such features in HWiNFO for broad use. I did this only per request of some friends for machines, where no other tool was able to accomplish this.

Regarding the ASRock Z97 Extreme 6 board and HWiNFO64: This board uses the Nuvoton NCT6791D sensor chip, according to HWiNFO64. I noticed a previous version of HWiNFO64 reported the Vccin (CPU Input Voltage) as 1/2 the actual value, you've fixed that at least one version ago, thanks, IMO the value displayed is correct.

My ASRock Z87 Extreme 6 board's sensor provided the Z87 PCH temperature (via Auxiliary in HWiNFO64), but I don't get an Auxiliary value with the Z97 board, or anything in its place that could be the Z97 temperature.

FYI, various voltages detected by HWiNFO64 are valid values and map to these voltages:

VTT and VIN5 both are the CPU Digital IO Voltage. (If I change this value in the UEFI or Windows utility, I see the change in HWiNFO in both values.)

VIN11 is the CPU Analog IO Voltage. (Verified the same way as above.)

VIN13 is the System Agent Voltage. (Verified as above.)

VIN6 is unknown or bad data, varies between ~0.008V and ~0.840V.

VIN12 is unknown or bad data, it changes between 0.000 - 1.000V.

VIN14 is unknown or bad data, a constant 0.216V.

Are you sure the Auxiliary value was really PCH? AFAIK, neither ASRock tools, nor the BIOS reports such temperature.
Thanks for the voltage additions, I'll update this in the next build. I assume this is for the Z97 series, correct ?
Is VIN6/12 maybe Vcore ? Some boards (apparently capable of measuring Vcore on Haswell systems) report very low Vcore values when a particular core is in deep C-state.

I appreciate the Total DRAM Power readings, which change with CPU usage, OC, etc. I noticed a strange thing about it. As I over clocked my memory from 1600, to 1866, and to 2000, I saw an increase in DRAM power, which makes sense IMO. When I OC'd to 2133, the DRAM power dropped to a much lower value, even lower than at 1600. The memory voltage was the same for all these speeds. I know you would need to see the log data to possibly explain this, which I have not saved yet.

I don't think the log data would help. I have no explanation for this, could be a glitch in the DRAM power measurement logic in CPU...

I like the network chip detection and download/upload speeds. Two boards I have with dual NIC chips (one two Intel chips, the other one Intel and one Realtek chip) are detected and displayed fine, even when teamed.

I had the chance to test a Plextor M6e SSD, which is an M.2 interface (PCIe) SSD, and comes with a PCIe adapter card for use in a PCIe slot. I used it both ways on the Z97 board, and HWiNFO64 detected and displayed its information fine.

That's more than enough, as usual HWiNFO64 is the best, thanks as always!

 

[parsec]

Thanks for the answers Martin. Yes, I was referring to the ASRock Z97 Extreme 6 board throughout my post, except where I specifically said otherwise.

So the persistent driver has no advantage besides a faster start of HWiNFO? I always use an admin account, so I wouldn't notice any problems with a standard account.

You are correct about ASRock not displaying the PCH temperature in the UEFI or in their utility programs. The auxiliary temperature reading from HWiNFO64 on my ASRock Z87 board seemed to be the PCH temperature, I forget now what else convinced me that it was. I get a similar temperature reading via HWiNFO64 on my ASRock Z77 board, again the auxiliary reading, although the temperatures are different on the two different chipsets. The temperature readings would change depending upon case fan speeds, ambient temperature, etc. I was just surprised not to see an Auxiliary value present in HWiNFO64 when used with the Z97 board.

I forgot to mention that the Analog and Digital IO voltages are displayed in the UEFI and ASRock utility on the Z97 board, as well as the System Agent voltage, so I'm confident about the mapping of those voltages to the VINx readings.

Unfortunately, the ASRock Z97 UEFI and utility still only display VIDs. When using all C States including C7, the VIDs go down to ~0.640V in HWiNFO64.

I suspect that VIN6 might be the true VCore voltage, but I'm not sure. VIN6 does change with CPU load (as VIDs do), but if ASRock does not use this reading for VCore, how can I be sure it is valid? OTOH, would it surprise me if it was a true VCore voltage but is simply not used by ASRock? Not really... :dodgy:

I have a special program that can set the MSR registers for the C States in Windows, enabling or disabling them. With C States disabled, VIN6 is very slightly above the core VIDs. That would seem to indicate that it is not a true VCore IMO. I am using the Haswell Adaptive Voltage setting, which can do surprising things with the VIDs/Vcore, which makes me wonder if it might be correct. VIN6 goes as low as 0.008V in HWiNFO, which seems to low to be true. But, I get iGPU VID readings of 0.00000V at times (set to 5 decimal digits.) Perhaps some slight adjustments in HWiNFO are needed when interpreting these values?

VIN12 might also be a true VCore, it is very similar to VIN6 except just below the VIDs with C States disabled. I hope I'm not just dreaming about these readings, if you know they are just random or coincidental, I rather know the truth.

Yes, the DRAM power difference I saw that dropped at 2133 only made no sense to me. That does not bother me, and I can't get the memory any higher than 2133 to see if it changes again...

At least I can identify some of the VIN values for you, and I'll keep trying to figure out others. If you have questions let me know, thanks Martin!