Hi red-ray

Here's the result of your utility on my Rise iDragon CPU, under Win98SE.

The CPU has the Chris Norrie Easter Egg !

Vincent

Grosvince wrote on 2021-01-25, 22:31:

The CPU has the Chris Norrie Easter Egg !

Hi Vincent, Thank you for checking out SIV and I thought the Easter Egg would be there. In general SIV did OK, SIV V5.56 Beta-01 should report more for your ATI Rage 3D Pro PCI GPU, but to fully report it NT4/W2K/WXP would be needed.

I was surprised SIV did not report the disk SMART information, do any utilities manage to do this please?

red-ray wrote on 2021-01-25, 23:04:

I was surprised SIV did not report the disk SMART information, do any utilities manage to do this please?

Don't know about SMART function on this setup. This is a fresh Win98SE setup with an old Fujitsu disk. This "computer" is just some parts connected together laying on my workbench... Not even a sound card. I figure to test every Socket 7 possible CPU, just because know I can...

I can test the SIV with a Win2000 setup, if I found all drivers for this install.

Vincent

Grosvince wrote on 2021-01-26, 09:07:

Don't know about SMART function on this setup. This is a fresh Win98SE setup with an old Fujitsu disk. This "computer" is just some parts connected together laying on my workbench... Not even a sound card. I figure to test every Socket 7 possible CPU, just because know I can...

I can test the SIV with a Win2000 setup, if I found all drivers for this install.

I have W2K on a Socket 7 board with the same chipset as yours and the only drivers I needed were for GPUs.

I wonder what is happening with SMART, I suspect doing SIV32L -DBGSDM > SIV_DBGOUT.log from a command window may provide some insight.

red-ray wrote on 2021-01-26, 10:43:

Grosvince wrote on 2021-01-26, 09:07:

Don't know about SMART function on this setup. This is a fresh Win98SE setup with an old Fujitsu disk. This "computer" is just some parts connected together laying on my workbench... Not even a sound card. I figure to test every Socket 7 possible CPU, just because know I can...

I can test the SIV with a Win2000 setup, if I found all drivers for this install.

I have W2K on a Socket 7 board with the same chipset as yours and the only drivers I needed were for GPUs.

I wonder what is happening with SMART, I suspect doing SIV32L -DBGSDM > SIV_DBGOUT.log from a command window may provide some insight.

SMART was turned off in Setup !

Grosvince wrote on 2021-01-31, 15:46:

SMART was turned off in Setup !

Thank you for the update, so I guess SMART is now being reported, is it? What's strange though is that SIV should have enabled it anyway, I would need to see the SIV32L -DBGSDM > SIV_DBGOUT.log output to tell more.

Schule04 wrote on 2021-03-20, 20:24:

The Mainboard is an "Aquarius systems 5BLMP10E".

Thank you, SIV worked as I hoped and I did not spot any errors, did you? SIV did not report the Cyrix 5530 Kahula/Geode Video Controller as a GPU as no driver was installed, is there an XP/2K driver for it? I wonder if any on the W2K ones from http://www.video-drivers.com/companies/268.htm can be adjusted to work.

The SMBIOS did not report the motherboard as an Aquarius systems 5BLMP10E which is why SIV did not, see [Machine]. I had a look in the .DMI file, given what I found was able to adjust SIV so it should and also to report the BIOS version as VER:5.06.13 rather than 4.51 PG. Does >= SIV 5.57 Beta-08 report these?

Out of interest does >= SIV 5.57 Beta-08 report your C7-M ULV + P54CQS as such? If not I will need SIV save files to fix it.

I suspect the P54CQS motherboard does not have SMBIOS/DMI data which is why CPUZ is unable to report the motherboard.

Here is the report for the C7-M ULV. The CPU is detected properly. The mainboard is called "VIA Demo Board" by all profiling tools so I guess that name comes from the BIOS. The actual name of this machine is a "Wyse Xn0L" laptop.

There seems to be an issue with ACPI.

Here's the Geode machine again. I noticed some strange things in the machine window:

- Max CPU frequency is only 200MHz, should probably be 333MHz instead

- The memory controller is not EDO, it's SDRAM. This board has two memory slots. The slot labeled A0 is a full size SDRAM slot while slot A1 is SODIMM.

Schule04 wrote on 2021-03-22, 20:26:

Here's the Geode machine again. I noticed some strange things in the machine window:

- Max CPU frequency is only 200MHz, should probably be 333MHz instead
- The memory controller is not EDO, it's SDRAM. This board has two memory slots. The slot labelled A0 is a full size SDRAM slot while slot A1 is SODIMM.

Thank you for checking out the changes I made which all worked as I intended.

Yes, the BIOS sets up some strange values up in the SMBIOS/DMI, SIV merely reports what the BIOS sets and to fix these an updated BIOS would be needed. I expect all of AIDA64 + CPUZ + HWiNFO will report the same strange values.

Schule04 wrote on 2021-03-22, 19:50:

Here is the report for the C7-M ULV. The CPU is detected properly. The mainboard is called "VIA Demo Board" by all profiling tools so I guess that name comes from the BIOS. The actual name of this machine is a "Wyse Xn0L" laptop.

There seems to be an issue with ACPI.

Thank you for checking out SIV and you certainly have some interesting systems that I have not tested SIV on before.

It looks like some of the ACPI methods timed out while processing the [acpi-eval] section. This happens one in a while so I have adjusted >= SIV 5.57 Beta-09 to skip the problematic methods.

I also noted SIV failed to report the DIMM SPD data as it did not have support for the VIA VT8237S SMBus so I added this, please may I have new save files?

Looking further only the PCIe information was reported on the initial screen so I added the AGP information.

I will see if I can find a way to detect it's a "Wyse Xn0L", what description does it report when it posts? If "Wyse Xn0L" then I may be able to get SIV to scan the System ROM, if you do Menu->File->Save->Save System ROM does the .ROM file contain "Wyse Xn0L"?

Here's the report and bios dump. The BIOS shows a custom WYSE boot logo so they definitely have customized it

Schule04 wrote on 2021-03-26, 01:30:

The BIOS shows a custom WYSE boot logo so they definitely have customized it

Thank you for checking out the changes which all worked as I intended. I can't find anything in the .ROM file to show it's a WYSE BIOS and AFAIK adding a boot logo does not result in a textual name in the BIOS.

I did spot the BIOS reports Single Channel when SIV reported Dual Channel so have fixed this for 5.57 Beta-10 and also added the command rate. I can see tCL must correct as both SIV and the BIOS report 4 and am wondering about tRCD + tRP + tRAS + tRFC. Does SIV report the same as HWiNFO and/or CPUZ please?

Here are the reports

Schule04 wrote on 2021-03-29, 13:04:

Here are the reports

Thank you, all four utilities report the same 😀. I asked as looking at the SPD I expected 4-4-4-12 rather than 4-5-5-13 for the 266 MHz clock and wonder what would happen if you interposed the SO-DIMMs as strange things happen from time-to-time when unmatched DIMMs are used.

Schule04 wrote on 2021-03-20, 21:46:

Pentium P54CQS also is not detected correctly https://www.vogons.org/download/file.php?id=105773 […]
Show full quote

Pentium P54CQS also is not detected correctly

I suspect P54C may be correct as looking at https://en.wikipedia.org/wiki/P5_(micro ... re)#P54CQS the P54CQS is a 0.35μm TCP package.

It's also possible that the 0.60μm and 0.35μm parts return identical CPUID information which makes reporting them correctly tricky/impossible. If you have both a P54CQS + 120MHz P54C please will you post what SIV reports for both?

For the wikipedia article and the source it cites, all 120MHz parts should be P54CQS.

Schule04 wrote on 2021-03-31, 22:52:

For the wikipedia article and the source it cites, all 120MHz parts should be P54CQS.

Thank you for the link, but the article says the "Pentium Is First CPU to Reach 0.35 Micron, New Manufacturing Process Pushes Intel Processor to 120 MHz" so it's possible 120MHz P54C CPUs were produced after the article was published on 27-Mar-1995 as https://en.wikipedia.org/wiki/P5_%28microarch … ls_and_variants seems to report.

It does state "The ’CQS is based on the C2 stepping of the P54C" so I suspect it's a 0.35µm P54CQS if it's stepping 5/C2 @ 120MHz so will implement this for SIV 5.57 Beta-10 . AFAIK a stepping 6/E0 CPU @ 120 MHz is a 0.60µm P54C.

Looks like this system is stepping C2, so the smaller core.
Is there a way to distinguish between a 120MHz C2 core that has been overclocked to 133MHz and a real 133MHz part?

Schule04 wrote on 2021-04-01, 18:11:

Is there a way to distinguish between a 120MHz C2 core that has been overclocked to 133MHz and a real 133MHz part?

Thank you for checking out SIV on your P54CQS, I have just uploaded SIV 5.57 Beta-10 which fixes/improves things and should report as below.

Yes, AFAIK a real 133MHz is a P54CS which has a higher stepping and I expect Beta-10 will report your P54CQS as such even if you overclock it to 133MHz.