GL1zdA wrote on 2022-10-31, 22:45:

are you sure the 540 doesn't have the device and 320 has?

Yes, 100% sure as there is no Display class device on the 540

Irinikus wrote on 2022-10-30, 15:57:

It's a pleasure! 😀

Thank you again and I have just uploaded SIV V5.67 Beta-08, please use this rather than the Dump-08 test SIV.

Here's the latest output from my Socket 8 system, which is still a build in progress:

Irinikus wrote on 2022-11-04, 15:22:

Here's the latest output from my Socket 8 system, which is still a build in progress:

Thank you, I can see the SIMM size reporting is now fixed 😀 Why did the time zone change from South Africa to Egypt?

red-ray wrote on 2022-11-04, 20:27:

Irinikus wrote on 2022-11-04, 15:22:

Here's the latest output from my Socket 8 system, which is still a build in progress:

Thank you, I can see the SIMM size reporting is now fixed 😀 Why did the time zone change from South Africa to Egypt?

I’m in Pretoria, which is on the same timeline as Harare Zimbabwe and Cairo Egypt. I’ve just done a reinstall and selected Cairo, as it’s the first city on the list with a +2 hour timeline.

Irinikus wrote on 2022-11-05, 03:31:

I’m in Pretoria, which is on the same timeline as Harare Zimbabwe and Cairo Egypt. I’ve just done a reinstall and selected Cairo, as it’s the first city on the list with a +2 hour timeline.

OK, I was half wondering if this was a side effect of the clocks going back an hour. I was aware those time zones are all UTC+2 and wonder why they all exist, maybe the summertime changes are at different times.

Here's the information from the Alienware system:

I unfortunately can't give you the saves, as it encounters the following error when I try to do so:

So I've given you wats found in the various fields:

This is what CPUZ and GPUZ show:

red-ray wrote on 2019-08-22, 10:17:

If one of the save sections has an issue do Menu->File->Save->Save Selected and deselect that section before pressing [ Start ].

Irinikus wrote on 2022-11-06, 07:17:

I unfortunately can't give you the saves, as it encounters the following error when I try to do so:

Thank you for trying and posting the screen shots. Please do Menu->File->Save->Save Selected, deselect [gpu-g-sync] then press [Start].

There should also be a SIV64X.dmp file, please zip and post it so I can look into why there is an issue with [gpu-g-sync] and fix it. The Menu->Windows->Parameters->Error Reporting panel will tell you the location of SIV .DMP files.

If general SIV looks to have done OK and once I get the partial save files I will adjust the SIO voltage scaling so such as +12 is reported correctly.

Here you go:

Irinikus wrote on 2022-11-06, 09:19:

Here you go:

Thank you, using SIV64X.exe.1076.dmp I could see the exact cause the issue and should have fixed it in the attached SIV64X V5.67 Sync-09 test SIV. Please will you check if Menu->Machine->GPU Detail->GPU G-Sync (same as [gpu-g-sync]) now works?

I have also adjusted the SIO voltage reporting. I may be able to further improve the SIO reporting if you post a screen shot showing the temperatures, fan speeds and voltages reported by the BIOS.

VBAT +2.88 seems rather low, how old is the CR2032 battery?

It now works!

I hardly ever have this machine plugged in, so that's why the battery's almost flat!

Irinikus wrote on 2022-11-06, 13:01:

I hardly ever have this machine plugged in, so that's why the battery's almost flat!

Thank you for confirming the fix worked and OK about the battery, I get the same with many of my systems and need to replace batteries more often than I would like.

The fixes will be in SIV 5.67 Beta-09 which I plan to release in 3 or 4 days' time.

Cool! 😀

It's hard to give all of the machines the attention they need! (You can only play with one machine at a time! Some of them haven't been booted in years!)

Irinikus wrote on 2022-11-06, 13:25:

Cool! 😀

I have just uploaded SIV 5.67 Beta-10 which should tidy up [GPU G-Sync].

Next time you have the Alienware 6A61ITT4C-00 booted please will you generate new save files for me?

Hey, picked up an 11800h HM570 motherboard combo and tried SIV64 with it.

Voltage numbers aren't correct

mastergamma12 wrote on 2022-12-30, 05:53:

Voltage numbers aren't correct

Thank you for the report and apologies for the delay in replying, I was travelling home yesterday and then had things I needed to do.

It's the usual thing of needing to adjust the ITE IT8613E scaling + mapping for the way the Intel HM570 motherboard uses it, how does that attached SIV64X 5.68 Test-05 look to you please?

red-ray wrote on 2022-12-31, 14:52:

mastergamma12 wrote on 2022-12-30, 05:53:

Voltage numbers aren't correct

Thank you for the report and apologies for the delay in replying, I was travelling home yesterday and then had things I needed to do.

It's the usual thing of needing to adjust the ITE IT8613E scaling + mapping for the way the Intel HM570 motherboard uses it, how does that attached SIV64X 5.68 Test-05 look to you please?

Looks good now

Irinikus wrote on 2022-11-04, 15:22:

Here's the latest output from my Socket 8 system, which is still a build in progress: […]
Show full quote

Here's the latest output from my Socket 8 system, which is still a build in progress:

Wow, XP on a socket 8 rig, you must have an awful lot of patience. I found even 2000 to be nearly unusably slow on my PPro rig when I still had it, and it was a 1meg black-top chip running overclocked to 233MHz. I know you've got the PIIOD there, but even at 333MHz I'd think XP wouldn't be a fun time.

I'm probably not being exceedingly fair here, but SIV gets practically everything wrong for my 486 Blue Lightning build. The CPU is not Intel, it's not socket 3, it's not a regular DX (the FPU is a IIT 4C87DLC at 33MHz, not integrated) , and it's not at ~58MHz -- it's an IBM 486BL at 2x33 for 66MHz. The only thing it (surprisingly) manages to report correctly is the L1 cache size -- everything else says it's 8K, but it is indeed 16. I have 256K L2 that it fails to pick up on, but that's forgivable since the latency is such that even Cachechk seems to think it's not there, though there is a difference in <=256K access times and >256K that tells me it's at least enabled. It doesn't support CPUID, so nothing can really detect it properly, to be fair. Given how off-beat this CPU is -- I like to call it a 386 on stilts as that's pretty much what it is -- I'm surprised it's not even more incorrect, really.

The video card in use here is a VLB variant of an S3 Trio64V+, specifically an STB PowerGraph 64VL. It's got 2MB of VRAM and drivers are quite difficult to find. S3 only supported this card with drivers for Windows 9x until early 1997, and even those latest drivers suffer from some corruption in screen savers. It's rather quick, as you'd expect a late VLB card to be. I'm in 800x600 at 75Hz, 256 colors.

I tried to save a text file from SIV but it seems to hang after collecting all the information and putting a bit of garbage data in the txt file it prepared for use. The rest of the program is fine, if not especially performant.

Sedrosken wrote on 2023-01-05, 10:40:

I tried to save a text file from SIV but it seems to hang after collecting all the information and putting a bit of garbage data in the txt file it prepared for use.

Thank you for checking out SIV and I would like to try and improve what SIV reports, but to sensibly do this need a save file.

I see you used the SIV 5.67 release, the "bit of garbage data" is a UTF-8 BOM and all the other data is missing as W95 does not support CP_UTF8, please use SIV32L 5.68 Beta-04 or later. With luck 5.68 Beta-04 may report your GPU. If you don't wish to use SIV32L 5.68 Beta-04 then uncheck Menu->Options->UTF-8 Save before doing Menu->File->Save Local.

On the [Graphics] panel the 2 x 23013 is clearly wrong, this is what W95 reported as is down to a bug in W95, with W98 it would be correct.

Given you system only has 16MB I am pleased to see that SIV runs and suspect it would be less slow with more memory.

BTW I also have XP SP3 on my 128MB PentiumPro @ 233MHz. As I recall it took a some time to install, even longer to install all the updates and takes a while to boot. It takes SIV about 8 seconds to start-up.

Sedrosken wrote on 2023-01-05, 10:40:

it's not at ~58MHz -- it's an IBM 486BL at 2x33 for 66MHz.

Looking at the CPUZ thread I noticed CPUZ also reports ~58MHz. As i486 class CPUs don't have RDTSC support the speed calculations are done using AAM instructions which usually take 15 cycles. Do you know how many cycles it tales on your IBM 486BL and/or have the datasheet. If we can figure out how to detect it's a IBM 486BL then it should be simple enough to adjust the speed calculation. How much do the speeds on Menu->Help->Conversions->CPU MHz vary?

I don't have info for detecting 486BL or 486BL2, but 486BL3 has MSRs. BL and BL2 may, but I'd have to research or test that. For BL3, I have the following info:

1;   MSR 00001004h - IBM 486BL3 - PROCESSOR CONTROL REGISTER			+
2;										+
3;	Size:	24 bits								+
4;										+
5;	Bitfields for IBM 486BL3 Processor Control Register:			+
6;										+
7;	Bit(s)	Description							+
8;	 63-24	reserved							+
9;	 23	OS/2 boot (0=DD1 hardware, 1=DD0 hardware)			+
10;	 22	MOV CR0,x Decode						+
11;		0: DD0, DD1A, DD1B, DD1D hardware				+
12;		1: DD1C hardware						+
13;	 21	reserved							+
14;	 20	Cache Low Power (DD1 only: cache disabled when not in use)	+
15;	 19	reserved							+
16;	 18	NOP timing							+
17;		0: 2 cycles on DD0, 3 cycles on DD1				+
18;		1: 3 cycles on DD0, 2 cycles on DD1				+
19;	 17	bus pipelining for 16-bit accesses				+
20;	 16-5	reserved???							+
21;	 4	MOVS split							+
22;	 3	power-saving cache feature					+
23;	 2	reserved							+
24;	 1	enable MOV CRx decode						+
25;		(reserved on DD1B, DD1C)					+
26;	 0	reserved							+