OMG, is it true!? Someone has MR BIOS for SIS 471!?

Please dump it!

We can still change chipset registers to disable shadow ram 😀

1uint8_t val;
2
3int main(int argc, char*argv[]) {
4 outp(0x22, 0x52); // index = Reg 52
5 outp(0x23, 0);    // disable any shadow ram 
6
7 outp(0x22, 0x53); // index = Reg 53
8 val = inp(0x23);  // read
9 val &= 0x70;      // disable caching & rom combining(?)
10 outp(0x23, val);  // write changed
11 return 0;
12}

GigAHerZ wrote on 2026-02-08, 16:09:

OMG, is it true!? Someone has MR BIOS for SIS 471!?

Please dump it!

T48 should arrive Thursday per fedex estimates. Hopefully using one is not too hard to figure out. Documentation looks quite good.

Mov AX, 0xDEAD wrote on Yesterday, 06:48:

We can still change chipset registers to disable shadow ram :) […]
Show full quote

We can still change chipset registers to disable shadow ram 😀

Copy code to clipboard

1uint8_t val;
2
3int main(int argc, char*argv[]) {
4 outp(0x22, 0x52); // index = Reg 52
5 outp(0x23, 0);    // disable any shadow ram 
6
7 outp(0x22, 0x53); // index = Reg 53
8 val = inp(0x23);  // read
9 val &= 0x70;      // disable caching & rom combining(?)
10 outp(0x23, val);  // write changed
11 return 0;
12}

Unfortunately as soon as I run that (or do the same register change using debug) I can no longer access the C: drive and the system is just dead. Apparently it doesn't like that being changed. I see nothing in the datasheet indicating why turning shadow off should cause any problems to the running system, but it seems to upset DOS at least.

lsorense wrote on Yesterday, 16:44:

Unfortunately as soon as I run that (or do the same register change using debug) I can no longer access the C: drive and the system is just dead. Apparently it doesn't like that being changed. I see nothing in the datasheet indicating why turning shadow off should cause any problems to the running system, but it seems to upset DOS at least.

I did find a claim that they are modifying some of the code as it is copied to shadow ram and it doesn't work without the modifications done. if that is what is happening, then that is a clever way to prevent access to the rom chip contents. I suppose they might even be doing it based on features of the board as an optimization with putting in the correct code for the board to avoid some branches having to be checked each time.

Well only 3 days until I should get the T48 and should be able to dump the actual chip itself.

I managed to write a program that does give me a BIOS dump with checksum 0x0.

1#include <stdint.h>
2#include <conio.h>
3#include <stdio.h>
4#include <stdlib.h>
5#include <i86.h>
6#include <dos.h>
7
8uint8_t val;
9
10uint8_t orig_reg52;
11uint8_t orig_reg53;
12
13#define BIOSLOW ((char far *)MK_FP(0xF000, 0x0000))
14#define BIOSHIGH ((char far *)MK_FP(0xF800, 0x0000))
15
16char far *bios_low_ptr = BIOSLOW;
17char far *bios_high_ptr = BIOSHIGH;
18
19void disable_shadow() {
20    outp(0x22, 0x52); // index = Reg 52
21    orig_reg52 = inp(0x23);
22    outp(0x23, 0);    // disable any shadow ram
23
24    outp(0x22, 0x53); // index = Reg 53
25    orig_reg53 = inp(0x23);  // read
26    val = orig_reg53;
27    val &= 0x70;      // disable caching & rom combining(?)
28    outp(0x23, val);  // write changed
29}
30
31void reenable_shadow() {
32    outp(0x22, 0x52); // index = Reg 52
33    outp(0x23, orig_reg52);    // disable any shadow ram
34
35    outp(0x22, 0x53); // index = Reg 53
36    outp(0x23, orig_reg53);    // disable any shadow ram
37}
38
39int main(int argc, char*argv[]) {
40    FILE *file;
41    int i;
42    size_t size = 32 * 1024;
43    char* bios_low = (char *)malloc(size);
44    char* bios_high = (char *)malloc(size);
45    uint8_t checksum = 0;
46
47    disable_shadow();
48    for (i=0;i<size;i++) {
49        bios_low[i] = bios_low_ptr[i];
50        bios_high[i] = bios_high_ptr[i];
51        checksum = (checksum + bios_low[i] + bios_high[i]) % 256;
52    }
53    reenable_shadow();
54    printf("Checksum: %.2x\n", checksum);
55
56    file = fopen("bios471.bin", "wb");
57    if(file) {
58        fwrite(bios_low, 256, 128, file);
59        fwrite(bios_high, 256, 128, file);
60        fclose(file);

61    }
62    return 0;
63}

So I now have what I think is a valid dump. I was thinking I should put it on vogon drivers but I would need an account and it seems to do that I have to PM SquallStrife aka RetroSwim but I don't think I have been around long enough/posted enough to be allowed to PM. Oh the punishment for only reading stuff for years and not participating.

Or if there is another suggestion for a place to put a 64MB rom dump I am open to suggestions.

lsorense wrote on Yesterday, 19:25:

So I now have what I think is a valid dump. I was thinking I should put it on vogon drivers but I would need an account and it seems to do that I have to PM SquallStrife aka RetroSwim but I don't think I have been around long enough/posted enough to be allowed to PM. Oh the punishment for only reading stuff for years and not participating.

Or if there is another suggestion for a place to put a 64MB rom dump I am open to suggestions.

I of course meant 64KB. Sheesh.

And I did my first ever dos program, and first time ever actually using the open watcom compiler. And I still remember how to use ftp to transfer files.

lsorense wrote on Yesterday, 19:25:

So I now have what I think is a valid dump. I was thinking I should put it on vogon drivers but I would need an account and it seems to do that I have to PM SquallStrife aka RetroSwim but I don't think I have been around long enough/posted enough to be allowed to PM. Oh the punishment for only reading stuff for years and not participating.

Or if there is another suggestion for a place to put a 64MB rom dump I am open to suggestions.

Can't you just attach the file to a comment you're making here? Just like in all the other bios threads were the bios files are attached to the comments people post.

another option would be to go to theretroweb discord and do an improve board request to add the biosfile to the specific board page.

weedeewee wrote on Yesterday, 19:59:

Can't you just attach the file to a comment you're making here? Just like in all the other bios threads were the bios files are attached to the comments people post.

another option would be to go to theretroweb discord and do an improve board request to add the biosfile to the specific board page.

Hmm, for some reason I didn't think that was allowed, but it does appear many people have done so.

Well here it is. Maybe someone can give it a try.

lsorense wrote on Yesterday, 20:06:

Hmm, for some reason I didn't think that was allowed, but it does appear many people have done so. […]
Show full quote

weedeewee wrote on Yesterday, 19:59:

Can't you just attach the file to a comment you're making here? Just like in all the other bios threads were the bios files are attached to the comments people post.

another option would be to go to theretroweb discord and do an improve board request to add the biosfile to the specific board page.

Hmm, for some reason I didn't think that was allowed, but it does appear many people have done so.

Well here it is. Maybe someone can give it a try.

The attachment MRBIOS-SIS471-V3-21.zip is no longer available

Is it ok if I do a submission on theretroweb in your name of the bios file ?
Asus VL/I-486SV2G is your board, right ? https://theretroweb.com/motherboards/s/asus-v … 86sv2g-rev.-1.0

weedeewee wrote on Yesterday, 20:15:

Is it ok if I do a submission on theretroweb in your name of the bios file ?
Asus VL/I-486SV2G is your board, right ? https://theretroweb.com/motherboards/s/asus-v … 86sv2g-rev.-1.0

Perfectly fine with me. That is my board. Rev 1.5, although that probably doesn't matter. Of course the original BIOS was not MRBIOS. I think it was Award. I bought that as a replacement back in 1995.

lsorense wrote on Yesterday, 20:32:

Perfectly fine with me. That is my board. Rev 1.5, although that probably doesn't matter. Of course the original BIOS was not MRBIOS. I think it was Award. I bought that as a replacement back in 1995.

Thank you so much for the dump — I think this indeed is one of the most sought after images!

I've noticed your board has 1MB cache and 48MB RAM — did you use that configuration back in the day, or did you upgrade it now?

lsorense wrote on Yesterday, 20:32:

weedeewee wrote on Yesterday, 20:15:

Is it ok if I do a submission on theretroweb in your name of the bios file ?
Asus VL/I-486SV2G is your board, right ? https://theretroweb.com/motherboards/s/asus-v … 86sv2g-rev.-1.0

Perfectly fine with me. That is my board. Rev 1.5, although that probably doesn't matter. Of course the original BIOS was not MRBIOS. I think it was Award. I bought that as a replacement back in 1995.

rev 1.5 ? the retroweb only has an entry for rev 1.0.
Can you look at the page and check if there's any clear differences ?
There seems to be mention of some revisions having a 3v3 regulator.

If your board has a 3v3 regulator, you might want to post a photo of the front& back as well, so a new entry on trw can be created 😀

jheronimus wrote on Yesterday, 20:34:

lsorense wrote on Yesterday, 20:32:

Perfectly fine with me. That is my board. Rev 1.5, although that probably doesn't matter. Of course the original BIOS was not MRBIOS. I think it was Award. I bought that as a replacement back in 1995.

Thank you so much for the dump — I think this indeed is one of the most sought after images!

I've noticed your board has 1MB cache and 48MB RAM — did you use that configuration back in the day, or did you upgrade it now?

It started out life with 256K cache and 16MB ram back in 1992. It was originally my farther's CAD system. With some upgrades to disk and such it eventually got a CDROM drive and a larger disk to go to windows 95, which required the BIOS upgrade to get past the 504MB limit. In 1997 he replaced it with a Pentium Pro 200 and I took over the 486 at university. The ram got upgraded to 48MB later when another machine I helped someone with died and it had 32MB ram in it I could use (it is all 16MB 72 pin) and the cache I bought new chips 3 or 4 years ago when I decided I wanted a DOS/windows 98 gaming system again and I figured why not max out the cache. Also switched to IDE to CF adapter at that point since the old IDE disks are not very happy anymore. After I finished university it ran as a file server for my dad for many years until we eventually wore out the bearings on an 18GB Quantum Fireball. It had about 80000 power on hours according to SMART data, so I think it did OK. The Pentium Pro took over the file server job and a Core i7 became a new cad system (there was an Athlon 64 system in there at some point too). These days the file server is some little i5 I put together for him. Original power supply is still doing fine, and clearly so is the motherboard and CPU. Even the nice PC Power and Cooling heat sink fan is still running perfectly after all these years on the 486.

weedeewee wrote on Yesterday, 20:38:

rev 1.5 ? the retroweb only has an entry for rev 1.0. Can you look at the page and check if there's any clear differences ? Th […]
Show full quote

lsorense wrote on Yesterday, 20:32:

weedeewee wrote on Yesterday, 20:15:

Is it ok if I do a submission on theretroweb in your name of the bios file ?
Asus VL/I-486SV2G is your board, right ? https://theretroweb.com/motherboards/s/asus-v … 86sv2g-rev.-1.0

Perfectly fine with me. That is my board. Rev 1.5, although that probably doesn't matter. Of course the original BIOS was not MRBIOS. I think it was Award. I bought that as a replacement back in 1995.

rev 1.5 ? the retroweb only has an entry for rev 1.0.
Can you look at the page and check if there's any clear differences ?
There seems to be mention of some revisions having a 3v3 regulator.

If your board has a 3v3 regulator, you might want to post a photo of the front& back as well, so a new entry on trw can be created 😀

I don't see any changes. The boards with 3.3V are SV2GX4 as far as I know, since those can support the DX4 chips. Mine doesn't have that although it does have a white sticker covering up the X4 part, and a number of voltage regulator components are not populated.

lsorense wrote on Yesterday, 20:43:

I don't see any changes. The boards with 3.3V are SV2GX4 as far as I know, since those can support the DX4 chips. Mine doesn't have that although it does have a white sticker covering up the X4 part, and a number of voltage regulator components are not populated.

Ok.
Submitted.

weedeewee wrote on Yesterday, 20:47:

Ok.
Submitted.

So looking at the board, it is visually identical to a sv2gx4 rev 2.0, but does not have some jumpers and voltage regulator parts installed, and the X4 in the name is covered with a square of white paint. So it seems they were making changes to support the DX4 but kept using the same board for the older model by just leaving out the parts that it didn't need. The gx4 bios list even has mentions of a rev 1.7 board, so they went through quite a few revisions it seems.

Oh and can anyone believe that Sony 3.3V coin cell from 1992 still has full voltage?

I also miss those days when Asus model numbers meant something. VL/I = Vesa Local Bus and Isa slots. SV2G - Sis chipset, 2 Vesa slots, Green. And the X4 added for the DX4 supporting variant. Or the Pentium Pro board we had (well I guess I still have it, but it's not working right, I think it needs a recap): ASUS P/I-P6NP5, PCI and ISA sots, Pentium Pro (P6) Natoma (440FX) chipset, 5 PCI slots.