VOGONS


Asus PVI-486SP3 known revisions

Topic actions

First post, by Comos

User metadata
Rank Newbie
Rank
Newbie

Hi,

over the years I went through the board revisions which I have and Im looking the other ones to update my hardware list.
Currently I own the following revisions: 1.0, 1.2, 1.21, 1.6, 1.8. Beside that Im aware about rev 1.22 aswell.
Does somebody have a revision outside this list and can provide a picture please?

Reply 2 of 34, by Comos

User metadata
Rank Newbie
Rank
Newbie
Zerthimon wrote on 2021-12-14, 22:06:

Should revision 1.21 be after 1.8 ? Or is it not a chronological order ?

I expect that the revision 1.8 is the last one.In chronological order it suppose to be 1.0, 1.2, 1.21, 1.22, 1.6, 1.8

Reply 3 of 34, by debs3759

User metadata
Rank Oldbie
Rank
Oldbie

I have a 1.8 that is working, and a 1.22 that needs work (might have had the wrong jumper settings when I last tried it). I always thought the 1.22 was the last revision, will have to look at the date codes on the chipsets 😀

See my graphics card database at www.gpuzoo.com
Constantly being worked on. Feel free to message me with any corrections or details of cards you would like me to research and add.

Reply 5 of 34, by Comos

User metadata
Rank Newbie
Rank
Newbie
pshipkov wrote on 2021-12-15, 04:12:

If I am not mistaken revisions go like this
1.0 1.2 1.6 1.8 1.21 1.22

Haven't seen other versions.

If revison 1.21 and 1.22 went after rev 1.8, then that would not make much sense regarding JP1 which is unpopulated on later revisons like 1.6 or 1.8.The manufacturing date on the chips is also not accurate, because I have seen chips with 96 date on rev 1.8 and also on rev 1.21/1.22 so it seems they were assembling what they had on stock 😀

Reply 6 of 34, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

It is difficult to conclude really. Quantity of PVI SP3 boards out there is not enough for a good sampling rate.
I drew my conclusion based on the observation that the latest B4 stepping of the SiS 486/497 appears predominantly on 1.21 / 1.22 versions and some 1.8.
If the PCB versions were like how you suggested - 1.0 1.1 1.2 1.21 1.22 1.6 1.8 - it will be difficult to explain the above.

As for JP1 - didn't notice board models missing it, but it sounds like you saw a discrepancy there. Can you clarify?

retro bits and bytes

Reply 7 of 34, by Comos

User metadata
Rank Newbie
Rank
Newbie
pshipkov wrote on 2021-12-16, 04:43:

As for JP1 - didn't notice board models missing it, but it sounds like you saw a discrepancy there. Can you clarify?

Yes, on the later revisions like 1.6, 1.8 I don't have the jumper populated, where I have it on rev 1.0, and rev 1.2 ,1.21.It's also visible on the photos, that I have in my archive somewhere from the internet.

Attachments

  • Rev_1.8.jpg
    Filename
    Rev_1.8.jpg
    File size
    12.77 KiB
    Views
    1215 views
    File license
    Fair use/fair dealing exception
  • Rev_1.22.jpg
    Filename
    Rev_1.22.jpg
    File size
    25.76 KiB
    Views
    1215 views
    File license
    Fair use/fair dealing exception
  • Rev_1.21.jpg
    Filename
    Rev_1.21.jpg
    File size
    39.67 KiB
    Views
    1215 views
    File license
    Fair use/fair dealing exception
  • Rev_1.2.jpg
    Filename
    Rev_1.2.jpg
    File size
    7.67 KiB
    Views
    1215 views
    File license
    Fair use/fair dealing exception

Reply 9 of 34, by libby

User metadata
Rank Member
Rank
Member

Here's a 1.22. It has the jumper being mentioned, which iirc is for disabling the entire onboard SMC SuperIO chip on these, useful for troubleshooting IRQ conflicts mostly. Later revisions switched to a different SuperIO which isn't able to be disabled.

PVI486SP3_122.png
Filename
PVI486SP3_122.png
File size
651.49 KiB
Views
1177 views
File license
Fair use/fair dealing exception

This is also that desirable unicorn B4 revision chipset I suppose. I have heard that sometimes a magic combination of revision and BIOS will allow the board to support EDO, but info seems rather sketchy. I guess this will throw out the theory of chip dates being tied to revisions, as this is B4 but chip dates are 94/95. My suspicion would be that ASUS probably had multiple runs of the blank PCBs over time and would just put on whatever chips they had in stock when building more, or replaced faulty chips with different revisions in the event a board was RMAed.

This came with a 3.6V Cyrix 5x86-120GP in it. It runs at 133MHz without issue.

I have 3-4 more of these boards floating around and will add more whenever I dig them out in the process of slowly testing inventory.

Reply 10 of 34, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

So far didn't stumble upon PVI SP3 that does EDO.
At best they can complete POST with some specific EDO RAM modules but BOOT is not guaranteed, let alone reliable system.
In fact the same applies to late SiS 471 revisions.

Chipset dates are 94/95. Didn't notice specific pattern there. Steppings and years are mixed up.

As for the super io - the PCB models with UMC chip don't have the J1 jumper.
SMC > UMC for the IO system.
The only board version I have seen so far with UMC superIO chip is 1.8.
1.0 1.1 1.2 1.6 1.7 1.21 1.22 seems to be always with SMC.

Also all boards I have seen so far come with VIA clock gens.

The board on the picture you posted is the unicorn indeed.

retro bits and bytes

Reply 11 of 34, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

Is that image 256 colour?

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 12 of 34, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

I think he posterized it - basically reducing the numbers of colors, effectively working around PNG's low levels of compression.
Better will be conversion to jpg or another algorithm that does better at lossy compression.

retro bits and bytes

Reply 13 of 34, by libby

User metadata
Rank Member
Rank
Member
Anonymous Coward wrote on 2021-12-17, 07:48:

Is that image 256 colour?

The forums keep compressing things in weird ways, sometimes reducing the dimensions, sometimes reducing the colors.

Here's a JPEG of it that the forums decided not to compress.

ASUS_PVISP3_122.jpg
Filename
ASUS_PVISP3_122.jpg
File size
592.88 KiB
Views
1142 views
File license
Fair use/fair dealing exception
pshipkov wrote on 2021-12-17, 08:02:

I think he posterized it - basically reducing the numbers of colors, effectively working around PNG's low levels of compression.

She, fwiw. 😀

Reply 14 of 34, by kixs

User metadata
Rank l33t
Rank
l33t
libby wrote on 2021-12-17, 05:33:

This came with a 3.6V Cyrix 5x86-120GP in it. It runs at 133MHz without issue.

Does it work in Windows 95?

I've tested a few and the ones that work fine in DOS and DOS games, won't load Windows 95. Didn't use any special programs for tuning the CPU.

Requests are also possible... /msg kixs

Reply 15 of 34, by libby

User metadata
Rank Member
Rank
Member
kixs wrote on 2021-12-17, 09:30:
libby wrote on 2021-12-17, 05:33:

This came with a 3.6V Cyrix 5x86-120GP in it. It runs at 133MHz without issue.

Does it work in Windows 95?

I've tested a few and the ones that work fine in DOS and DOS games, won't load Windows 95. Didn't use any special programs for tuning the CPU.

Have not tried it yet. I'll toss the combo in a case since I'm keeping this board, and see if it'll install OSR2 this weekend. Will probably try seeing if I can get some EDO memory working in it as well, I have around 50 pounds of SIMM RAM here, something out of it has to work.

Reply 16 of 34, by frudi

User metadata
Rank Member
Rank
Member
Comos wrote on 2021-12-14, 22:11:

I expect that the revision 1.8 is the last one.In chronological order it suppose to be 1.0, 1.2, 1.21, 1.22, 1.6, 1.8

That's not how versioning typically works. A bigger number normally means a later version, even when the numbers switch from single to double digit. If 1.21 and 1.22 were just minor fixes of 1.2, they would typically have been called 1.2.1 and 1.2.2 instead.

As is, without having knowledge that Asus used some weird own versioning system (which they very well might have for all I know, I'm just saying we don't know that they did), it should be assumed that revisions in this case of this motherboard should go in order 1.0, 1.2, 1.6, 1.8, 1.21 and 1.22.

Reply 18 of 34, by libby

User metadata
Rank Member
Rank
Member

I operate under the assumption that 1.21 and 1.22 are dramatic updates of the board over a 1.8.

I can't begin to fathom ASUS's janky numbering system back then, but the 1.21 and 1.22 boards seem to have the later more desirable chipsets, which is enough for me to believe they are revisions of the board that were being sold or sent as warranty replacements through the end of 1996 into early 97.

The presence of the SMC disable jumper is likely not indicative of board age. The chip may simply have stopped being available for a period, so they used something else and didn't populate the jumper as it would have undesired effects with the different chip. The SMC chip provides floppy, parallel, both serial ports and basic IDE capability, but I don't think the IDE portion is being used or if it is, it's this chip handling interrupts while data is being processed by the SiS bridge, perhaps using the VLB. This may also be a reason the disable jumper exists - this chip does exactly what a full blown VLB multi-I/O IDE and floppy card did, and it would massively conflict with one should someone want to install one for whatever reason. If a third party expansion card gets installed in the system which has parallel, serial etc, you would need to disable the onboard SuperIO in order for that card's ports to be BIOS supported.

Reply 19 of 34, by debs3759

User metadata
Rank Oldbie
Rank
Oldbie
libby wrote on 2021-12-17, 09:07:
pshipkov wrote on 2021-12-17, 08:02:

I think he posterized it - basically reducing the numbers of colors, effectively working around PNG's low levels of compression.

She, fwiw. 😀

I think it confuses these lads to see us lasses in a forum like this 😀

See my graphics card database at www.gpuzoo.com
Constantly being worked on. Feel free to message me with any corrections or details of cards you would like me to research and add.