I am only interested in information about SiS chipset revisions, I hope this thread does not get hijacked and I don't see any "whom you muted, made this post" in this thread.

Perhaps this post can help you:
Re: Soyo SY-4SAW2 Rev WA3

red-ray wrote on Yesterday, 13:05:

So far I have found Soyo SY-4SAWs for sale with the QR + RU + ST revisions and suspect that these are later than PR so should support EDO.

Sorry, but where do you read this board contains QR + RU + ST revisions?
I do not read anything about that revision on the description and the chips are couvered with stickers.
It also contains the warning: *No liability for techn. Data / *No Liability for Technical Information

I was also searching for a SIS chipsetted board with edo support a few years back. All I can confirm is that PR worked with edo, and I don't think anything else did, though I can't recall if i had any of those "later" revisions to test.

In the end EDO support only brought a very very minimal improvement speed wise on this chipset - this is not where the holy grail lies. In my case I was running the board at a 60mhz FSB - and results were still dissapointing. At more common lower FSB speeds the advantage was almost nil

bertrammatrix wrote on Today, 17:33:

I was also searching for a SIS chipsetted board with edo support a few years back. All I can confirm is that PR worked with edo, and I don't think anything else did, though I can't recall if i had any of those "later" revisions to test.

In the end EDO support only brought a very very minimal improvement speed wise on this chipset - this is not where the holy grail lies. In my case I was running the board at a 60mhz FSB - and results were still dissapointing. At more common lower FSB speeds the advantage was almost nil

Your tests at 60 MHz match to the test of my pal mkarcher: 486 board with UMC 8881E/8886B: The winner is: EDO without L2 (if your only other option is L2 at 3-2-2-2)
Did you notice any difference in how many EDO modules you have tested and in the speed of the EDO modules on your SiS based mainboard?

He got best results with EDO when all L2 cache modules were NOT INSTALLED:
Memory(!) timing 3-1-1-1 at FSB 40 with 1 50 ns EDO module.
Memory timing 4-2-2-2 at FSB 60 with 3 60 ns EDO modules.

bertrammatrix wrote on Today, 17:33:

I was also searching for a SIS chipsetted board with edo support a few years back. All I can confirm is that PR worked with edo, and I don't think anything else did, though I can't recall if i had any of those "later" revisions to test.

In the end EDO support only brought a very very minimal improvement speed wise on this chipset

Thank you for the information, for me speed is not the only reason I would prefer to use EDO, it's down to me already having 4 x 64MB of EDO SIMMs and I would like to have EDO for testing SIV 's SiS 496 reporting. I am also wondering what the EDO Detect bit does, does it just get set while detecting EDO SIMMs or does it remain set after the detection is complete.

Do you still have the PR and/or do you know the PCI revision please? Thus far I have also seen 0x31 from a friends NV 496 and 0x02, I don't know which revision reported this.

red-ray wrote on Today, 18:17:

bertrammatrix wrote on Today, 17:33:

Do you still have the PR and/or do you know the PCI revision please? Thus far I have also seen 0x31 from a friends NV 496 and 0x02, I don't know which revision reported this.

Sorry I don't. I moved away from sis boards altogether and sold off whatever I had.