I never even knew there was a difference between these 2.

By pure coincidence I've taken pics of the 2 Trio64 VLB cards I have (I bought one and received 2, but one of them was already broken. Never got around to test these cards), but I can't read the print on one of these card's BIOS sticker.
I don't think they are Diamond though, but they are the same manufacturer.

Want me to go write down the markings of both stickers and post them here? They are out of storage atm anyway and was about to put them back into storage.

I don't have one of these BIOS reader thingies though.

Ftr, until I read your thread here, I didn't even know these was 2 different Trio64's and I have one of either. Perhaps I can help out a bit?

I wonder what is the performance difference between Stealth64 (968, 868) and Trio64 on VL-bus? Would Trio64 be on the same level as 868/864 or better?

As far as I know the Trio64 should be more or less identical to the 868/864 in DOS. The Windows performance would depend on which RAMDAC the 86x was equipped with. If it has a 16-bit RAMDAC, then the performance should be similar, however I believe there is at least one 86x based card that supports 4MB, whereas all the Trio64 based cards only support up to 2MB.

As for the difference between the 86c764-P and 86C764X, I would guess the P indicates the chip supports a PCI interface (which is not used on a VLB card of course)

Anonymous Coward wrote:

As for the difference between the 86c764-P and 86C764X, I would guess the P indicates the chip supports a PCI interface (which is not used on a VLB card of course)

Were you able to click the image links? The VLB cards came with, both, 764-P and 764X chipset options.

Ftr, I only got the bare cards.

The BIOS stickers on both read as follows:
(the 86C764-P)
MIRAGE P64 V5.01
(c) PHOENIX (c) SPEA

(the 86C764X)
SPEA V 5.52 S
V 7 MIRAGE P64
(c) SPEA (c) PHOENIX

The layout of both cards are identical, I couldn't find any of thier date codes though? But the 86C764X has a date code of 9546 while the 86C764-P has a date code of 9441 (about a year older).

Dunno if this is of any use to you, but like I mentioned, I had them close by anyway.

My benchmarks show that in Win95 the S3 driver that is included in DirectX5 is the fastest.

I have also seen the 764-P and 764X on PCI cards. I couldn't locate any datasheets on the 86c764, so I wonder what they changed and if the BIOSes are compatible.

I just ran across a Hercules Terminator 64/DRAM image online which utilises a 764-P chip with a 1996 datecode! This may imply that the 764X chips did NOT supersede the 764-P chips. Thus far, the 764-P chips I have found contain a late 1994 or very early 1995 datecode, whereas the 764X chips have an early-to-mid 1995 datecode.

According to http://www.plasma-online.de/english/identify/picture/s3.html
The X and -P is mostly additions due to chip revisions and PnP capabilities (whatever this means exactly).
Usually the fixed bugs in later revisions are rarely published.
I got also Spea Mirages P64 with the X and -P and no notable difference.
http://retronn.de/imports/hwgal/hw_s3trio64.html
http://retronn.de/imports/hwgal/hw_graphics_c … _p64_front.html
and if you go for date codes, here is a 764X with mid-late 1996 http://retronn.de/imports/hwgal/hw_s3_trio64.html

Maybe the X version had some DirectX bug fixes?

The X should denote PnP compatibility.
The -P usually means that the chipset supports PCI bus.

A rough guess what PnP compatibility could mean: The card can enable/disable the VGA IRQ?

Maybe it is possible to find the difference when directly comparing a 764 and a 764X card.
I would guess as well from the plasma designs list that the X is Rev 02. and Rev 03 has already DEV ID 8013 then.

Does it seem unlikely that the X BIOS would work on the -P cards?

When I try to run PCPBENCH on my Diamond Stealth 64 DRAM T with chipset 86c764-P and BIOS v2.01, I get an error that VESA Mode 100 could not be initialised. I don't recall seeing this error on my other S3 Trio/ViRGE cards. Is this something that 764-P cards with BIOS v2.02 might fix?

feipoa wrote:

Does it seem unlikely that the X BIOS would work on the -P cards?

There are some differences in the chip initialisation routine (aka vga wake-up). If the VGA BIOS is made to support the both chips, it would work with the both of them. Otherwise it looks like they can be programmed in a way incompatible in both directions: P won't work with X and the other way around. Some details here: http://tyom.blogspot.de/2017/03/hercules-term … rikes-back.html It looks like X may work with 765 (aka Trio64V+) BIOS though.

Does anyone happen to have BIOS version 2.02 (found on 764-P cards) and version 2.09 BIOS (found on 764X cards)? If so, I can try them in my Trio64 VLB.

Another thought - would a 764X BIOS found on PCI cards be the same as that on the VLB version (Diamond Stealth 64 DRAM T)?

Looks like 2.09: http://retronn.de/imports/hwgal/hw_graphics_c … dram_front.html
also 2.09: http://retronn.de/imports/hwgal/hw_s3_trio64.html

I don't have 764X or 764-P from Diamond as VLB.
The 764X should be the PnP version and -P usually names later chip revisions with fixed chip bugs.

Do you think BIOSes from non-Diamond branded VLB 764-P cards will work on the Diamond card?

feipoa wrote:

Do you think BIOSes from non-Diamond branded VLB 764-P cards will work on the Diamond card?

If you have an EPROM burner you can try. VLB cards are like ISA with nothing like PCI ID, where there might be a check.
What are you looking for exactly? Better VESA support?

It is actually not so unusual that cards have no built-in VESA support in BIOS (like for PCPBench Mode 100). However you can always load a TSR.

As far as the Windows drivers go, use the one delivered with DirectX5.
From my tests they are the fastest and most compatible for games regarding DirectDraw and specific modes.

Yea, I was actually hoping for some VESA support from the BIOS. Which Trio64 86c764-P based card to you have and would you be willing to share the BIOS image? Yes, I have an EPROM writer.

The picture of the Trio64 VLB card with the 96 datecode was mine. Would you like a copy of the BIOS image? I am pretty sure I dumped it recently.