[Simon82]

What a wonderful past age! I remember that bilinear filtering was so connected to the idea of "acceleration" that when I saw a Moto game accelerated by the G100 I did not understand why pixel was not filtered adn when a friend talk me about the lack of this feature it make me say omg...
I remember that my first 3d acceleration card the wonderful(l of sh.. ) VIEWTOP 3D TITAN 4000 S3 Trio3D Rev.b based card ( ) capable of Direct3D acceleration even if no drivers could accelerate even one game without blank shaded texture or a number of problems, make me so sad that when I saw one of that Superbike games running on a Voodoo2 some years after that, I remained shocked... it was like born another time...

[[EOCF] Tim]

Nice scan Swaaye!! Excellent!

Here is a little piece I wrote on it, along with my Tasmania 3D board (Thanks T. ! ), never seen any other Tasmanias.

A very early 3D accelerator based on the Yamaha YGV612 Rendering Polygon Accelerator.


The history of the Tasmania 3D starts in the mid-to-late 1980s, with Western Digital adding the Paradise subsidiary to their Multimedia Products Unit in 1986. Paradise produced some of the best VGA cards of the era. The next evolutionary step for many companies was dedicated 3D acceleration.

On Sept 11, 1995, Western Digital and Yamaha announced that they formed a strategic partnership to develop and market 3D products for the PC.

At the time of announcement, 3D acceleration was considered as the next step in real life graphics for the pc, and with the advent of new 3D APIs, and faster interfaces, a number of companies dared to step into this new realm of PC computing.

The 3D game board would be based on the Yamaha YGV612 3D Rendering Polygon Accelerator, and would be marketed under the Paradise brand name.

In the press release, officials from WD and Yamaha explained why they chose to partner up:
"Three dimensional graphics is the next technology wave in multimedia," explained Leonard Sharp, who was at the time the vice president of marketing for Western Digital's Multimedia Products Unit. "Yamaha's current 3D technology is the right solution at the right time for consumers to realize the excitement of 3D.
“The Yamaha and Western Digital partnership will deliver a new dimension in reality and `twitch response' to PC games that will allow them to rival the best console game machines," said Robert Starr, general manager for sales and marketing for Yamaha Systems Technology Inc. "Our core 3D accelerator expertise combined with Western Digital's board-level knowledge is already attracting OEMs, resellers and game developers."

On Sept 25, 1995, the official announcement came when Western Digital announced its first 3D accelerator, based on the Yamaha YGV612, called the Paradise Tasmania 3D.

The actual Yamaha YGV612 chip was announced Nov 14, 1994, when Yamaha announced two new 3D accelerator chips, the Vram based YGV611 RPA, and a cost reduced, Dram based YGV612 RPA2 that would be released at a later date.

Running at 50Mhz, and produced on a 208-pin QFP package, the YGV612 was one of the first 3D chips to market, and performed polygon rendering, shading, hidden surface removal, texture mapping and Z-buffering. Performance was measured at 300,000 Gouraud-shaded polygons per second and Z-buffering and 150,000 shaded/texture mapped polygons/sec.

The highest resolution supported in 3D was 640 x 480 with 65k colours and supported a maximum of 4Mb Dram.


WD tried to give their first 3D card an edge over the soon to be released competing products by keeping the costs low while still satisfying the consumer’s desire for 3D performance.

At $250, the Tasmania 3D gaming board gave software developers a quick path to porting their 3D games while maintaining compatibility with any VGA board, by using a 9 pin loopback cable, similar to the later released Voodoo Graphics.

The Tasmania controlled the switching from 2D acceleration to 3D acceleration using an on-board analogue MUX. This back-end circuitry takes in the analogue data from the VGA board via a loop-back cable and either routes it directly to the display or switches to the 3D engine.

At the time, 2D graphics cards were still vastly superior to the integrated 2D/3D solutions, so combining 3D acceleration with any 2D board that a consumer already had was a logical move, and also cut down the cost considerably.

The first boards came with a PCI interface, though Western Digital did plan to offer a VL-Bus based board.

The company chose to use the more common polygon rendering approach of the Yamaha chip, rather than employ an alternative rendering technique, such as the curved surfaces (quadratic texture mapping) used in the NV1, produced by their competitor NVidia.

On Sept 27, 1995, only two days after the Tasmania 3D was announced, Western Digital announced that they sold their Multimedia Products Unit to Phillips, which re-established the Paradise name. The sale was finalised on October 31, 1995. The Paradise brand was then owned by the Multimedia PC Group of Philips Semiconductors, which was a Philips Electronics company.

The board began shipping in the same month, but was expected to be widely available in October.

With the upcoming arrival of Direct3D in Windows 95, Yamaha announced support for the new API already back in April, 1995, and subsequently Phillips confirmed that announcement once again with their Tasmania card on April 10, 1996.

The card supported games written for DOS, Windows 3.1 and Windows 95 and was backed up by the major API's at the time, namely RenderWare, BRender and RenderMorphics.

Shipping with Tasmania 3D were two popular PC games: FX Fighter by GT Entertainment, and Domark ‘s Tank Commander . In addition, for a limited time, Tasmania 3D customers could register with Philips to receive three more games from Mindscape: SU-27 Flanker, CyberSpeed and Air Power.

The card was discontinued August 1996, and Direct3D drivers were never released. (More sources needed on D3D drivers and date discontinued)

[[EOCF] Tim]

And of the card itself.

[Simon82]

EOCF we really share the same passion! You've wonderful rare cards!

[[EOCF] Tim]

Originally Posted by Simon82

EOCF we really share the same passion! You've wonderful rare cards!

Thanks man, I guess we do!

(cuz you are Italian, you might be interested in this, I'm selling an Italian box, no card)

http://cgi.ebay.co.uk/ws/eBayISAPI.d...m=110366917050

[Davros]

a box ive seen it all now

[Dave B(TotalVR)]

But more importantly...

Who is licencing the SGX543MP16, and could this multi-core chip be multi-chipped too?

Listed here:
http://en.wikipedia.org/wiki/PowerVR...XT_.28SGXMP.29


I WANT ANOTHER POWERVR AIB!!!!!!11one

[Ailuros]

Originally Posted by Dave B(TotalVR)

But more importantly...

Who is licencing the SGX543MP16, and could this multi-core chip be multi-chipped too?

Someone was just speculating for your 1st question and IMHLO no for the 2nd (unless of course you'd think of two SoCs in the same device which doesn't make much sense).

[Scali]

Isn't the Matrox M3D the only card Matrox has ever made, which didn't use one of their own graphics chips? It uses a PowerVR PCX2, right?
I have a Videologic Apocalypse 3Dx card, which is essentially the same thing as the Matrox M3D I suppose.
It was my first 3d accelerator. I liked the hardware, but at the time, people were moving away from vendor-specific API support to things like miniGL and Direct3D. The driver support wasn't very good imho. The only game I ever played 'natively' on the card was Tomb Raider, but that looked incredible and ran really smoothly.
The card also gave me my first taste of programming Direct3D. I think we were at version 5 back then.

[Simon F]

Originally Posted by Scali

Isn't the Matrox M3D the only card Matrox has ever made, which didn't use one of their own graphics chips? It uses a PowerVR PCX2, right?

Yes.

The only game I ever played 'natively' on the card was Tomb Raider, but that looked incredible and ran really smoothly.

That's probably mainly because the original Tomb Raider software renderer code was very well written (thank you, Core) which made porting it relatively painless.

(FWIW, we also started on a port of a different company's game whose source code was inspired by a bowl of pasta... Funnily enough, the port was abandoned.)

[TEXAN*]

Originally Posted by Dave B(TotalVR)

But more importantly...

Who is licencing the SGX543MP16, and could this multi-core chip be multi-chipped too?

Listed here:
http://en.wikipedia.org/wiki/PowerVR...XT_.28SGXMP.29


I WANT ANOTHER POWERVR AIB!!!!!!11one

From what I've heard, SEGA is developing a HD casual system to compete with the successor to the Wii.

They might use the SGX543MP16 or they may just base it off the RINGWIDE or RINGEDGE.

[Davros]

Originally Posted by Simon F

whose source code was inspired by a bowl of pasta

my code is usually inspired by spaghetti

ps: tim check your pm's

[Grall]

Originally Posted by Simon F

(FWIW, we also started on a port of a different company's game whose source code was inspired by a bowl of pasta... Funnily enough, the port was abandoned.)

Surely there would be no harm in revealing the name of the company/game after so many years... Don't be coy with us please, nobody will mind/care.

Btw, I loved my PCX1/2 boards. Unfortunately I fried the PCX2 by trying to overclock it, but it did go (somewhat) faster with a slightly higher frequency oscillator... Kids today have it easy, back then you overclocked by breaking out the soldering iron, lol!

[Scali]

Originally Posted by Simon F

That's probably mainly because the original Tomb Raider software renderer code was very well written (thank you, Core) which made porting it relatively painless.

Yea, my '2D' card was actually a Matrox Mystique. There was also a 3d binary for that one... Tried that aswell, so I saw the full unfiltered, stippled glory of it all. I guess Matrox figured there were two ways to go about it, for a cheap all-in-one solution like the Mystique: You can either make it look fancy with bilinear filtering and all, and end up with something like the S3 Virge, or you could just try to make it fast.
The PowerVR made it look good AND fast.

Funny that you say the code was so well-written. I never considered the software renderer itself very fast, and it also was rather unstable iirc... wobbly polygons etc, didn't seem to have any subpixel correction. Didn't quite compare to the Quake engine in terms of quality. So I never thought the 3d engine was all that special.
But I guess at least the internals of Tomb Raider were well-designed, since it was ported to all major 3d accelerators at the time, and both quality and performance were great.

[I.S.T.]

You're forgetting Tomb Raider was originally a Saturn game...

[Scali]

Originally Posted by I.S.T.

You're forgetting Tomb Raider was originally a Saturn game...

No, I just don't see what that has to do with the software renderer for the PC version.

[Ailuros]

Originally Posted by TEXAN*

From what I've heard, SEGA is developing a HD casual system to compete with the successor to the Wii.

They might use the SGX543MP16 or they may just base it off the RINGWIDE or RINGEDGE.

Really?

[Simon F]

Arghh! Hit the key combo for exit 90% the way through a long reply so here comes a shortened version...

Originally Posted by Grall

Surely there would be no harm in revealing the name of the company/game after so many years... Don't be coy with us please, nobody will mind/care.

I'm sure they would but, to be honest, apart from it being a racing game that also had a 3dfx port (as we were using that as the starting point) I can't remember the details. The source code was a mixture of C and ASM with, IIRC, the 3dfx calls done with assembler. Also, the draw calls were distributed throughout the code.

Btw, I loved my PCX1/2 boards. Unfortunately I fried the PCX2 by trying to overclock it, but it did go (somewhat) faster with a slightly higher frequency oscillator

IIRC the NEC ASIC libraries turned out to be fairly conservative so I think over clocking possibly not that difficult. I think I also replaced the crystal on mine (but I had a socket for the crystal).

Originally Posted by Scali

Funny that you say the code was so well-written. I never considered the software renderer itself very fast, and it also was rather unstable iirc... wobbly polygons etc, didn't seem to have any subpixel correction.

Well, the TR code was all in C, and IIRC the polygon draw calls were restricted to only one or two functions making that aspect fairly simple to replace. Of course, there was a lot of other things that had to be done such as adding texture format conversion (palette to 16bpp + MIP mapping), identifying opaque/translucent sub-textures in each atlas, dynamic texture management (as it wouldn't all fit into the texture memory), translucency sorting (as we replaced punch-through with alpha blending), and adding the water effects.

As for the wobbly polygons in the software renderer the original code had an integer transform pipeline which we replaced with a floating point one which gave sub-pixel precision and, on the new-fangled Pentium, was faster. We also noticed that the original code, curiously enough, was waiting for 2 vsyncs per frame which we promptly reduced to one. Perhaps it was a hangover from Console+TV days <shrug>

[Davros]

I think the game was P.O.D

[Simon F]

Originally Posted by Davros

I think the game was P.O.D

Definitely not. I remember that at least.

Actually, I think I've now found what it was but it would be inappropriate for me to comment further.

[Simon F]

Originally Posted by Simon F

... integer transform pipeline which we replaced with a floating point one which gave sub-pixel precision and, on the new-fangled Pentium, was faster.

The other thing we did was throw out the XY clipping as we, virtually, had an unbounded guardband on PCX.

[Captain Chickenpants]

Ah, the good old Apocalypse 3dx. I can quite honestly say it was directly responsible for me having the job I have today.
I was downloading updated drivers and spotted the vacancies link.

I would like to see the Cyberk screensaver revived.

CC

[Scali]

Originally Posted by Simon F

As for the wobbly polygons in the software renderer the original code had an integer transform pipeline which we replaced with a floating point one which gave sub-pixel precision and, on the new-fangled Pentium, was faster.

Yea, I recall some demos with early 3d acceleration which stuck to an integer pipeline, so you had wobble even with 3d acceleration
But as I say, there were also software engines that DID have subpixel correction. In fact, the first subpixel-corrected renderer I wrote was aimed at 486 and didn't even use floating point for it (was too slow on anything pre-Pentium ).

Originally Posted by Simon F

We also noticed that the original code, curiously enough, was waiting for 2 vsyncs per frame which we promptly reduced to one. Perhaps it was a hangover from Console+TV days <shrug>

Hah, that's interesting... Perhaps the software renderer itself wasn't as slow as it appeared then... Perhaps one day when I'm bored, I'll try if I can locate the vbl loops and patch them out, see what happens

[Simon F]

Originally Posted by Scali

Yea, I recall some demos with early 3d acceleration which stuck to an integer pipeline, so you had wobble even with 3d acceleration
But as I say, there were also software engines that DID have subpixel correction. In fact, the first subpixel-corrected renderer I wrote was aimed at 486 and didn't even use floating point for it (was too slow on anything pre-Pentium ).

Oh, it probably did use fixed-point but I think the final projection rounded/truncated the fractional values back to integer.

Hah, that's interesting... Perhaps the software renderer itself wasn't as slow as it appeared then... Perhaps one day when I'm bored, I'll try if I can locate the vbl loops and patch them out, see what happens

Good luck.

[Davros]

Originally Posted by Simon F

Actually, I think I've now found what it was but it would be inappropriate for me to comment further.

Oh come off it it was 14 years ago