Live data from Hacker News

The 3dfx Voodoo1

fabiensanglard.net

131–140 of 241 posts

Re: The 3dfx Voodoo1

#131
post #54

Earlier quoted context omitted.

at least you didnt throw away Voodoo5 9000 https://www.youtube.com/watch?v=_3iHV0NvLPI

Did this card genuinely exist or is it some kind of joke?

It's an April's fool joke. Voodoo 5 doesn't support DirectX 9 so it can't run Crysis.

Re: The 3dfx Voodoo1

#132
post #101
post #95

The main reason for the success of 3dfx were drivers that let you talk to bare metal hardware registers instead of dozens of GL abstraction layers. The simplicity of it all is very impressive. This is something programmers absoluteley love. Hence many games supported it and it was very easy to write drivers. Maybe it is time to reimplenent the glide interface in an FPGA ang make a super powered Voodoo card.

I miss software rendering, games with that as an option are the only old games that still run on modern PCs.

OpenGL games should run fine, some old games that used static sized buffers to hold extensions (like all Quake/Quake2 games that dump the available extension string to the console and the console having a hardcoded buffer for each line) will crash it because of how long the extension string is nowadays, but most drivers have an option to limit the string size (AFAIK Nvidia and AMD drivers check the executable name and do it automatically for known faulty games).

For Direct3D games you must use a 32bit color mode otherwise Windows 8.x/10 will force software rendering which is very slow. Even better, use dgVoodoo2 which is a reimplementation of DirectX 1 to 7 (with a some bits of 8) in Direct3D 11 and provides much better compatibility (also gets rid of the Direct3D 7 2048 surface width limitation, making it possible to play games at 2560x1440 and up).

For Glide, dgVoodoo2 is also very good and you can "cheat" the game to force higher resolutions than what the game thinks it is running at.

I have a lot of old games and everything that isn't DRM encumbered works fine in Windows 10 using dgVoodoo2 and/or some game-specific hacks (Tomb Raider 1 for example is normally a DOS game that you can play using a Glide-enabled build of DOSBox but there was also a Windows version made that used an ancient proprietary 3D API by ATI - someone reimplemented that API and placed extra hacks in there for high resolutions and widescreen support).

Re: The 3dfx Voodoo1

#133
post #2

The Voodoo was, very literally, a game-changer. The difference in quality between Quake and GLQuake was astounding. Just a few seconds of gameplay was enough to get someone to make the purchase.

The two main issues that GLQuake has over software one are that the textures are squashed down to the nearest power-of-two and lightmaps are clamped in the "0..100%" range whereas the software rasterizer has a (sort of) "0..200%" range. You can see this in the comparisons in the site where bright areas are dulled.

Modern source ports have fixed both issues though (FWIW i recommend Quakespasm)

Re: The 3dfx Voodoo1

#134
post #67

I was the lead and pretty much only programmer for the Canopus Pure3D. Ours had 12 RAM chips for 6 MB of memory. 2 MB extra Woo hoo! We also featured a Chrontel chip for composite and S-Video output. I wrote our DirectX driver (performance enhancements) and other sundry software for the Pure3D (pictured) and Pure3D II. Pretty cool to think back on those days.

Hah! That is super cool.

I think about that card from time to time. The extra 2MB made me feel badass, though I'm not sure any games actually used it, did they?

The Pure3D and Pure3D II were the first and last times I ever owned best-in-class graphics hardware.

The Rendition-based card I owned previously wasn't half bad, but those early voodoo cards were really revolutionary.

Re: The 3dfx Voodoo1

#135
post #22
post #16

It’s crazy how fast stuff moved back then. Voodoo 1, to Voodoo 2 (both hits) to Voodoo 3 (meh), to Voodoo 4/5 (debacle), to bankruptcy was less than 4 years. In the same timeframe, we went from a 200 MHz Pentium MMX to a 1 GHz Pentium 3. What’s happened since 2015, by contrast? Many people still swear by their 2015 MBPs, the last model before the touchbar debacle. That Haswell based machine was getting long in the to…

The obvious comparison to the GPU arms race is the VR hardware arms race. Same number of companies rising and falling in about the same period. Carmack is at the forefront of both, too.

The problem is that VR performance is tightly coupled to GPU performance. Currently, adding more pixels requires an equivalent increase in GPU performance to maintain quality.

I'm prepared to believe that e.g. foveal rendering will give step-wise improvements without more GPU power, but it's not the same as the years of continual process improvements were for GPUs.

Re: The 3dfx Voodoo1

#136
post #67

I was the lead and pretty much only programmer for the Canopus Pure3D. Ours had 12 RAM chips for 6 MB of memory. 2 MB extra Woo hoo! We also featured a Chrontel chip for composite and S-Video output. I wrote our DirectX driver (performance enhancements) and other sundry software for the Pure3D (pictured) and Pure3D II. Pretty cool to think back on those days.

Loved this card, still have it! Need to build an old box to drop it in.

Re: The 3dfx Voodoo1

#137
post #132
post #101

Earlier quoted context omitted.

I miss software rendering, games with that as an option are the only old games that still run on modern PCs.

OpenGL games should run fine, some old games that used static sized buffers to hold extensions (like all Quake/Quake2 games that dump the available extension string to the console and the console having a hardcoded buffer for each line) will crash it because of how long the extension string is nowadays, but most drivers have an option to limit the string size (AFAIK Nvidia and AMD drivers check the executable name an…

Software rendering will only be slow until everyone has manycore processors with SIMD. The die area of threadripper isn’t far from a GPU’s die area.

Re: The 3dfx Voodoo1

#139
post #67

I was the lead and pretty much only programmer for the Canopus Pure3D. Ours had 12 RAM chips for 6 MB of memory. 2 MB extra Woo hoo! We also featured a Chrontel chip for composite and S-Video output. I wrote our DirectX driver (performance enhancements) and other sundry software for the Pure3D (pictured) and Pure3D II. Pretty cool to think back on those days.

Would love to know more. I recollect the Pure3D being by far the best V1 card. Because of this many gamers patiently waited for the Canopus version of the V2, expecting it to be a similarly significant leap up from the other OEM cards. IIRC it took ages to come out, and was barely different functionally (just smaller and with some kind of LED to indicate it was switched)? And people were pretty annoyed to have waited all that time not joining the V2 party :)

Re: The 3dfx Voodoo1

#140
post #104
post #71

Earlier quoted context omitted.

32-bit colour is still just 24-bit colour with 8 bits of stencil or whatever (or at least was back when 3dfx was relevant). The only company I know of that was doing higher bit depths was Matrox who had 10-bit colour iirc. I remember back in the day "serious gamers would never use a combined 2D/3D card". Then nVidia came out with their Riva chipset and suddenly 2D graphics cards weren't really a thing any more.

I had a Riva 128, what a piece of crap it was... I used to assemble gaming boxes for friends and delaying delivery a day or two to take a ride on the Dual Voodoo setups. Edit: well, according to Wikipedia it was somewhat competitive although drivers and support for rendering stacks was erratic for a while. I guess that’s what soured me up, and I was probably too n00b to have a clue

Huh... We had a Diamond Viper V330 or something and it was rock solid. Worked fine on everything I threw at it, seemed comparable with a Voodoo 2 in terms of speed/quality. Maybe it was the card vendor rather than the chipset?
Post reply on HN