Ah, Gary Tarolli, Dr Voodoo, Mr T-Buffer.

Gary is one of the original founders of 3dfx Interactive and is currently its Chief Technical Officer/Chief Scientist (his official designation, although he's actually one of the 3dfx's foremost geeks, heh). He is one of the chief architects behind the original Voodoo Graphics, 3dfx Interactive's smashing debut success product and he continues to be behind every generation of the Voodoo-architechure line of products since then (Voodoo2, Banshee, Voodoo3, Voodoo4 and Voodoo5). He has been in the graphics line for the last 18 years and was the second to receive the EE Times' "Celebrating the Engineer" award in 1992.

Heard about the T-Buffer technology powering 3dfx's latest product, the Voodoo5? Well, that's "Tarolli" in the "T" - Gary is the smartie-brains behind the T-Buffer technology that enables such cool effects such as Full-Scene Anti Aliasing, Motion Blur, Soft Shadows and Reflections as well as Depth-of-Field.

I managed to wrench Gary away from his potion-mixing duties (hey, he's a Voodoo Scientist) into answering a bunch of questions from the whole Beyond3D crew. Let's hit it.

Gary Tarolli, CTO, keep us pointed in the right technical direction and subject myself to interviews.

Wavey: Exactly what have you been working on at 3dfx lately? You appear to be a bit of an enigma at 3dfx right now - we know you were directly involved with the original Voodoo1, but after that you appear to have slid away. The T-Buffer is obviously your concept, and that's gone into VSA-100 nicely; but what are you actually doing now? You broke silence on the Voodoo5 NG earlier this year to talk about Microcoded T&L units - does this mean you've had some direct input on the widely-and-likely rumored T&L part in Rampage? Can we expect more 'Tarolli'isms' on future 3dfx hardware...?

I worked directly on Voodoo1 and Voodoo2, writing lots of C code for simulation and verification - real engineering work! Afterwards, I did research into FSAA and other areas which helped define certain features in our next generation products. You will see the results of that work soon. I don't do any hands-on simulation and verification work anymore. Now I mostly review our product plans and research, and keep us pointed in the right direction. I am also getting involved in our licensing efforts into the brave new world of small devices, an area that I find very exciting. In addition, I run our Technology Advisory Board which plays a key role in defining our future direction. You may see a few more Tarolli'isms on future hardware, but you will also see many features that others played the major role in creating.

Dave: 3dfx's stock price had been very low lately. At one point in recent history it was a bit under $4. This, of course, is of serious concern for investors. Currently, the stock price seems to be hovering at the $4.5-5 range (note : with regards to 3dfx's stock price, this was asked at the end of September. The price may have changed since then. - Rev). What are 3dfx's plans for the short and long-term future for raising the stock price and returning to profitability?

Release exciting new products in a timely manner, that's the goal in engineering. Beyond that answer, you should talk to our CEO and/or CFO.

Kristof: 3dfx has been suffering from great looking yet unappreciated babies like the 22bit post filter and FSAA. Both were (still are - Rev) unappreciated because both are something you have to see and experience to understand, 1000s of words can not explain how much of an impact both technologies have. Has this problem of marketability of these features hurt your products' success?

Yes, I think our products' success has been hurt. I agree with you that these features are still not appreciated by as many people as we would like, and they are something you have to see in person to truly value. It's been a challenge marketing these features, but we haven't given up. The truth will prevail! We have recently added some great demos to our web site that show off FSAA.

Kristof: 3dfx seems slow in the driver department compared to the many "leaky" NVIDIA revisions. Wouldn't it make sense to have Beta versions around quicker (and more frequently - Rev) so people can use them even if they are a bit buggy ? Some people just like this, others who want to play safe can ignore these ? Maybe even include an easy switch program to swap between the official and beta driver?

3dfx values the quality of our software and we offer frequent driver updates to correct bugs and improve performance. If you want us to comment on the competitors' drivers I'd have to ask you if you have read all the numerous chat room complaints on the many problems people are having with them. I just read one today who finally gave up with someone else's card and bought a Voodoo 5500, here is what he had to say, "If nothing else seems to work, try plugging in a Voodoo 5500. Worked for me!". What was doubly rewarding for me to read was that he was thrilled about the FSAA image quality.