Questions for Gary T (yeah, that Gary T)

[Typedef Enum]

Hey fellas,

For those of you in the dark as to who I'm referring to...think 3dfx. He was the CTO @ 3dfx back in the day, and is since been employed by nVidia.

I have arranged a little Q&A with Gary Tarolli, and I'm asking for suggested questions.

About the only thing that won't be asked is/are things pertaining to 3dfx. Things like opinions on current/future hardware are fair game, etc. Something like, "Hey, when are we going to see some Mojo action!" would probably be turned down

I had been thinking about doing this for a while, as I was bored some months ago, and decided to go back in time, and read some of the old interviews...I found myself really engaged by what Gary had to say, and how accurate (in many ways) his responses turned out to be...

Anyhow, enough of the babble...If you have a question you would like to have answered, leave it here, and I'll add it to the pile.

 

[Ailuros]

I don´t even know what Tarolli´s position at NV is right now. I´m afraid though that all the questions I would like to ask, he wouldn´t be able to answer due to NDA restrictions. Maybe I´ll come up with a question that he will be able to answer....

 

[Typedef Enum]

Yeah, I know...There are a lot of questions that I wouldn't mind asking him about 3dfx too...But I also realize that it's a sensitive issue, given the fact that he now is employed by nVidia.

I think things like...opinion on DRAM technology, multichip philosophy (has it changed...), bandwidth...better Antialiasing techniques..etc. would be definitely fair game.

By the way...I'll have to get clarification on this, but I don't think his position has really changed all that much in going from 3dfx->nVidia. I'll clarify that tomorrow.

 

[Ailuros]

Hmmm then ask him how he feels about TBR in general taking under account future prospects. (I figure he had a very close look to what Gigapixel brought along).

It doesn´t have to be PC centered entirely. Since NV is dealing with consoles with the Xbox, how much it would be an option for NVIDIA in the console market to put GP patents into use, especially after the recent problems that Microsoft brought up what cost concerns.

 

[Reverend]

Gary is officially known as "3D Architect" at NVIDIA. He probably won't be able to answer questions about NV20/NV25/NV30 since he isn't too much "in the know" about these products. He may not even know Cg in depth.

Instead, asking him about specific 3D technologies would definitely spark his interest. And maybe about general 3D gaming stuff.

I have a question for him though :

When's he getting rid of that beard because beard is hard to render.

 

[Dolemite]

How about the nVidia position on embedded memory, or simply what role he sees for such technologies in the industry in general.

I dunno. That was the first thing that came to my head. All these interviews are pretty restrictive as to what can be asked/answered, which is a constant source of frustration for me. To me the most obvious and interesting question would be when we'll start to see 3dfx tech in nVidia chips and what form that will take, which has been asked and answered in various ways. I figure he would the guy to ask about that, but again I doubt whether he would really be at liberty to say anything interesting about it.

Other issues to consider asking about: increased color precision, pixel/vertex shader standards, OpenGL 2.0, interface specs beyond AGP.

 

[bbot]

Does Nvidia plan to design the GPU for the next version of Xbox (Xbox2)? Sony have said that their next console (PS3) would be 1000x more powerful (presumably this refers to the computing power of Sony's next Emotion Engine) than PS2. Are you confident that Nvidia will be able to design a GPU as powerful as, or even more powerful than Sony's next EE+GS? Do you worry, however, that Intel won't be able to design a CPU with powerful enough vector units?

 

[psurge]

1.) Having worked for a company with a very different approach to 3d (TBR) from the usual IMR approach, and now working with a company known for producing IMRs, you must have a fair amount of knowledge on how each of these architectures was to evolve in the future. Has seeing a very different architecture from what 3dfx was working towards, and merging with it, changed your view of the optimal accelerator? Has it opened your eyes to new issues/new ways of doing things that are better than what either the IMR or the TBR approach had to offer?

2.) Do you foresee threading technology such as SMT (hyper-threading in the p4) having much relevance in the GPU market?

3.) As graphics chips become more and more programmeable, some inspiration is sure to come from CPU/DSP design. Which of the parallel CPU types (SMT, CMP, a stream processor such as Imagine, or a network of simple cores such as RAW do you think is best suited for 3d rendering?

 

[Pete]

Definitely ask if nV would consider a multi-chip or SLI solution. Wouldn't it have been to their benefit, with the NV30 rumored to arrive late, to be able to just tell people to buy another GF4 and SLI for double the performance? That way chips would be in produciton longer, OEMs would make more money as their product cycles would lengthen, and consumers could choose either reasonable performance now, and more later, or extreme performance right off the bat, for an extreme price.

I thought 3dfx made quite a bit of money off the Voodoo 2. It surprised me that they didn't continue that dual-card mentality. Particularly now that dual-monitor PCs are almost commonplace--the benefits of two separate 2D engines, as well as two 3D engines that would be smart enough to operate as one (via an SLI cable), would be fantastic. 3D speed, and the ability to simulatneously do some 2D work on your other monitor using the extra 2D portion of the second "GPU." Would the engineering hurdle be too high for such a solution? (They could always offer a dirt-cheap model without SLU capability, for the mass market.)

 

[Reverend]

Okay, here's a real question from me :

=========

In a past interview, of which it was during your time with 3dfx, you said the following in reply to one of the interview's question :

"I am disappointed by the quality of journalism, including reviews, on the Web. The Web is great in that it allows everyone to be heard and communicate with each other - the ultimate free speech medium. But one of the problems it creates is that everyone can become an "instant journalist", resulting in a tremendous amount of misinformation and sensationalism. Unfortunately, it seems that sensationalism sells and that truthfulness of information is often judged by how often it's repeated and linked to, rather than its contents. Thus, misinformation that is sensational has a tendency to become credible. The root problem is that there is no quality control.

Getting back to video card reviews, I don't mind our products getting bad reviews when the review is done fairly and objectively and diligently. However, there are many reviews that are just shoddy journalism. They are not diligent in their investigation, they are biased in their views, and they don't attempt to show both sides of the story. My favorite example is that when we get beat by 3 fps (frames/second) we "get our clock cleaned", but when we lead by 3 fps in another benchmark, we "just barely beat" the competition. It's not so much the data that is inaccurate as the spin that is put on the data. We also go through great pains to request that reviewers benchmark certain modes that we feel are representative of actual game play - for example high-resolution with FSAA (full-scene-anti-aliasing) enabled, e.g. 1024x768 or 1280x1024. Yet many review still focus on 640x480 without FSAA. In general, I am very disappointed by how many reviews ignore FSAA or ignore the different types of FSAA. Even if a reviewer doesn't value FSAA himself/herself, they have a duty to report FSAA results to readers who do value FSAA, and trust me, there are many out there.

The primary job of a 3d video card in terms of 3d is to run games. Therefore, I think your emphasis on games is warranted. For example, instead of being concerned about whether a card supports a particular feature or not, e.g. hardware T&L, just review the games. The final result should be judged by the overall user experience, not whether a function is performed in hardware or software. If software can perform a function faster than hardware, more power to it! In addition, some reviews focus only on one game or one benchmark, which isn't very thorough. It wouldn't be bad if this was pointed out in the review, but most times this is simply ignored and conclusions are drawn based on a fairly poor sampling of data. Driver and card stability, image quality, and smoothness of game play should also be taken into account. As you state, the overall experience is what matters, individual game benchmarks are just one part of that."

From journalism to quality of video card reviews, to the recommended focus on FSAA, to the emphasis on games instead of video card technology in video card reviews... has any of what you said in that interview changed?

=========

 

[Brimstone]

Do Voxels have an important role to fullfill in future GPU's? If yes, what are some of the pros and cons to voxels from a 3d architects point of view?

How harmonious is Ken Kutagragis (Sonys) vision of 3d graphic architecture compared to Nvidias?

 

[Typedef Enum]

Keep them coming...real good thus far.

Reverend, do you have a link to that interview? It seems like a fair question to ask, but due to the length, might be more apporopriate to summarize + post link...

 

[Vince]

How harmonious is Ken Kutagragis (Sonys) vision of 3d graphic architecture compared to Nvidias?

It's Ken Kutaragi and ask him if he believes that RT Grid computing is possible.

Ohh, and maybe a question concerning a more philisophical look at 3D advancement such as with computing power increasing and the ability to draw more and more polygons untill the point at which every pixel can sustain a triangle and you become more or less resolution limited, then what.

Maybe a question concerning the increase in programmability thats afforded by the rise in transistor budgets and whether or not he feels that hardwired implimentations are a thing of the past that will be replaced by what? Architectures that are fully programmable (100% SIMD) or a hybrid such as the P10, or what?

 

[Mephisto]

As Gary designed the Glide API (AFAIK), I would like to know his view on the upcoming high level 3D APIs like DX9/DX10 HLSL, Standford Shading Language, OpenGL Shader and OpenGL 2.0. What do we need for a real-time Renderman SL? Do we need automatic multipass of complex shaders?

 

[Reverend]

The interview is at http://www.3dgamingnews.com/3dpulpit/Articles/tarolli_2000_8.html

If you want to quote that link, you may need to be quick as that link may not exist for very much longer. You may want to simply tell Gary that "In a past interview with Reverend and the Beyond3D crew, you said..."

 

[EFA11]

I am interested to know what his take on Motion Blurr etc is. Did the FX of the T-Buffer come too early for the public and developers, that caused them to be not widely used or are they simply FX that will be unlikely to find a place in the gaming arena. Where does he stand on T-Buffer Type FX and their use today.

 

[Colourless]

As Gary designed the Glide API (AFAIK), I would like to know his view on the upcoming high level 3D APIs like DX9/DX10 HLSL, Standford Shading Language, OpenGL Shader and OpenGL 2.0. What do we need for a real-time Renderman SL? Do we need automatic multipass of complex shaders?

Brian Hook was the original designer of Glide.

 

[It is I]

well...

1 Wll we see much back from the 3dfx rampage architecture in the new Nv 30?

2 was 3dfx mojo a result, or maybe a port from the Sega gamecube Blackbelt project to the Pc market?
I mean, The blackbelt project was about a GPU which was desined around Tiltled architectue, and had i thought 63 bit colors, don't remeber good.

3Are there maybe some oldtech papers about the 3dfx mojo?, and if so, could we see them?

4What where the most intresting things about the mojo 3dfx had planned to design?

5 about all those rumors about 3dfx starting as oanother company, like 3d power and some other comanies. Where they possible join ventures which failed, or where they just rumors?

6what is so special, diffrent from the FSA the vsa-100 used and that from the vsa-200?

7 where there more things that cuased 3dfx to stop besides thestopped 3dfx blackbelt project and the cycle missed 3d cards?
What about the rumors of microsoft wanted to invest money in 3dfx?

 

[Ty]

Brian Hook was the original designer of Glide.

He designed it solely by himself? Or did he just happen to work on its development (along with a couple of others I thought)?

 

[Geeforcer]

Re: well...

1 Wll we see much back from the 3dfx rampage architecture in the new Nv 30?

2 was 3dfx mojo a result, or maybe a port from the Sega gamecube Blackbelt project to the Pc market?
I mean, The blackbelt project was about a GPU which was desined around Tiltled architectue, and had i thought 63 bit colors, don't remeber good.

3Are there maybe some oldtech papers about the 3dfx mojo?, and if so, could we see them?

4What where the most intresting things about the mojo 3dfx had planned to design?

5 about all those rumors about 3dfx starting as oanother company, like 3d power and some other comanies. Where they possible join ventures which failed, or where they just rumors?

6what is so special, diffrent from the FSA the vsa-100 used and that from the vsa-200?

7 where there more things that cuased 3dfx to stop besides thestopped 3dfx blackbelt project and the cycle missed 3d cards?
What about the rumors of microsoft wanted to invest money in 3dfx?

About the only thing that won't be asked is/are things pertaining to 3dfx.