50TH ANNIVERSARY OF THE COMPUTER SCIENCE DEPARTMENT AT THE UNIVERSITY OF UTAH

By 2030 we will have progressed to a few atoms per transistor in bulk CMOS, to wafer scale integration [3] and probably replaced most metal wires with free space optics [2]. We will see the first mass-manufactured SFQ [1] and RSFQ [4] superconducting Josephson junctions. Maybe we will see the first carbon nanotubes and superconducting circuits [5]. [1] ▶ this video [2] link [3] link [4] Rapid single flux quantum link [5] link

I think it would have helped if you linked to the timestamp[1] where he started talking to encourage discussion, 5h47m in for anyone wanting to watch it. I recommend at 2x. It's a very interesting presentation, from the history to the mechanics of how it works, to the possibilities. [1] ▶ this video It's interesting to hear about how it originated in a research group in Russia led by KK Likharev, and how the students were poached to the USA and are now at retirement age there. That it was funded by the USSR to out compete IBM superconducting and Sutherland thinks[2] his group did with SFQ . [2] ▶ this video Sutherland was interested in wafer probers for the purpose of SFQ research that work at superconducting temperatures, he asked where he could get one and was told four were shipped to China[3]. Relevant relating to how USA is falling behind possibly, and there are sanctions on nVidia sending the latest in GPU/compute to China, but this happened. [3] ▶ this video One of the most interesting topics I have been introduced to and a great presentation and Q&A (watched at 2x speed of course) on it.

My source for the story is indirect, via Steve Parker, who’s worked with Turner. ▶ this video Scrub to 3:52:19 for the Turner Whitted story. BTW, mostly unrelated, but scrub that video to 5:47:10 for an amazing talk by Ivan Sutherland (“father of computer graphics”) that is not about graphics (he politely refuses to talk about graphics anymore :P), but about the active research he’s been doing (at 85 years old) into Single Quantum Flux circuits (an alternative to CMOS). Also- Jim Clark at 4:48:40 Jim’s idea you mention is still correct & compatible with Turner’s idea. Jim’s point is that we only need to render finite pixels. You might need an x-factor more triangles than pixels because of sampling and depth complexity and secondary lighting, so 100M polys is probably in the ball park, as long as we can quickly pick the right 100M polys in real time… In a way, Turner was talking about a lower bound, while Jim is talking about an upper bound, albeit slightly different things but they are similar, both relate to how much compute is needed for real time rendering.

Recently Utah had a small commemorative conference where the photo in the picture was taken. Ivan Sutherland attended, and gave what I thought was the best talk, about nothing to do with graphics. He’s actively doing research on Single Quantum Flux circuits at 85 years old! [1] He’s trying to spread the word that the US is about to miss out on the next hardware revolution because we’re not paying attention to SQF. [2] The talk is awesome and worth watching. Anyway, I feel like this was a sneak peek into one of the biggest reasons why that early graphics team was so successful, Ivan’s amazing. Combined with some other magic ingredients, their environment was special, and I think, sadly, unreproducible. [1] ▶ this video (For Ivan’s talk, scroll to 5:50:00 - that’s 5 hours and fifty minutes in.) [2] link

2500 points. Wow. I haven't seen upvotes like this... ever. Unmentioned so far- I was watching the University of Utah's 50th Anniversary of CS sessions[1], and- apologies I forget who- someone mentioned that it was Carver Mead who largely promoted "Moore's Law" as an idea. Just some random trivia, to highlight how such a strong piece of identity around such a more dynamic & wide-ranging man came about. Hopefully not the "Douglas Engelbart: creator of the mouse" identity. [1] ▶ this video