Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Comment: It's summer-time, no need to heat your office! (Score 1, Interesting) 156 156

Nah... the sweet spot is still on the 750Ti, which is a nice 60W GPU that has plenty oompf and costs just $180,-
Why would you want to heat your office with an extra 0.25KWatt heater if you don't live in Alaska?
Top Tip: Pass on this one, and take the 750Ti.

Comment: Fusion will save the day (Score 1) 343 343

People are overly concerned about nuclear fission due to radioactive waste.
Nuclear fusion, on the other hand produces no or little waste, and yields much more energy to boot.

We've done fusion before, with hydrogen bombs.
All we have to do is contain it in a powerplant, and all the worlds energy problems are over.

Comment: Re:Computable? Simulatable? (Score 1) 199 199

I think the quantum eraser experiment is a strong indication that our universe is (lazy) computed.

A photon shot at double slit is a wave.
But only if you do not measure which slit was taken.
Or if you do measure: only if you destroy the measurement result.

I think the universe does lazy evaluation on which slit is taken.
If the result turns out to be 'not used' then the computation is skipped, and the light is approximated with a wave.

So not only is it a sim, the extra dimensional higher order entity running the sim cares about speed of computation.

Comment: Same architecture, better process? (Score 4, Interesting) 353 353

Why not reuse the cell design: use the exact same chip, but manufacture it with current lithography technology, smaller structures, higher clockrate, more SPUs. It may do the trcik, and there is no new learning curve for devs. I have programmed SPUs, and they can do wonders if used correctly.

Comment: Re:rasterizing is on it's way out anyway (Score 1) 412 412

Rasterization is good enough for 0% of all rendering that requires a full solution to the rendering equation.
Primary rays are handled efficiently in ray tracing. Both primary and secondary rays are pretty much independent of geometry complexity. Ray tracing scales efficiently with geometry, rasterizing does not.

Also... for the last 20yrs, our screen resolution has been between 1 and 2 Mpixel: it does not change. In the last 20yrs, polycount went up enormously. In the end, ray tracing will prevail.

I'm still waiting for the advent of the computer science groupie.

Working...