Maxwell Render | Official public forum

Intel Corporation researchers have developed the world’s first programmable processor that delivers supercomputer-like performance from a single, 80-core chip not much larger than the size of a finger nail while using less electricity than most of today’s home appliances.

http://www.intel.com/research/platform/ ... us|k8098|s

I would like about 100 of these implanted in my brain so I can render with Maxwell all the time.

Maxer wrote:I would like about 100 of these implanted in my brain so I can render with Maxwell all the time.

But then you'd need buy 200 licenses..........

Don't be sad... You should have enough licenses now to run at least one of these chips...

I only have enough for 3 chips.

Sounds impressive

Also on topic, some news from IBM.

Sounds like, it might be possible to have a realtime Maxwell shaded viewport

ASCI Red was the first computer to benchmark at a teraflops (1996). That system used nearly 10,000 Pentium® Pro processors running at 200MHz and consumed 500kW of power plus an additional 500kW just to cool the room that housed it. Although not a general purpose computing device, this Teraflops Research Chip delivers 1.0 teraflops of performance and 1.6 terabits aggregate core to core communication bandwidth, while dissipating only 62W.

I'm telling ya --- someone's been haning out with aliens. Moore's Law ..... hah, ..gibberish.

Mihai wrote:

i remember that story ......

We've got skynet by the ba$$s!

Maxer wrote:I would like about 100 of these implanted in my brain so I can render with Maxwell all the time.

With this much power, your brain will be able to remodel all the things you see in 3d and render them in maxwell, all in realtime of course!

well, let's do a little math with these TeraFlops.

i may not take all factors, but it shows my thoughts.

wikipedia says that a normal Pentium4 @ 3 GHz reaches around 6 GigaFlops.
Maximus' pentium 4 3 GHz takes for the speedtest around 4 hours.

one of these chips is a little faster (factor 1 x 1000 / 6) than the pentium, based on frequency and number of cores:
would be around: 86.4 seconds.
that's 1.5 minutes. if we compare now to a computer today with 8 cores (25 minutes) we see a speed increase of factor 15.

for the chip release in (i don't know) six years, this seems to me like totally moore's law.

100 of these chips would calculate the test in about a second. so we still are faaaar away from anything «realtime» with large resolutions, heavy geometry @ 24 fps !