washimul said: even if the raytracing is done on a single CELL it wont take more than 10 secs. shut up you ****** .......u wannna teach me something......just shut up.. I know a lot about CELL...........did u even read the links i provided. CELL outguns just about anything in heavy computational taks including FFT, TRE, CYPTOGRAPHY, STENCIL and LINEAR MATRIX MULTIPLICATION etc. first you have to know before posting. I know more than your entire ****** *****. just shut up and read the articles on CELL............produced not by illiterates like you but IBM, MC, UC etc. THESE BENCHMARKS reveal that CELL is around 30x more powerful than the high end pc cores available in the market. However in heavy computational tasks CELL clearly smash teh competitors by Bigger margins. http://www.cs.berkeley.edu/~samw/projects/cell/LBLTalk.pdf http://www.cs.berkeley.edu/~samw/projects/cell/EDGE06_abstract.pdf CELL murders even the X1E superscalar !!! ABOVE so no need to mention the pc cores. http://cag.csail.mit.edu/crg/papers/eichenberger05cell.pdf .......this is from MIT. NOTE: - MIT states that with proper compiler the performance of CELL could be boosted by 6--7x. although i doubt you wd understand anything!!!! http://www.mc.com/literature/literature_files/CellPerfAndProg-3Nov06.pdf http://www.mc.com/literature/literature_files/Cell-Perf-Simple.pdf .................a single SPE outguns intel EMT 3.6 by massive margins . The article is a little old . But even as of today if you compare quad-core with CELL +optimizers( iRT +octopilers etc).............u wd still have a magnitude of 30x. MEATHANDS u havent been reading any info i provided.NO i dont want to listen to your gar*** quotes.undefined CELL beats 7800gtx by 6:1 http://gametomorrow.com/blog/index.php/2005/11/30/gpus-vs-cell/ imagine this with iRT. the 256 gflops take into account the performance in HEAVY COMPUTATIONAL task. In AI, GI (less intrinsic algorithms) CELL alone would do 2 TF CELL alone can do raytracing heres the video http://www.youtube.com/watch?v=lr-R4bUZIQw ....................look at the photorealistic landscapes.......have you ever seen anything like that on any game released till date. NO u wd need 100 xenons to achieve that...lol so just wait for some time.............and witness the CELL power!!!! BY june we would have a debate over this. i f you want to argue any farther then you wd have to bring concrete proofs with Benchmarks.

Meathands originally said: Washimul, I don't expect you to actually understand this but I'm bored so what the hell.

washimul repsonded with: Nonsense, nonsense, nonsense, you didn't read my links (despite quoting directly from your links in my post), the ps3 will roxor everything, nonsense, oneElevenOneOneEleven.

washimul said: even if the raytracing is done on a single CELL it wont take more than 10 secs. shut up you ****** .......u wannna teach me something......just shut up.. I know a lot about CELL...........did u even read the links i provided. CELL outguns just about anything in heavy computational taks including FFT, TRE, CYPTOGRAPHY, STENCIL and LINEAR MATRIX MULTIPLICATION etc. first you have to know before posting. I know more than your entire ****** *****. just shut up and read the articles on CELL............produced not by illiterates like you but IBM, MC, UC etc. THESE BENCHMARKS reveal that CELL is around 30x more powerful than the high end pc cores available in the market. However in heavy computational tasks CELL clearly smash teh competitors by Bigger margins. http://www.cs.berkeley.edu/~samw/projects/cell/LBLTalk.pdf http://www.cs.berkeley.edu/~samw/projects/cell/EDGE06_abstract.pdf CELL murders even the X1E superscalar !!! ABOVE so no need to mention the pc cores. http://cag.csail.mit.edu/crg/papers/eichenberger05cell.pdf .......this is from MIT. NOTE: - MIT states that with proper compiler the performance of CELL could be boosted by 6--7x. although i doubt you wd understand anything!!!! http://www.mc.com/literature/literature_files/CellPerfAndProg-3Nov06.pdf http://www.mc.com/literature/literature_files/Cell-Perf-Simple.pdf .................a single SPE outguns intel EMT 3.6 by massive margins . The article is a little old . But even as of today if you compare quad-core with CELL +optimizers( iRT +octopilers etc).............u wd still have a magnitude of 30x. MEATHANDS u havent been reading any info i provided.NO i dont want to listen to your gar*** quotes.undefined CELL beats 7800gtx by 6:1 http://gametomorrow.com/blog/index.php/2005/11/30/gpus-vs-cell/ imagine this with iRT. the 256 gflops take into account the performance in HEAVY COMPUTATIONAL task. In AI, GI (less intrinsic algorithms) CELL alone would do 2 TF CELL alone can do raytracing heres the video http://www.youtube.com/watch?v=lr-R4bUZIQw ....................look at the photorealistic landscapes.......have you ever seen anything like that on any game released till date. NO u wd need 100 xenons to achieve that...lol so just wait for some time.............and witness the CELL power!!!! BY june we would have a debate over this. i f you want to argue any farther then you wd have to bring concrete proofs with Benchmarks.

Sintinel said: Washimul we have tried to be patient, but you are coming dangerously close to earning yourself a one-week vacation. Telling users who disagree with you shut up repeatedly is not how you argue a point, and only serves to degrade the level of discussion.

CELL outguns just about anything in heavy computational taks including FFT, TRE, CYPTOGRAPHY, STENCIL and LINEAR MATRIX MULTIPLICATION etc.

so just wait for some time.............and witness the CELL power!!!! BY june we would have a debate over this.

mrstickball said: Didn't you learn ANYTHING from getting banned? Jeez. I doubt Heavenly Sword and Lair are delayed due to massive graphical upgrades. Their being delayed because they can't get the bugs worked out of them.

sieanr said: Quake 3 with raytracing, playable on just about any modern PC with good framerates OMFG P4 IS TEH HAXOR PROCESSOR CELL IS NOTING!!!eleven!!!! CELL outguns just about anything in heavy computational taks including FFT, TRE, CYPTOGRAPHY, STENCIL and LINEAR MATRIX MULTIPLICATION etc. But it fails in many other areas, like branch prediciton. And too bad it's an in order processor but I guess out of order is too advanced and complex for CELL BM(tm), I'll stick to my pentium pro with "backwards" out of order execution. so just wait for some time.............and witness the CELL power!!!! BY june we would have a debate over this. Yes, because K8L will be out then and demolish Cell in almost every way imaginable except meaningless specs. BTW, Crysis looks better than any PS3 game or the Cell raytracing demo you linked to. EDIT: I had no idea pages now scale to posted pics. I thought it would just clip off the edges of the images like before, so sorry if this now wider web page causes problems.

The RPU is a fully programmable ray tracing hardware architecture, with support for programmable material, geometry and lighting. The RPU combines the efficiency of GPUs with the advantages of ray tracing. The instruction set of the RPU is GPU like, which is optimal for shading purposes. In addition the RPU supports fast ray traversal through an k-D tree using a dedicated hardware unit and recursive function calls, usefull for recursive ray tracing. To increase efficiency always 4 rays are handled in a packet and multi-threading allows for high utilization of the hardware units. A working prototype of this hardware architecture has been developed based on FPGA technology. The ray tracing performance of the FPGA prototype running at 66 MHz is comparable to the OpenRT ray tracing performance of a Pentium 4 clocked at 2.6 GHz, despite the available memory bandwith to our RPU prototype is only about 350 MB/s. These numbers show the efficiency of the design, and one might estimate the performance degrees reachable with todays high end ASIC technology. High end graphics cards from NVIDIA provide 23 times more programmable floating point performance and 100 times more memory bandwidth as our prototype. The prototype can be parallelized to several FPGAs, each holding a copy of the scene. A setup with two FPGAs delivering twice the performance of a single FPGA is running in our lab. Scalability to up to 4 FPGA has been tested.