3 ms·
AMD Radeon HD 6990 already has over 1 TFLOPS performance double-precision, and there's no problem buying it, Intel is too late.
by r3demon 15y ago
AMD Radeon HD 6990 already has over 1 TFLOPS performance double-precision, and there's no problem buying it, Intel is too late.
- phamilton 15y agoYes there is no problem buying it, but have you ever tried programming in OpenCL? Complexity aside, GPGPU hits a big bottleneck when dealing with large datasets. There just isn't enough memory available on the GPU, and transfers to and from the device are costly.
- buff-a 15y agoI promise you that if you don't pay an equivalent amount of attention to data availability on these intel chips you wont see teraflop speed. If there's one thing that GPGPU did it was force idiots to have to thing about presenting data to the processor instead of leaving it all over the fucking place and letting the cache "sort it out" (read: "run really slow"). That said. I'll take one =)
- marcf 15y agoThere are AMD video cards with 4GB of memory these days and special cards have up to 16GB I understand.
- phamilton 15y ago16GB isn't adequate for the work I've done on them. Genome assembly ( an O(n^2k^2) process ) generally has 100GB of data, each segment of which needs to be compared against each other segment ( O(n^2) ), and when comparing two segments each datum needs to be compared to each other datum ( O(k^2) ). So while you can just transfer 4GB of data to the card at a time, it really doesn't cut it.
- r3demon 15y agoI'm sure Intel won't cut it either, PCI Express speed is limited, and RAM woudn't catch up with simple computation speed. Find a better algorithm.
- rayiner 15y agoFar more limited architecture.
- r3demon 15y agoBy this logic all RISK architectures are very limited, but they can compute. Parallel architectures have to be seriously limited anyway, you just can't access 100Gb data with 1000 parallel processes at the same time.