Well, we’ve left the full month given the last episode of NVIDIA’s as well as Intel’s ongoing open feud, though it looks similar to Intel has right away stoked the abandon once again (albeit inadvertently) in a paper presented during the new International Symposium on Computer Architecture. That attempted to debunk a “100X GPU vs. CPU myth,” though it also contained the tidbit which GPUs have been “only” up to 14 times faster than CPUs in using application kernels, that NVIDIA has some-more than the happily latched onto. In the blog post, NVIDIA’s Andy Keane says that it’s a “singular day” when the aspirant states which their technology is only 14x faster, and which he can’t recall another time when he’s “seen the association foster rival benchmarks that are an sequence of bulk slower.” Of march, he afterwards serve goes on to note that Intel’s tests were done with NVIDIA’s previous era GeForce GTX 280, as well as which the codes were simply run out-of-the-box but any optimization — but, still, he seems more than happy to accept this bit of “approval.” In Intel’s defense, however, a altogether finding of the paper (linked below) is that the performance gap in between a GTX 280 GPU as well as Core i7 960 processor is actually only 2.5X “on average,” which NVIDIA hasn’t highlighted for some reason.