> there is no virtual memory translation happening, which in theory, will significantly cut latency (and hence boost performance and efficiency). This means that there is one cycle to address the SRAM, so “this saves half the power right off the bat just by getting rid of address translation from virtual memory.”<p>In protected mode (i.e., what the kernel is using), will an Intel processor not also disable virtual memory lookup? Couldn't we just recompile scientific software to a protected mode environment to get those same benefits?<p>Also, I think it is more useful and fair to compare against a GPU than a general purpose CPU.<p>(As an aside, I don't see where the reduced latency gives such a big advantage. There will be latency anyway, so in any case your software has to deal with waiting in an efficient way (doing useful stuff in the mean time). Shaving off some latency will only help if your software design was bad to begin with.)