Turing's thoughts on Merced
Can his theorem solve the Intel enigma?
Updated (The IA-64 articles in the last few days have spawned some interesting correspondence. If you'd like to contribute your thoughts, why not post them in the thread we've opened on our forum? Ed.) One of our Big Blue readers has penned some thoughts on how Alan Turing might have regarded the IA-64 architecture. Turing (1912-1954), for those readers who may not have encountered him, helped found computer science and in the Second World War cracked Nazi Germany's Enigma cipher. Here's what our IBM reader has to say: "Almost nobody has related Intel's proposal with the theoretical works of Turing. "The problem is that the only way to predict what a program will do is to run it (that's a consequence of the Turing theorem). A classic processor executes the program code and knows what happens with the current data (loops, jumps, etc). "It can infer locality relationship from its instruction flow and make on-the-fly optimizations. Meanwhile, the compiler of an non-optimizing VLIW processor such as the IA-64 can only look at the code beforehand and make assumptions about what kind of data that code will process. "And the Turing theorem predicts that the IA-64 pre-compiled optimization will always be inferior to a good on-the-fly RISC processor. At best, the precompiled optimization will yield an optimal path for a given set of assumptions, which may not be true on every run, since run-time data can defeat these assumptions. "I'd really like someone to look closely at the IA-64 architecture in the light of the Turing theorem. I understand that Turing's tumultuous sexual life make him a questionable reference in some Yankee circles, but the oh-so-British Register shouldn't have the same qualms. :-)" There's an Alan Turing home page here. ®