baum@Apple.COM (Allen J. Baum) (09/13/89)
[] In article <blahblahblah (John R. Levine) responds to my defense of the VAX as merely a product of its time, rather than a bad design >.... The architecture was predicated on a >technological bet that microcode memory would always be faster than regular >memory, and when that turned out to be wrong the penalty for the heavily >microcoded implementation that the Vax demands has become very large. >Also, it has the distinct flavor of having been designed by a bunch of guys >sitting around saying "yeah, that would be nice, we'll put it in microcode... >I don't doubt that the OS and compiler guys were in on the design, but a >little more scepticism about what the software guys asked for would have >helped a lot. Well, technological bet sounds a bit worse than technological prediction. They may have predicted wrong, but they based the design on a prediction, rather than just design it according to what was around then. And they did ask OS & compiler guys what they wanted. Both of these may be considered innovations, AT THE TIME THE DECISIONS WERE MADE. Now, they're trite. >For comparison, consider the considerably older IBM 360 architecture. They >made a few real goofs, notably 24 rather than 32 bit addresses... Actually, we could spend a lot of time dicussing /360 architectural goofs. Like, no PC relative branches. Like, no immediate operations... >There are relatively few instructions that nobody uses (MVZ and LNR, >perhaps.) I think instruciton frequency lists might prove you wrong. The top 10 insts. account for 90% of the executed insts., like all other machines. This leaves a lot of those instructions used almost never. >IBM has certainly had more success with really fast 360s than DEC >has with really fast Vaxen. They sell them for a LOT more money, too. Perhaps even DEC could come up with a fast VAX if it sold them for $2 mil. -- baum@apple.com (408)974-3385 {decwrl,hplabs}!amdahl!apple!baum