Author: Steven Edwards
Date: 01:33:02 11/19/03
Go up one level in this thread
On November 19, 2003 at 03:45:47, Daniel Clausen wrote: >On November 19, 2003 at 03:21:36, Steven Edwards wrote: >>The other day I was able to visit an Apple Store to try out my chess toolkit on >>a dual 2GHz PPC G5 machine. Running a single thread instance of the toolkit >>showed that there was a nearly exact linear speed up of a factor of 2.5 when >>compared to performance on a 800 MHz PPC G4. >> >>Note that there was no real testing of 64 bit operations as the code was >>compiled for 32 bit mode and at this time, the G5 operating system Mac OS 10.3 >>still presents only a 32 bit space to the user. > >Where did you compile your toolkit? I would expect a compile on a G5 itself >should give you a few percentage of speedup apart from the linear speed up. (not >much though) I used the gcc compiler, the same under both G4 and G5. There's no difference in the generated binaries. I am currently using Apple's Xcode setup with gcc 3.3 that is the same (modulo Altivec output) for G3, G4, and G5 platforms. >>The Apple rumor mill suggests dual 2.4 GHz PPC G5 machines in early 2004 and >>dual 3 GHz PPC G5 desktops by late summer the same year. > >Well, the 3GHz G5 roadmap (at 3GHZ by the end of late summer 2004) was stated >from Steve himself. But as always, it counts when they're out, not before. And >with Apple, it counts when they actually deliver, not when they're officially >released. :p (although that probably holds for the top-models of AMD/Intel-based >machines as well) Well, the G5s hit retail very quickly after their announcement. If Virginia Tech handn't nabbed the first 1,100 dual G5s, then regular consumers could have had them a few weeks earlier. With IBM pushing the PPC 970 in their own machines, I expect supplies to be reasonably plentiful, and surely far more plentiful when compared to Motorola's less than stellar production record. >>If IBM can keep on their POWER architecture roadmap moving towards 8 GHz >>chips by the end of the decade, then the days of x86 dominance of the >>desktop market may be limited. > >As much as I would like that to happen, forget it... do you think AMD/Intel will >stand still? (well AMD maybe, after Intel has killed them...) And if the PPC >would be the fastest comp on Earth, it would not count since it's not running >Windows. (no flamewar please :) Of course, Intel will not stand still as long as they have competition. AMD has been a big friend to Intel users; without AMD they would be paying twice as much for chips that would be running half as fast. I wouldn't be surprised to see AMD as a second source for PPC instruction set CPUs in times to come. Intel remains tied down to supporting 20+ year old legacy non orthagonal CISC and this only adds to the bulk and power requirements (along with lower fab rates) of their CPUs. For Windows, and even Microsoft itself, each passing day of further development of Linux and other open source software works against costly and proprietary code. MS gets wounded in battle after battle in emerging software markets and it appears that they're losing all of East Asia and South America. Even in developed countries MS is facing pressure because of high total cost of ownership numbers and concerns about security.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.