On Sun, 2003-08-10 at 13:26, David E. Fox wrote: > > 1. After reading this article ... > > http://seattletimes.nwsource.com/html/businesstechnology/2001436280_linuxgroup08.html > > Please put that back. :) I couldn't access it. :) > > > I am left wondering; where does Intel stand in the debate? Are they > > officially on the M$ side of things are are they attempting to play > > Even if that's relevant, AMD makes better processors. See tom's > hardware. AMD (ca 2001 when I put this box together, at that time > the Pentium 4 had just surfaced) is better and faster at the same clock > speed than P4. There are still serious issues like cache and pipeline > stalls in the P4, unless Intel fixed that since. If Intel is more > MS-friendly than they should be, I don't know. I'll stick with AMD.
This is an excellent decision and you are absolutely correct in what you are saying about the P4. The P4 had a bunch of error correction and lookahead prediction circuitry ripped out so that they could clock the chip higher. This amounts to retrograde evolution. Meanwhile the Athlon series has continued to build on past experience and refine it's designs instead of ripping them out in order to clock higher. The clockspeed == performance propaganda has been one of the greatest lies foisted on the gullible computer user public that I've seen. The lie is evident when you compare an Athlon running at 2 gig to a P4 running at the same speed. The Athlon smokes them every time. The ONLY way a P4 can get ahead of the game is by clocking higher; meanwhile because of it's inferior pipeline, cache handling, error handling, and prediction systems, it is wasting huge amounts of cpu cycles. The Athlon, by comparison, throws away almost no valuable cpu cycles because it's design gets the job done correctly on the first cycle. (generally speaking.) Therefore naturally you see more work getting done correctly in the Athlons than in the Pentiums at the same speeds; and this truth is always reflected in the benchmarks. When you see one or two P4's at the top of the Quake benchmarks as the highest clocking processors in the benchmarks turning out what look to be better numbers, you then need to examine carefully what performance you are getting per cpu cycle. It is a fact that the Athlon will come out on top every time. And in order to get that P4, you have to shell out a huge amount of cash compared to what you pay for an AMD. Most of the time the people that pump out inordinate amounts of cash for high-clocking P4's are using their pocketbooks more than they are their brains. One more thing I've noticed over the years. The AMD based equipment is vastly easier and more flexible to work with. It's more tweakable, more adaptable....just plain easier to work with. The processors are more overclockable. This is why the AMD's have a HUGE contingent of fans on the net, where whole websites have been devoted exclusively to AMD. There is a large population of overclockers that buy AMD almost solely on matter of principle. There's alot of trust involved between AMD users and their public. > I, > for one, don't want a computer that keeps track of my existence and > reports to MS - I know I exist, and don't need any help in that area. Of > course, I'm referring to the increasing possibility of DMCA-aware > computing, with processors that detect or refuse to run stuff.) > > I can sum up the SCO effort in 2 words: unjust enrichment. > > I read another article which made the point that at one time SCO (they > acquired Caldera, right?) was actively trying to market a "commercial" > Linux distro. I never opted for Caldera, but I read the reviews and > such. Now, they did much of that through GPL, no? Did SCO commit any > *non GPL* changes to the Linux kernel? No - they would have been > rejected by Linus. (Anyone got his input?) > > AIX is a totally different animal than Linux. The kernels are different. > How dare SCO tell IBM that every customer is now using an > illegal product. (we use AIX + Pick at work). But isn't that SCO's > argument? That Linux got "polluted" by source from AIX? > > Redhat, Mandrake, debian, etc. have not signed contracts with SCO to my > knowledge. I know I didn't. I can't be in violation of a non-existant > license. SCO will never see any money from me. Might as well tell them > that if Linux did not exist, millions of people probably would be > pirating SCO, or perhaps Windows. > > > neutral in all of this. I have never used an AMD processor and I wasn't > > AMD is superior. > > > license for software they shipped me under the pretense that it was > > 'free'? > > Was it shrinkwrapped? Did it have a non-GPL type license to it? > > There have been precedences in the user community reactions over the > years - involving en masse switching to the competition. For instance, > rememmber SEA versus Phil Katz? The result of that was thousands of > BBSes switching to ZIP, almost overnight. GIF licensing threats from > Unisys engendered a similar reaction, more recently. > > So if SCO wins, I'll switch to FreeBSD :). I'd rather not, though. > FreeBSD is fine, but I think it would require a lot more post-install > setup than Linux distros do ATM. I could be wrong on that - I haven't > tried it in a very long time. > > > -=Thinker > ------------------------------------------------------------------------ > David E. Fox Thanks for letting me > [EMAIL PROTECTED] change magnetic patterns > [EMAIL PROTECTED] on your hard disk. > ----------------------------------------------------------------------- > LX -- ����������������������������������������������� Linux Mandrake 9.1 Kernel 2.4.21-0.13mdk "Filter That, Beach!" --Lanman, MDK Newbie List ������������������������������������������������
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
