A decent enough answer & thank you for the info :) Femme
civileme wrote: > FemmeFatale wrote: > > >Is there a LM 8 ver that is optimized for P3 machines? > > > >Or is it the ones with the 586 in their RPM names? > > > >Thx > >Femme > > > > > > > >------------------------------------------------------------------------ > > > >Want to buy your Pack or Services from MandrakeSoft? > >Go to http://www.mandrakestore.com > > > No, there isn't a LM 8 for P-III machines. At this juncture, it is > easily demonstrable that one loses more than he gains by optimizing for > 686 code. There is some evidence that optimizing for K7 can produce > some gains, but it is still being discussed whether the significance of > those gains is worth the effort. > > <rant> > Part of this is the compilers available and their limited use of the > code. But a large part of it should be attributed to ingenious > marketing departments who tell you that activating some trash codes > (which may have been there in slightly different form in earlier > processors) will get you into the internet instead of on it. In other > words extra instruction codes are not necessarily an improvement except > in the eyes of the sales staff. Computer architecture may have changed > in the automated digital implementation of instructions a great deal, > but, in the sense of defining instructions to make a computer both > edfficient and a pleasure to program, well that is still an art. And > adding useless instructions is a common technique for meeting deadlines > or improving sales. > > It is just about like 56x CD-ROMs versus 8x. The 8x are often faster on > single look-ups and definitely do not wear the media as much, but the > 56x gives users the impression that they have something smokin' (well, > maybe, but not the sort of smoke you want to see). > > I know that this is not something a lot of people want to face, but a > significant portion of this industry is sales hype. If you don't > believe that, look at the processor speed ads of a few years ago where > the system bus stayed at 66MHz and the pocessors got faster and faster. > A P133 laptop routinely outrperformed the P150 laptop of the same brand > and model except perhaps on number-crunching because the 150 was on a > 60MHz Bus and the 133 on a 66. SO the throughput of the 133 had 10& > more bandwidth to play with in accessing other parts of the system > besides the processor... And of course a little old company named Cyrix > beat the tar out of the P200 for real-world apps with a processor that > barely kept from burning itself up at 150MHz, simply by boosting bus > speed 14% from 66 to 75MHz. As consumers became wise to the gimmick, > the marketeers had to let the engineers boost bus speeds and today we > see 200, 266, 333 on the Front-Side Bus. But that doesn't mean the > management of most companies doesn't look for a neat way to play up a > cheap or even useless feature to sell more product... Look at the > "Hardware" IDE-RAID, which isn't hardware at all. > > </rant> > > OK I feel better. I hope it wasn't at the expense of making you folk > feel worse. > > Civileme > > ------------------------------------------------------------------------ > Want to buy your Pack or Services from MandrakeSoft? > Go to http://www.mandrakestore.com
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
