What is a Duron?

Why is the Duron interesting?

Yield ofimático and professional

Graphic yield


And the Celeron?


Released page

Juan Herrerías Rey

Articles about microprocessors

What is... the microprocessor?

principal page
what is...
to update
Contact Us
forum and chat



The distribution of the present pages is prohibited without the express assent of the authors.

All the marks are registered by his respective owners.

For more legal details, pulsate in the following icon


AMD Duron: more for much less



Making his other virtues separate, the Duron stands out undoubtedly for his high possibilities of success in the overclocking (but as always, he should remain be said that THE OVERCLOCKING IS DANGEROUS AND NORMALLY IT ANNULS THE GUARANTEE OF THE MIKE, WE DO NOT ASSURE THAT IT IS SUCCESSFUL, DO NOT BECOME RESPONSIBLE FOR POSSIBLE DAMAGES, etc, etc).

As the Athlon, Pentium III and the Celeron, the Duron is a mike with the fixed multiplier; in case of 700 MHz that, to 7x (to these effects, there take 100 MHz of the "physical" bus, before doing "2x" or DDR). Nevertheless, in this case (and also in the Athlon) it can "to be liberated" the multiplier by means of a very simple trick.

... Trick that by no means we have invented, the merits go to other places as Tom’s Hardware and others. And again, we have done it, but it seems somewhat risky and unnecessary to us while the factory multiplier one could raise the speed of the bus supporting fixedly (which usually gives major yield, certainly). Detail of an AMD Duron 700 showing the bridges L1 to be joined to liberate the multiplier

Well, to the grain: the trick consists of joining a series of electrical bridges that has the mike in his right top part, called "L1". It would be necessary to use a pencil of repair of circuits, but a soft pencil of well sharp graphite can work (we have used that, but undoubtedly it is not the best thing). Eye, sharp good: the bridges measure less 1 mm wide and I would prevent how I was that some were joining others, we have not studied the electrical scheme of the invention but probably it would have pernicious consequences.

Well, next a few results with both methods, raising the bus supporting 7x original and changing the multiplier supporting the bus (of course, it is possible to do - and we have done - both things simultaneously, but we were tired of so many tests):

Comparative of the yield of the Duron 700 with / without overclocking

Selecting in our KT7-RAID a tenth of additional voltage we have come to 850 MHz (8,5x100) and 840 MHz (7x120) with an almost perfect stability and even partial successes over 850 MHz... that if it does not turn out to be obvious, is: an overclocking of 21 %, neither more nor less!! The comments remain...


And the Celeron?

Ejem... Well, undoubtedly, the logical thing in the comparative one on a cheap mike is to mention how he behaves with regard to the cheap mike of his competition. But the case is that we could not have obtained any Intel Celeron 700 on time (and without buying it, that we already buy very much), so we will give a few general ideas and will take given information of this big web site called AnandTech.

As for price, the Celeron is located in the same segment as the Duron, something more expensive but with slightly cheaper basic badges; again, eye with "too cheap" (with chipsets too integrated and/or little possibilities of enlargement), that for Celeron there are many.

As the Duron, the Celeron is a good candidate for the overclocking, and it has the added advantage of removing poquísimo heat, is refreshed very easily. But it has two problems:

  • for motives exclusively of marketing, it keeps on using a 66 MHz bus (Pentium III has it of 133 MHz, the 200 MHz Duron);
  • his cache memory is smaller (32 KB of L1 and 128 KB of L2, against 128 KB of L1 and 64 KB of L2 in the Duron) and of (worse) different design.

For it, at "low" speeds as 533-600 MHz it is a very interesting mike, but to 700 MHz or more his yield goes down. How much? Well, since according to AnandTech (and if he says it, surely that it is like that), a Celeron 700 is 23 % worse than a Duron 700 in games (average of 3 different games) and 14 % worse in office computerization (Winstone2000). Really, it is not that it is a micro evil, the fact is that the Duron is better and in addition Intel limits the yield...



After everything previous, little stays already for saying, but we can say it higher: the AMD Duron is a FANTASTIC, EXCEPTIONAL, INCREDIBLE mike! And thanks to whom? To Intel, undoubtedly.

... Because if it was not because AMD has so formidable competition, we doubt very much that it had never offered us this mike, and of course not at the ridiculous price to which it does it. It is the best mike in quality / price from the epoch of the overclockeables Celeron 300A/333, a perfect and powerful mike for almost any use (good, neither not for servants nor edition of professional video, but yes perhaps enclosed for not very ambitious workstations).

Well, already he knows: if you are going to buy a new PC or update the current one, make use of this circumstance, it is never known how much can last the good thing.


To return

Pulsate in the picture desplegable or in "Returning" - Sitemap - Contact us Best website of 2009 & 2010 voted by us are:
, , , , , ,,