Since we have commented one day, with the current badges the difference of yield between badges equipped with the same chipset is usually invaluable, being more important to state the quality and the stability of the badge (or the absence of both). Of nothing it serves to run very much, if suddenly we shock ourselves against the wall of a "hang" of the system...

Nevertheless, in this case we have two innovations that they can affect to the yield: the presence of the checker UltraDMA66 and the use of more than one mike. Will there be a lot of difference? Will it be invaluable? What will be, be?

The configuration used in the tests has been the following one:

• Microprocessor (is) Celeron PPGA 400
• 64 ó 128 MB of memory DIMM PC100
• Hard disk UltraDMA66 Quantum Fireball CR of 8,4 GB
• Hard disk UltraDMA33 Seagate Medalist 8420 de 8,6 GB
• Graphic card Intel i740 8 MB SGRAM - Drivers v. 2.1.
• Resolution 1.024x768 with 16 bits of color (65.536 colors)
• Operating systems: Windows 98 and NT 4.0 with Service Pack 4

For the tests mikes Celeron have chosen so much for his excellent relation quality / price, as simply because the BP6 neither admits others (is not even missing that it does to him, in my opinion). It is Celerons in format PPGA ("Socket 370"); to the being the BE6 a motherboard Slot1, we have used a PPGA converter to Slot1, precisely the Slotket of the same company ABIT (the ABIT catalog cannot be enormous, but it is full of useful products). This class of aparatitos costs approximately 2.500 pts, and often they are worth it.

The results are interesting; first, both present a stability of the first category, even with overclockings reasonable and them 35ºC of the summer of Madrid. The tests used Winstone are not at all light, one treats as hundreds of gentle of the most used programs of the market (Office, WordPerfect, Netscape Navigator...) that are executed during at least 15 minutes... and like that for hours. In whole there have been 4 days of continued tests, crushing both badges, connecting and disconnecting mikes and discs, living by means of pizza... through the truth, have supported it better than I.

It is a question of average values of both badges, since the differences in this test were very scarce, slightly logically because they are based on very similar designs. The UltraDMA66 offers an increase of the scarce yield, only 3,2 %, but there is not doubt that in the future it will be essential, when the discs of 7.200 are standardized rpm.

As for the use of two mikes in the BP6, only it is worth it if the computer is going to be used like servant or the application needs a lot of calculation potency. Since one sees in the following table, the applications that more make use of two mikes, all very professional ones, they can come to 38 % of progress, while in office computerization we obtain not even 2 %: