Celeron 366 and 400: the accessible speed
Overclocking: purifying more... or not!!
For the not initiated ones, the overclocking is a consistent skill in raising the clock speed over the face-value of the mike. This practice can be realized by the way or have been a victim of a trick, although the above mentioned it is quite strange; in any case, it contains certain risks for the mike overclockeado. Nevertheless, many people prefer to confront these risks (on the other hand avoidable if a suitable fan is used and of quality to refresh the mike) to obtain a more rapid mike to a very low cost.
Of the case that occupies us, Intel has made a very interesting mike to realize overclocking, since simply raising the nominal speed of the 66 MHz bus to 75, we would manage to convert a Celeron to 366 MHz into a Celeron to 412,5 MHz, obtaining a yield superior to that of Pentium II to 400 MHz nearly more of the third part of the price of this one, what turns out to be undoubtedly a tempter.
Conscious of this, Intel has promised that soon it will be impossible to do overclocking with his mikes (nice the boys: no?). In fact, for some time the Celeron and some Pentium II have the multipliers of the chip already fixed, in such a way that a Celeron Mendocino (5x66) only admits to 333 MHz a multiplier of 5x, what does that for overclockearlo (with pardon) it is possible only to play with the speed of the bus, for example putting it to 75 MHz (5x75=375 MHz)... if the badge leaves us, since someone only allow the official buses, 66 and 100 MHz (a riddle: do you believe that the badges you mark Intel they enter this group?).
The difference is that the new Mendocino to 366 and 400 MHz it is supposed that they are going to include, in addition to a fixed multiplier (5,5x and 6x respectively), a circuit that prevents the chip from working with a bus superior to 66 official MHz. Personally it looks like to me a very bad move on the part of Intel, but they are in his right...
Any way mikes still have not been seen with this limiter of the bus, for what if they hurry up in buying them still they will be able overclockearlos, in whose case the cheapest would recommend to take, 366 MHz that, and it to put to 5,5x75=412,5 MHz, what should not give excessive problems. The following step (5,5x83=458 MHz) can turn out to be too risky for the mike and the videocard (that also the overclocking suffers, for good and for evil), apart from the fact that the memory should be necessary PC100 (what on the other hand is the enclosed most advisable thing if it is not done overclocking).
And if the pernicious Intel promise is fulfilled, since we still have left the possibility of taking a Mendocino to 300 ó 333 MHz (that for the time being only are known by multiplier's limiter to 4,5x and 5x respectively, but not of bus) and to put them with a bus of 75, 83 ó even of 100 MHz. There is no doubt that the mike suffers, sometimes he needs a little extra voltage and this way they all nor are capable of obtaining it, but a 300 MHz Celeron Mendocino running to 450 MHz (the maximum that is possible to reach without safely destroying the chip) is an unbeatable buy in quality / price... although the mike warms up so much that it is convenient to have a very good, and better fan if it comes with temperature alarm.
To end, we go there with the economic considerations. The first thing, to emphasize that Intel does not want that the new Celeron seem so good as they are. And why? Man, because then the bargain finished them of selling mikes as Pentium II to 450 MHz to the fortune of 467$ (approximately 67.000 pts or 400 euros), and that in minimal quantities of 1.000 units.
In fact, Intel does not have the most minimal interest to make mikes of big cheap services as there are the Mendocino; the problem is in that he does not want to lose market share in favor of AMD either, and for it it is ready to what it is. The curious thing is that the mark Celeron gained itself to pulse a very bad reputation in the computer world when they did not have any cache memory L2, what there have dragging his brothers Celeron Mendocino despite his good yield.
But remember that although in the shops of computer science the Celeron it will be very cheap, fortunately that has not anything in common with his yield, so I was useful to buy more memory RAM, to put a better videocard or to buy this monitor of mark that so much excites him.
In particular, Intel sells the 366 MHz Celeron Mendocino 131$ the version for Slot1 and only 123$ the version Socket370, while the 400 MHz models sell 166$ and 158$, according to the type of connector. As we said already, the models Socket370 are cheaper, although the motherboard turns out to be limited seriously in his expansion capacities.
This topic is one of the most complicated: the election of the motherboard. To big features, two types of possible buys exist:
And from any form, unless we choose a badge Socket370 without possibility of reaching a 100 MHz bus, the memory PC100 100 MHz (SDRAM) turns out to be an essential buy, especially now when it costs practically the same as the 66 MHz SDRAM.
As for the AMD K6-2, the answer is sadly obvious: unless they lower drastically the prices, it turns out to be more profitable to buy a Celeron that a K6-2. It is true that the K6-2 can be something better than Celeron Mendocino in some conditions (to 333 MHz or less in office applications or with games optimized for 3DNow!), but in general lines his useful life is finishing, what does not wonder being on the point of going out the K6-3 and to a few months of the entry (probably rompedora) of the K7.
What is out of any doubt is that very few people requirirá to buy a Pentium II, and even those who need it should hope that Pentium III should go out or at least to the AMD K6-3, both more rapid than everything what exists at present on the market. The above mentioned: Pentium II has died... alive the Celeron!