So, what does this mean?
This dissertation is my opinion, and I have an opposing point of view I'll be presenting shortly. I believe that these new "boxed" processors will not achieve nearly the overclockability of the previous SL32A's. I've tested a total of 242 of the SL32A's and have had one single processor that would not overclock to 450Mhz.
While the SL2WM has also been a good overclocker, coming in at close to 90% in earlier weeks, it has always had a higher percentage that needed a goose in voltage, and the later weeks have dropped to about 75% success. It's logical to assume that those now being boxed for sale are from the later weeks. With production of the 366 and 400 in full swing, and these CPU's being cut from the same wafers, it only makes sense that the quality of the 300A core can do nothing but deteriorate, the later the chip was manufactured.
Now, on the other hand, I heard from the folks at Mushkin and, according to them, they have been selling the boxed SL2WM version of the processor for "several months with an overclocking failure rate that is close to zero (according to our customers)." Mushkin is a very experienced group and one of the best vendors on the web, so I am certainly not going to doubt their word on this, but...
It is highly possible that these were earlier weeks, prior to 46 when these seemed to turn, and that this is why the success rate has remained high. Personally, I hope that they are later weeks, and that they remain highly successful to the very end, but we shall see.
One final note before we move on. If your buying a CPU or motherboard/CPU combination from one of the vendors offering a 450Mhz guarantee, none of this really matters. The information in this, and previous reports, really only comes in to play if you're out hunting Celery on your own.
Now, a little about the Celery cooking process. Those of you have read previous Celery reports can feel free to skip this section, but I include it for complete and utter disclosure's sake.;-)
Cooking the Celery
I test each CPU under initial identical conditions prior to "burning in" my combos. I do an initial test run using the same Abit BH6, same RAM (single stick 64MB Micron, CAS3) and the same video card, a Trident 975, 4MB AGP. For those who are interested, this is a very fast 2D card that consistently scores better than 5 in Final Reality. Not very good in 3D, but for a business system, its a screaming bargain. I use an old 270MB Quantum hard drive (actually have 3 set up identically, so I can burn multiple combos at once). I use an old hard drive because I assume that these setups will be going in to a variety of situations with a variety of drives and that if itll work with this clunker, itll probably work anywhere.
I run a quick test at 450 using Business Winstone 97. Why such an old version, you might ask? Because Im not testing for performance, only stability. This works perfectly for that purpose. If it fails, I bump the core voltage up a notch until it passes. If the CPU wont do 450 or requires more than 2.2v core, I set it aside for later evaluation. After passing the first test on my evaluation board, I move the chip to the motherboard with which it will be mated.
I put each one through 6 full rounds of Winstone and 2 hours of Final Reality. By using these two tests, Im covering both the 2D business users and the 3D gamers. As an aside, for those who havent used it, Final Reality is not only a great test, it is visually stunning. One glitch at current voltage, and its bumped up a notch, until its completely stable. Because I believe that stability means having a little breathing room, I dont sell a chip at 450 if it requires more than 2.2v core to stabilize.