CB0 Survey Results

I’d like to thank all of you who sent me emails telling me what you bought and how you did!

Conclusions

1. The supply of cB0 chips out there has improved quite a bit. There’s still reason for some caution, though, some of you are still getting cA2 processors. Quite a few of you picked out your own CPU from a local store or computer show. Quite a few bought at PC NUT (and said quite a few nice things about the whole process). Others bought at places which offered cB0 steppings like atacom.com or new places like cpusforless.com. A few bought at googlegear.com blind and seemed to be alright.

(Buyer beware: ALWAYS check Resellerratings.com and read the comments about the place before you buy, quite a few people are not happy with customer service at atacom.com, and googlegear.com is a good deal worse. As for new companies, I know of two I’ve bought from that no longer exist).

I still think you shouldn’t order blind, simply because if you buy a 650 or 700, and you get a cA2, you probably won’t be able to run at 133 MHz. I’ve noticed the prices at my local stores have dropped quite a bit over just the last week. It does seem, though, if you want to play Russian Roulette for a cB0, there’s fewer bullets in the gun now. Ask yourself, “How lucky do I feel,” but if you lose, don’t blame me. 🙂

2. There doesn’t appear to be any significant difference between FCPGA/SECC2 OEM/Retail.

3. There aren’t a lot of 150 MHz wonders out there in 650 and 700 land. Most of you are getting 133 MHz from the processors, but about 20% aren’t, and most aren’t getting a whole lot more than that.

There seems to be some reasons for this. A lot of you are trying to do this with old PC100 memory. This is not a good idea at all if you’re hoping to get to 140 MHz or better; people aren’t. I see so many times that people first buy the processor, then they upgrade the RAM because they can’t get far enough.

It doesn’t make sense for most people to spend another $150 or more to get maybe 5-10% more performance. The Celerons are getting cheaper. I think a lot of people should start considering
them or Durons as an alternative, unless they have a really good reason to need the Coppermine, or they don’t plan to upgrade for quite a while.

Quite a few of you are trying to do this with the standard Intel heatsink/fan, this may be a hot idea, but not a good one, especially in 700 territory.:)

4. Don’t be at all surprised if you have to raise voltage to get into the 900s, especially the higher 900s. This is particularly true of the 700s. This shouldn’t be surprising, even Intel has to raise the voltage to get the 1 GHz to work reliably. I wouldn’t be worried about voltages up to 1.80 or 1.85, but this obviously increases the need for very good cooling.

5. Don’t assume you’re going to hit 1 GHz with one of these chips with even the best air cooling. You might, but you probably won’t. Jumping more than a hair above 1 GHz is Peltier territory. I don’t expect the 750s to do any better; I would not take 7.5 x 133 as a given.

On the whole, the results are about what I expected from the cB0 chips. If you’re building a new system and going with modern parts, you’ll probably do a bit better, if not, a bit worse.

Thanks again!

Email Ed


Be the first to comment

Leave a Reply