Why So Much AMD?

I’m sure more than a few of you have wondered why we have shifted our emphasis so much from Intel to AMD over the last month or so.

Basically, there are three reasons:

Nothing’s happening with Intel at the moment

The cB0 stepping chips have reached optimal price/performance levels. You’re going to get whatever you’re going to get out of a cB0 Coppermine for $250 or less. It’s not like you’re going to rack up a higher speed if you buy a Coppermine 750 or 800 right now.

The same pretty much holds true for the Celerons, too.

It won’t be until the cC0 stepping begin to become available in late August or early September that we’ll see better speeds coming from the CPUs. At that time, we plan on getting one and testing that. A bit further down the road, we’ll see DDR boards meant for the PIII. We’ll cover that, too, as well as .13 micron PIIIs and whatever new motherboards they may require.

We haven’t abandoned covering Intel, just put it on the back burner until something important happens. We do hope to get a PIII system running at about a 1Ghz going in the next couple weeks so we can get some comparisons between it and the AMD competition.

With on-die cache, the AMD chips represent a potential alternative to Intel chips

It’s always best to know all your options.

Since the new AMD chips have taken care of a number of disadvantages we saw in the older Athlons, we think it’s important to explore these chips far more than we have in the past, and identify their strengths and weaknesses.

This doesn’t mean we’re suggesting you all go junk your current systems and go buy this. For most of you, especially those who would be upgrading from a pretty current system, or contemplating just a CPU upgrade, that would be silly. These are not the Second Coming, and the level of improvement may well
not be worth what would be a relatively small level of improvement.

These are not perfect chips. The most obvious problem is a lot more power consumption and heat. This is a huge problem if you want to use aircooling; it’s even proving to be a problem for Peltiers.

A Duron may well be better than an equivalent Celeron, but it may not be such a great deal for you if you need a 300W power supply and heavy-duty cooling to make it work.

The overall stability of the motherboards remain a question mark; the A7V doesn’t look bad so far, but it’s still early in the ballgame. The relative lack of FSB overclocking is another disadvantage.

So while we are aggressively exploring what these chips can do, that doesn’t mean we’ll necessarily aggressively recommend them, because . . .

We’re Not An AMD Website

We’re not knocking AMD websites. They provide a great service and a lot of information. However, a website dedicated to AMD processing might be a little less willing to say things stink than we would.

Sure, we like the Duron over the Celeron. Then again, we don’t mind connecting dots and hooking up water-coolers and Peltiers and soldering motherboards, either. 🙂 You may not be willing to go that far, and that’s fine. There are other reasonable factors besides raw performance. For many of you, buying a Celeron probably makes more sense than buying a Duron, even if the Duron is a bit better.

We want you to get the system that is best for your situation, and the best chance you have to do that is to know all the viable options, and pick which is best for you. That’s what we’re trying to do. It may look a little lopsided right now, but then again, it was pretty lopsided the other way before. Things will balance out in the fall.

A Future See-Saw

Over the next few months, we expect to see rough overall equivalence between the offerings of Intel and AMD. AMD chips will probably cost less, offer a bit higher Mhz, and yield a bit better performance in at least some areas. Intel chips will be a bit easier to overclock, may offer higher FSBs, use less power and do better in other areas.

This is not to say one side won’t grab an advantage for a while, but any advantage will be short-term. We see personal circumstances having greater weight than raw performance in most buying decisions.

Email Ed