Hatching Eggs . . .

Add Your Comments

Remember x86-64? The stick AMD was supposed to beat Intel to death with?

Then Intel got themselves the same stick, and the would-be beaters found something else to talk about. Yes, AMD is doing better now, but not because of x86-64. Indeed, AMD’s increased fortunes have come well after Intel switched.

Microsoft came out with a Windows XP for x86-64, but the world hasn’t exactly stampeded to get it.

So what did we end up? A roar, then a mouse.

Of course, some will say, really, all should say, “Wait a minute, x86-64 isn’t dead in the water. When Vista arrives, so will x86-64.” And this is quite true.

But were the hypemasters saying that two years, one year ago? How many of you bought CPUs back then at a premium primarily because of x86-64? What did you get out of it?

This is not an exercise in AMD-bashing, but to illustrate a much bigger truth: hardware and software run on two different clocks.

It’s a lot easier for one group to scratch some new silicon paths and crank out a zillion copies of it than it is for a zillion programmers to tell those paths what to do.

This is true for x86-64, it is true for dual cores, it is true for whatever shader X.0 is being touted. It’s even true for items like solid state drives.

Since big increases in raw speed are no longer an option, any advances are increasingly dependent on having a software infrastructure that uses it. Until that happens, all anybody buying something with the new feature is doing is paying for the privilege of sitting on eggs waiting for them to hatch.

This doesn’t mean you should never hatch eggs. If you buy a computer meant to last five years, and you know an egg is going to hatch in two, it might be wise to sit on the nest for a while.

What it does mean is that you shouldn’t pay a lot for the privilege of sitting on those eggs. If you can justify the purchase and price you pay for an item with some eggs without considering the eggs (or at least not too much), then the eggs are a fringe benefit.

But don’t pay hundreds of dollars more for something with some feature that will sit idle for most of its useful life.

Ed


Leave a Reply

Your email address will not be published. Required fields are marked *

Discussion
  1. I believe Ed was asking a rhetorical question, "What has x86-64 done for you lately"? People that ran out and bought an x86-64 CPU with the idea that 64 bit was somehow going to justify their purchase were fooling themselves. Possibly still fooling themselves. What exactly has a 64 bit CPU done to improve your computing life? The average newegg shopper doesn't buy a dual opteron, slap linux on it, and start some highly specialized application. It's going to take Windows and some application that can make use of that extra address space to make an impact on most people's computing lives. Running 64 bit MS Word isn't it, btw. 64 bit (on the desktop) has turned out to be like MMX. Lots of hype. Little payoff.
    Really the problem is that no one really has that much incentive to code in 64 bit yet. Its expensive to take advatage of it in man hours and not very many people in the RW actually use it so why bother.
    AMD 64 served its purpose. Athlon 64s are good chips....buts its a lot easier to say to your average layman "This chip is better because its 64 bit instead of 32 bit" instead of "This chip is better because the ondie memory controller signifigantly increases the processors ability to do more work"
    One is just appealing to the simple "more is better" marketing thats worked so well in everything from SUV sales to Intels own P4. The other is a more tangible but harder to define concept.
    I'm not saying 64bitness is bad...I just think intel was correct when they said no one needed it right now. To bad for them no one really knows what they need right now!
    Yeah, I agree with the whole user thing. The vast majority of people probably think AMD's chips being better is a result of 64-bit, instead of an on-die memory controller and more efficient core architecture.
    It's easier to sell chips that way.
    The rest of us get 64-bit capable sempron 1.6s and pump them through the roof.
    I think the argument is pointless. Who here has chosen AMD over Intel based only on 64 bit? Not me. It never even crossed my mind. I bought AMD because I got more performance than an Intel that cost $100 more. I could care less that AMD had 64 bit before Intel, because they still were better without even using it.
    There are benefits to be reaped from 64bit, even on the desktop. I don't know if it's due to a differant compiler or what, but I definatly felt a differance in general snappiness under 64bit windows.
    In my opinion Amd did well in starting the transition, as it will take a long time (due to other parties).
    So no direct gain for now (ditched 64bit windows after 2-3 months due to lousy driver support).
    I think the argument is pointless. Who here has chosen AMD over Intel based only on 64 bit? Not me. It never even crossed my mind. I bought AMD because I got more performance than an Intel that cost $100 more. I could care less that AMD had 64 bit before Intel, because they still were better without even using it.

    Totally agree. 64-bit was the least of my concerns too. However, I did start out with the 64-bit Windows and eventually dropped that too. I wasn't able to run hardly anything on it. It just became a huge pain and wasn't worth it. I'm still happy with my X2 though regardless of 64-bit capabilities. More than anything it was THG's AMD vs. Intel stress test (here) that finally won me over for AMD. After seeing the results of that there was no doubt that AMD was the better choice.