Comrade,
I see the ignorant have arrived
I was just thinking the same thing. I would appear you missed the flow of my message, but it was sort of complex. Let's look at your comments:
," but is it cheaper to make than a comparable performance chip? " It has a smaller die than either the amd or intel chips, therefore almost by definition it it 'cheaper to make'.
Do you know what "comparable" means? You see, that was the entire point of my message. Transmeta is not a comparable chip. If you want to compare it in die size-performance that runs x86 (a rather narrow comparison), well, a comparable chip from Intel would be big--those first 366 MHz Celerons were large! (That's a joke) Intel and AMD fab their own parts, so their cost per die area is cheaper than fabless TMTA (and high volume helps costs, too). I don't know that AMD or Intel has a bin split low enough to match the Crusoe performance, but if they did I imagine you would be right. Luckily for them, they can sell the rest of their chips at much higher prices. I wonder what kind of margin you can get for a low power, low performance microprocessor, particularly when your only customers are your investors?
"wait for some weird benchmark to compare this relatively unimportant point" If you say so, but then why are amd/intel now trying to match crusoe's energy efficiency/ performance ratio?
Do you really think Intel and AMD were not already making lower power chips? (See StrongArm again, which exists for low power apps) Do you think Intel announcing a low-power part soon after Crusoe came out was due to them trying to match Crusoe? How long do you think it takes to design a low-power version of a microprocessor? These companies aren't trying to match Crusoe's power/performance; they are trying to lower power while maintaining performance, which is motivated by the (good margin) mobile market. They might be a little more vocal about it now that low power is a hot topic, but hell, that will actually help the Intel's of the world sell their chips to the better-educated and more-receptive audience. Intel should probably thank the Transmeta PR department!
BTW, have you seen what percentage of power is consumed by the microprocessor? I forget the breakdown, but the LCD draws more current than the microprocessor, so there is only so much impact to be made. . .
short if you like, but watch out in case it turns into a brcd or brcm. i would like to be long, but etrade didn't carry this one, so i'll just keep watching for now...
Personally, I think BRCD is also a big, bloated bastard of a stock, and you can ask Monday's BRCM buyers (at 220) just how much future growth means to them today (at 160). But, if you believe TMTA to be the next BRCM or BRCD (at their IPO, say), then you should have no fear picking it up now at a discount to its intrinsic value. Your opinion would mean be much more to me (and others, presumably) if you actually put your money where your mouth is.
Regards,
Steve |
