Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Tenchusatsu who wrote (118657)	11/20/2000 8:11:07 PM
From: Dan3	Read Replies (1) \| Respond to of 186894

Re: Anyway, there are no more processors coming out these days which can demonstrate substantially better performance without resorting to the compiler or ISA extensions. Athlon might have been the last. I'd amend that statement to include the cache added to P3 to make coppermine into a very decent chip. AMD is making noises that Palamino will demonstrate better performance without resorting to extensions. We'll have to see if it actually can do so when it is released. But if Palamino does improve over Thunderbird, then P4 will be the only new chip that hasn't demonstrated better performance without resorting to extensions. (unless you want to bring up Cyrix, Winchip, and transmeta - but lumping P4 in with those chips wouldn't be fair to P4, at least not yet) After 5 years, I really think everyone expected more from this chip. Dan

To: Tenchusatsu who wrote (118657)	11/20/2000 10:04:10 PM
From: Joe NYC	Respond to of 186894

T. By the way, you'll notice that even Herr Uberclockermeister is looking forward to SSE2-optimized software, now that AMD's Hammer series will support it. We talked about this a lot in the past. You think that Tom is pro-AMD and Anti-Intel. That's not true. Tom is pro-Quake, and subsequently for the company that will deliver the highest performance to him. You must have noticed that he was mildly positive on P4. Why is it? Quake (used to be Doom). He was the most pro-Intel guy since during K6 vs. PII, pro-AMD in Piii vs. Athlon, and now he is wavering, as Intel regained the Quake title. P4 got beaten in all the other 3D games, which explains the hesitation. And think about the huge flood of x86-64 optimized software when Hammer is released. Hammer will be bought mainly for it's superior 32-bit performance. 64bit will arrive eventually, but during the 1st year, Hammer will have to survive based on it's ability to run 32bit apps. You might have to narrow your definition of "integer-based" applications. Would hardware RSA encryption count? Pentium 4 w/ SSE2 will scream on this sort of app, thanks to 128-bit integer SIMD. What about Photoshop? Speech recognition? MPEG4? (The last two I'm not sure about.) I write custom apps that are used in real world, and I want my apps to work fast. To extrapolate from the benchmarks published on the web, these types of apps are P4s Achilees heel. Of course there are other areas where P4 will lead, but right now, it is a very narrow area where P4 may do exceptionally well vs. very broad crossection of apps where P4 does poorly. Intel made a bet that the direction of the computing is the narrow area where P4 does exceptionally well. Athlon is a much safer bet in that AMD didn't put all the eggs in one basket. The baskets and eggs seem like a good analogy. P4 and Athlon each have 10 baskets and 20 eggs. Intel put 11 eggs in one basket and one egg each in remaining 9, while AMD put 2 eggs in basket. The business apps are one of the baskets where Intel has only 1 egg, AMD has 2. I hope this fact will help AMD enter the commercial market, where palying of Quake, MP3 and MPEG is discouraged. Intel already demonstrated a "gee whiz" 2.0 GHz demo on 0.18u silicon. They plan to have this in Q3 of next year. AMD may come close, but they'll be resorting to a 0.15u half-shrink (which Intel doesn't have). When 0.13u comes around, Intel will be ahead of AMD process-wise and speed-wise, leaving AMD no other choice but to pin their hopes on SOI. That's one way to characterize the current situation. Another way is to say that AMD will have faster, smaller, energy efficient chips, and more of them thanks to the .15u shrink, .13u + SOI + Hammer core will "slam the door". But the truth will be somewhere in the middle. Joe