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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Joe NYC who wrote (85754)	7/24/2002 9:03:08 PM
From: wanna_bmw	Read Replies (1) \| Respond to of 275872

Joe, Re: "Just because it's there, we think people are going to use it," said Mark de Frere, AMD's Athlon brand manager. As usual, AMD is relying on the hope that if they build it, the development community will come. This approach didn't work for 3DNow, nor has it ever worked for any type of architectural improvement out there (that I can recall, at least). You need exceptionally heavy lifting to get people to embrace a new technology. Just because 64-bits has a cool marketing prospect, it doesn't mean that real software developer engineers will see it that way. I am again losing faith in AMD's ability to provide a solid adaptation plan for x86-64. Re: ClawHammer's 64-bit architecture--which gives the processor the ability to process twice as much data per clock cycle as current Athlon chips And now, people are already starting to believe that 64-bits, being twice as wide as 32-bits, automatically means twice the performance. AMD isn't managing expectations at all. Quite the contrary, they actually seem to be letting the public make any sort of optimistic claims they can about Hammer's conceivable performance. That works just great as a Hype Engine, but sooner or later, people are going to face reality, and Hammer will be a disappointment. Not that it will be a bad chip, but it can't possibly meet the exceedingly raised expectations that AMD has let aspire around their design. This is a classic marketing blunder, and it has blown up in similar situations in the past. There's a good chance that it can blow up in this case, too, if AMD isn't careful. wbmw

To: Joe NYC who wrote (85754)	7/24/2002 9:08:57 PM
From: hmaly	Read Replies (1) \| Respond to of 275872

Jozef Re..AMD fielding 64 bits for PCs news.com.com Think this sentence epitomizes the importance of Hammer. Because of ClawHammer's 64-bit architecture--which gives the processor the ability to process twice as much data per clock cycle as current Athlon chips and allows consumer desktops to offer significantly larger amounts of memory than they do now--AMD asserts that the chip will usher in a new class of high-performance games and other complex applications such as video editing and voice recognition. How much more data 64 bits can do per clock is up in the air, but doubling. However, if Hammer is close to doubling the IPC of NW, you can bet that the consumers will assume, the IPC advantage is because of 64 bits. Which would get the mhz monkey off of AMD's back; and put a performance monkey on Intels.

To: Joe NYC who wrote (85754)	7/24/2002 9:13:09 PM
From: Monica Detwiler	Read Replies (1) \| Respond to of 275872

Speed bumps The ClawHammer will also offer clock speeds of 2GHz or faster, along with performance enhancements such as a built-in memory controller. This will give the chip a 20 to 25 percent performance gain over the current Athlon XP, Crank said. Wasn't the Clawhammer supposed to be introduced at a quantispeed of 3400+? A 3400+ should be 55% faster than today's top Athlon - 2200+: (3400-2200)/2200 = 55%. Yet AMD is now back pedalling to a "20 to 25%" speed improvement for Clawhammer over Athlon. Sounds like the Clawhammer is going to be a disappointment - no?

To: Joe NYC who wrote (85754)	7/25/2002 3:38:56 PM
From: Petz	Read Replies (2) \| Respond to of 275872

Scalability of P4 from 2.4-->2.53 shows Clawhammer will equal or exceed 3.4 GHz P4. The company has already sent tens of PCs fitted with the upcoming chip to game developers and multimedia software creators, said John Crank, a senior branding associate for the Athlon. This is part of Operation Rolling Thunder, AMD's campaign to introduce the chip, he said. Smart move. The article also affirms > 2 GHz for Clawhammer. I didn't take the "20-25% greater performance than XP" as a negative. The ClawHammer will also offer clock speeds of 2GHz or faster, along with performance enhancements such as a built-in memory controller. This will give the chip a 20 to 25 percent performance gain over the current Athlon XP, Crank said. A 20-25% performance increase is exactly what is needed to give a 2 GHz Clawhammer a 3400 performance rating. The chart below shows the "scalability" of the CPU performance at the top of the P4 frequency range (from 2.4 GHz to 2.53 GHz) at the highest bus speed, using the fastest memory PC1066. It is calculated as a percentage, where 100% means that performance increases the same percentage as clock speed. I used tomshardware.com and following pages: Quake III Arena..................55% Quake III Arena1024..........34% Quake III ArenaNV15.........63% 3DMark2000 Pro...............60% 3DMark2001 Pro...............36% Newtek Lightwave.............94% XMPeg4.5.........................81% Sysmark 2002...................60% Sysmark 2002 Office........10% Sysmark 2002InterCC......102% LameMP3..........................88% (Athlon 2200+=P4 2.4/533) MP3 Maker.......................78% WinACE............................53% Cinema 4D XL R7............102% (Athlon 2200+ > P4 2.4/533) 3D Studio Max.................74% (Athlon 2200+ > P4 2.53/533) SPEC Viewperf DRV08....40% SPEC Viewperf DX07......32% SPECViewperfLIGHT05...69% (Athlon 2200+=P4 2.3/533) SPECViewperfPROE01...41% (Athlon 2200+=P4 2.4/533) SPECViewperfUGS01......25% (Athlon 850(!!!!) > P4 2.53/533) Comanche 4.....................74% If we ignore the 6 of 21 benchmarks where the Athlon 2200+ already outperforms a 2.33 GHz P4 with fastest bus and fastest memory, the average scalability of the P4 Northwood core is 58%. This is at 2.4 to 2.53 GHz. From 2.5 to 3 GHz it will certainly be worse, 50% at the best. A 2 GHz Athlon would already have performance equivalent to the 2.4 GHz, 533 FSB, 1066 RDRAM Northwood. Add on 20% performance boost to that and the P4 would have to be at 2.4 x 1.4=3.36 GHz. Using Crank's 25% number you would get 2.4 x 1.5 or 3.6 GHz as the equivalent P4 speed. Petz