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: wanna_bmw who wrote (61105)	10/30/2001 4:03:50 AM
From: Dan3	Respond to of 275872

Re: Pentium 4 still has a lot of room left for improvement.... You can say that again. When is a new core coming from Intel?

To: wanna_bmw who wrote (61105)	10/30/2001 6:54:12 AM
From: Bill Jackson	Read Replies (1) \| Respond to of 275872

wanna, The P-4 is quite bad and does indeed need improvement. As for the hammer being better at the start that the old horse P4, well, it has miles to run yet. All we can do now is speculate and hope that AMD does it well this time too. Bill

To: wanna_bmw who wrote (61105)	10/30/2001 7:52:35 AM
From: combjelly	Read Replies (1) \| Respond to of 275872

"and as far as I could tell, there is little difference micro-architecturally from the K7" True, but that misses the whole point. AMD is doing the same thing that the DEC/digital/Compaq engineers were doing with the Alpha. The EV7 is basically an EV6 core with a built-in memory controller and interfaces for forming a glue-less SMP and NUMA system. Much attention has been paid to things like memory latency and the path through the chip for minimal latency on each hop to the different CPU chips. Sure, the details are different (EV7 uses DRDRAM, for example), but the overall concept is the same. EV8 was going to build on that by having a new core. Considering the ties that AMD had with those engineers, it wouldn't surprise me if AMD has a similar plan. Assuming the numbers that AMD bandied about at MPF are correct, the integrated memory controller makes a huge difference, and it's SMP and higher capabilities will blow everything else away in it's class. The SMP and NUMA capabilities are a gamble. If it catches on, it would reduce the pressure to make processors ever faster with decreasing gain. Writing code that uses threads is a lot more common than it used to be, a multi-Hammer system can take ready advantage of that. Intel is trying to take advantage of that also, that is what Jackson Technology is all about...

To: wanna_bmw who wrote (61105)	10/30/2001 9:58:14 AM
From: dale_laroy	Read Replies (2) \| Respond to of 275872

>I don't know about that, Bill. AMD showed their Hammer slides at MPF, and as far as I could tell, there is little difference micro-architecturally from the K7. I don't believe that Hammer will have anything but incremental benefits over K7.< My SWAG is that they will achieve at least a 20% increase in peak clock rate, slightly higher average IPC for single tasking, and significantly higher IPC for multitasking and multiprocessing. But you are right, they did not significantly change the micro-architecture, just increased and rebalanced the pipeline plus making a few other changes that could be made with Athlon itself.