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: Umunhum who wrote (10525)	9/28/2000 12:23:36 PM
From: Petz	Read Replies (1) \| Respond to of 275872

uh-man, you are right, the CPU doesn't much care what kind of memory is sitting on the other side of the chipset, that statement is 90% true. It wants low latency and high bandwidth. latency = time to get the first piece of data requested bandwidth = data rate in GBytes/sec There are other memory characteristics such as the ability to keep multiple "pages" of RAM open concurrently which RDRAM excelled at. But recently, to get RDRAM to be less expensive, the multi-page capability of RDRAM was degraded to near-DDR levels. Athlon doesn't work with Rambus' RDRAM not because of the CPU but because none of the chipsets support RDRAM. However, since only 1% of current ram production is RDRAM, Athlon won't be left in the dust anytime soon. If RDRAM became popular, it would take six months to design a chipset. But it would take two years to convert even half of overall RAM production over to RDRAM. Its dangerous to only have chipsets supporting one memory technology, but in reality all the new chipsets for DDR SDRAM also support standard SDRAM. It won't be possible to support both on the same motherboard, or very difficult at least. Some specific comments: <garden hose analogy> His whole argument is about bandwidth under various conditions. The big problem with the RDRAM firehose is that it takes too long to turn it on! Furthermore, RDRAM is not like a firehose at all! It only transfers 16 bits at a time, but 4 times as fast. Kind of like a steel 1/2" pipe. The narrow spigot (16 bits) but high data rate leads to problems with overheating and synchronization. A big problem with designed RDRAM modules is that all the path lengths have to be absolutely identical. To use the hose/pipe analogy, designing a narrow pipe to handle a lot of water is not difficult, but the valves are a bitch! The statement that RDRAM systems can use a single module but DDR systems will need 4 modules is TOTALLY BOGUS. In reality, all Pentium 4 systems have to have EITHER TWO OR FOUR RDRAM modules, but SDRAM systems can have 1,2 or 3 modules. I'm not sure DDR modules need to be used in pairs or not, but they certainly don't need to be installed in quads. Hope this helped. Any claim you see by RDRAM fanatics should be taken with a big grain of salt, its good you are attempting to check out their claim of imminent doom for AMD. Petz

To: Umunhum who wrote (10525)	9/28/2000 12:57:39 PM
From: EricRR	Respond to of 275872

This whole thing is about the future. AMD is thinking about maximizing returns right now. They are going with DDR, which was widely recognized as a stop-gap measure years ago. But DDR will not meet their needs for the long term, and they will be forced to go through another costly design cycle to get rid of it. Intel is thinking several years into the future. They know the kinds of things they will need to be able to support, what kinds of markets they are going after, 5 and 10 years out. The P4 core is designed to take them 5 years out, perhaps to the end of the 32 bit processor. Timna is the low cost processor, and it also is designed to be used with RDRAM in value systems. AMD is taking advantage of the weakness in Intel right now, as they switch from the old product to the new, but a black eye is not the same thing as a death blow, and you can expect Intel to recover quickly. This guy's premise is really weak. Intel has bet on DDR for both their Itanium, McKinley, and Foster systems. Nintendo dumped RDRAM for DDR in there next gen game console. DDR will be used in the Xbox. DDR is here to stay. How many NEW RDRAM chipsets does Intel have on Willy's roadmap? How many new SDRAM/DDR willy chip sets are on Willy's roadmap (add VIA and ALI too). Its RDRAM whose future is in doubt. BTW the guy makes the point that Willy was designed for RDRAM, and by comparison Athlon won't be able to benifit RDRAM's bandwidth (I'm not going to touch the latency/bandwidth issue). Well how good is a Willy with SDRAM going to perform? PS- McKinly taped out. Itanium is dead. Long live IA-64! theregister.co.uk

To: Umunhum who wrote (10525)	9/28/2000 2:11:35 PM
From: hmaly	Respond to of 275872

uhmuhnum Re..AMD cannot simply switch to RDRAM at a moment's notice. <<<< Why would anyone want to switch to RDRAM at a moments notice. Even if RDRAM became all the rage, it would take at least a yr. for the manufacturers to ramp up supply to 20% of market. Where is this rumored supply of RDRAM going to come from; out of your a**? The sadder part is that with Athalon and DDR, AMD could have taken the performance crown from Intel for at least one full quarter, maybe two, maybe three, if DDR had been available on schedule. It would have been a major blow to Intel's market share. <<< The athlon has taken the market lead. While you were out driveling about rambus, Intel recalled their 1.13 ghZ P4 and now 1.1 t-bird rules the roost; and it has been a major blow to Intel, or haven't you heard, Intel announced last wk. they wouldn't make q3-00, and there are reports all over this thread about Intels falling market share. You can't take market share by tieing, even if you discount the price of the product. You have to make your product better. AMD has failed to do that. So they have taken a little market share, but it is likely the P4 will take that back and then some. <<< Where have you been all of your life. Businesses almost always take market share by discounting the price. What is difficult is to keep on charging exorbient prices for average products and not lose market share. Intel thought they could but, with their warning last wk. it seems Intel has fell also. The P4 can't take market share back until Rambus drops its prices and ramps up to 15% of market, or P4 gets a DDR chipset, which isn't going to happen by Intels own roadmap until 2h-01. Intel was not derailed by Athalon, and they are winding up for the counterattack with a Louisville Slugger. <<<<< Maybe Intel will do better designing and selling louisville sluggers. Designing X-86 cpu's doesn't seem to be Intel's forte anymore.