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: eracer who wrote (212047)	9/30/2006 4:09:52 PM
From: Rink	Read Replies (2) \| Respond to of 275872

Eracer, Currently ATI's stream software for physics is for the GPU. From what I read (yep Inq) they will (future tense) release a platform with three Direct x16 expansion slots of which two will be for GPU's and one for a new physics chip. Don't get me wrong: I'm not saying it's a 100% certain that this physics chip will arive in a much different form than the GPU. I am saying that for me your opinion doesn't count enough against the Inq who most of the time has some information to base their stories on (at least you certainly have none). This background information is not like the WSJ would require, but like a somewhat reasonably informed tabloid. So whatever it'll turn out to be worth, it's about all we have for now. Regards, Rink

To: eracer who wrote (212047)	9/30/2006 4:16:24 PM
From: DRBES	Read Replies (2) \| Respond to of 275872

...Sure... AMD doesn't have a dual-core chip either because there is only one memory controller instead of two. Until AMD has quad-channel DDR they don't have a "true" dual-core either. AMD's cores don't share L2 cache, another strike against being a "real" dual-core CPU.... i really do hate reading your posts...you are correct, of course...thanks once again for keeping the thread honest

To: eracer who wrote (212047)	9/30/2006 4:19:22 PM
From: plantlife	Read Replies (1) \| Respond to of 275872

Here is my opinion on why you were rather subborn on this point: If ATI created a unique physics chip rather than renamed a video card GPU it would made ATI's engineering accomplishments appear much more significant. That type of creativity and engineering prowess would certainly make ATI look like a much better buy at ~$5.5 billion, and therefore make AMD's decision to pay such a high price for ATI seem more reasonable. _______________________ I can't believe that you held that opinion and didn't state it. I wouldn't be able to answer it anyway because I was limited to 5 posts. Clearly, I didn't know if it was or wasn't a Physics chip, and I still don't. I also didn't know if it was or wasn't the 1900 that they were referring to. I put up a casual post based on a link from the Inquirer that I thought people here would find value in, and didn't expect to incur a hostile reaction from such am innocuous comment. As for ATI, I knew very little about that company other than their Video Cards were competitive with Nvidea. I did know about Torrenza, and when the Physics chip was mentioned on Inquirer, I made the logical conclusion that this was what AMD wanted. I guess if that was true, you would agree with my conclusion. So, I still don't know whether it's true or not, but with this new Stream computing system, it doesn't matter. That will do just fine, and it looks like the Co-processor candidate AMD is looking for. If somebody out there has a better one that works with Torrenza, I'm sure AMD would be receptive to it. Cell seems to be a candidate, according to IBM, but its only an accelerator.

To: eracer who wrote (212047)	9/30/2006 5:17:42 PM
From: heatsinker2	Respond to of 275872

Eracer "Rev. F and Rev. G are not expected to retake the performance lead in desktops or notebooks" I don't think your getting the AMD vision of the universe, as clarified a while back by Sir Combjelly. There's a high performance world where quad core will be king. Servers, sure, but also high-end desktops for gaming and business workstations. Then there's a low power world: laptops, generic business computers and home users who don't give a cr*p about gaming. So where does this leave top-performing dual core CPU's? Well this is a category that AMD thinks is not too important. That's why the green guys are taking forever to challenge Conroe. If nothing else, you have to admit that this strategy is coherent.

To: eracer who wrote (212047)	9/30/2006 7:46:57 PM
From: pgerassi	Read Replies (2) \| Respond to of 275872

Dear Eracer: Intel does not have a quad core chip. The QC CPU they will supposedly offer in November is a quad core MCM package composed of two dual core dies. AMD does in fact have a dual core chip. And if you look at any AM2 or socket F die, it has 2 DCTs, one for each DDR2 channel. If you claim that AMD's Rev G isn't dual core, because it hasn't 4 DDR2 channels, then every Intel CPU chip is a no core die having zero DRAM channels of any kind at all. And for the record, we are at the end of Q3. One year is end of Q3 2007 and Rev H (K8L) will be out and even Rev I with DDR3 and HT3.0 will be in the pipeline. Intel will be behind in all segments and playing catch up with nothing in the pipeline until 2009 at the earliest. Pete