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: Dan3 who wrote (257552)	12/25/2008 2:27:09 PM
From: wbmw	Respond to of 275872

Re: I don't want an intentionally broken platform with no compatible upgrade route, I want the best X86 servers I can buy. Are you referring to Dunnington in this statement? I've always thought of Dunnington as a good intermediate product on the way to developing an 8-core Nehalem. New architectures have a 4-5 year time line, so Intel was working on getting to an integrated memory controller and high bandwidth interconnects ever since AMD surprised with Opteron. It takes a long time to change fundamental roadmaps, but Intel's efforts in the mean time haven't been fruitless. Their server offerings since Woodcrest in 2006 have not only been noteworthy, but have also been far superior to AMD's server roadmap. Intel was the first to offer quad core on the server roadmap with Clovertown, and they enjoyed its performance leadership for more than a year, and AMD's first response with Barcelona was a buggy chip they had to shop shipments on until they stepped the parts. It resulted in 4-5 months of delay, and I'm sure many customers weren't happy about that. You may find now that Shanghai is now giving you better bang for the buck, but don't pretend that Intel hasn't been the leader in the server market in the mean time. Back in 2003 when Intel was pushing Itanium as the only 64-bit ISA, I can see why you'd be cynical enough to claim that they were prioritizing marketing over engineering. But it only took them until 2004 to release ISA extensions that were binary compatible with AMD64, so continuing to blame them for holding back the software industry sounds a lot like revisionist history to me. And by 2006, they recovered performance leadership - at least in the 2-socket space. 4-socket servers have been a toss-up between Intel and AMD since Intel launched Tigerton, which simply moved Core architecture to the high end. And using the same platform, they are now upgradable to Dunnington, which still has performance leadership over Shanghai on some benchmarks. Again, you may find Shanghai gives you more of what you want in a server, but after 5 years since Opteron, Intel now has a Nehalem architecture, which offers all the same advantages that Opteron has had, plus better performance. At least, that's what it looks like today based on the SAP benchmark. There's still more to come.

To: Dan3 who wrote (257552)	12/25/2008 2:51:43 PM
From: rzborusa	Respond to of 275872

Dan, I blame Intel for delaying the transition to 64 bits for years (some drivers and software are still not available). This has caused substantial harm to everyone who uses computers, and Intel did it as a marketing ploy, harming customers like me in an attempt to force users to an expensive, slow progressing Intel proprietary world. Dan and others who openly and honestly embrace the excellent AMD journey and enjoy seeing the 45nm phenom II, merry Christmas and Happy New Year. Additionally, I would wish the world happy inauguration of our next president. And, hopefully an end to the cult of personalities who have given a blind eye toward the abuse of power in recent history. I also have to say I respect the principles of the Arab investors who took up the slack for AMD, when AMD was driven to near the end of its rope, perhaps_due_in_part to illegal practices and policies allowed in this country.