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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Dan3 who wrote (150698)	12/1/2001 3:57:06 PM
From: Tushar Patel	Read Replies (1) \| Respond to of 186894

Dell price: $5,040 Rackmountpro price: $3,215 That's a difference of $1,825, or 75% more for the Dell. Especially if you're considering dozens or hundreds of units, that can become a very big deal. This is like saying that I can buy a car kit (with the warranty/support of each individual component provided by the manufacturer of that component) for less than a fully assembled one with a single dealership providing warranty support. [From the Rackmountpro website FAQ: Each individual manufacturer warranty will apply to each component. ] Why don't you compare similarly configured systems from the same supplier? i.e. Intel based systems from Rackmountpro vs. AMD based ones? After all if the quality & support provided by Rackmountpro is good enough for your company if the system is AMD based, surely the same applies if the system is Intel based also.

To: Dan3 who wrote (150698)	12/1/2001 4:10:23 PM
From: wanna_bmw	Respond to of 186894

Dan, I disagree with you on a couple point. Re: "That's a difference of $1,825, or 75% more for the Dell. Especially if you're considering dozens or hundreds of units, that can become a very big deal." OK, let's say that 100 systems are ordered, meaning that a company would spend $182,500 more for the Dell units. With 100 systems, the chance of something going wrong all of a sudden increases. Let's give AMD the benefit of the doubt, and say that there is an equal chance of both Intel and AMD systems going down for an amount of time. These front end systems usually claim reliability of 99.9% uptime. That's about equivalent to 8 or 9 hours per year (give or take a couple). Basically, it's a full 8-hour work day or more. A full day of compute resources down could cost a company millions, or even a small company tens of thousands. Dell's support structure, however, is meant to minimize the effects of these down times. Let's say a Dell team can debug the problem faster, thus reducing down time in half compared to the Rackserver debug hotline. 4 hours less downtime could conceivably make up most of those costs from the purchase of those servers. Additionally, if we made the wrong assumptions, and the AMD systems have more downtime than the Dell systems, this would only be amplified. Now, I'm only giving Rackserver the benefit of the doubt because I know how strongly you AMDroids fight for the stability of the Athlon platform. But most of the time, it's not the CPU that has the problem. Dell has been manufacturing and selling servers in high volumes for many years. Rackserver has a small fraction of Dell's experience (or any major OEM for that matter). It's only natural to assume that Dell's systems are more stable. Even if you disagree with this, it's only your personal opinion, and one which has had limited exposure (despite your profession). I don't want to tell you you're wrong about the reliability of an Athlon platform, but the fact is that they've had far less real world exposure, and that will limit the trials by errors that businesses can learn from using their systems. In all likelihood, Athlon platforms will fail more than Intel platforms, and it's not because the platform is designed any worse. Rather, it's because it is not ubiquitous enough to get the same level of support understanding. Re: "Now, consider that the AMD chip is available at speeds up to 25% faster than the Intel, which means that fewer boxes could be used, meaning lower administration costs." Excuse me if I misread, but aren't you comparing an Athlon 1.2GHz with a Tualatin 1.26GHz? While the Athlon tends to perform better on floating point apps, the Tualatin definitely leads on Integer at the same clock frequency (due to the extra cache, etc). I believe a 1.26GHz Tualatin would probably be 10-20% faster than a 1.2GHz Athlon in many server apps. Re: "P4's weren't available in a 1U form factor from Dell - dissipating 190 watts from 2 P4's in a 1U case seems to be too much to handle for Dell. The lower power consumption of Athlon MP lets them scale higher in small form factor cases. Of course, given P4's mediocre performance on server applications, it's something of a moot point, but buyers are still stuck with the "old" Intel processor compared to the "new" AMD processor." Intel has yet to offer Netburst based Xeon processors for servers yet, so I don't know where you are making the comparison. Xeon processors are only sold for workstations, due to the lack of a DDR chipset. By the time Intel launches DDR chipsets, they will have both Plumas and Prestonia. Prestonia will have 512KB of L2 and Hyperthreading, and Plumas will have dual DDR and wide I/O paths. It should be far stronger than any Xeon systems tested by the cheerleader sites against an Athlon MP. As far as the form factor, you are assuming that Dell tried and failed with a 1U configuration. In all likelihood, they haven't tried, since they aren't selling Netburst based Xeons to the server market yet, and Xeon based workstations don't need to be in 1U form factor. If they still don't have anything next quarter, you may have a point. Until then, you're only spreading FUD. wbmw