The point is simply that DP Xeons are much more expensive than desktop Core chips. Ancillary chips for DP systems (5000X, FB-DIMM) are also more expensive.
We've been saying that the Mac Pro (and the Dell Precision) are expensive, not that they are over-priced. Big difference.
That might have been your intent, but it wasn't your message. As soon as you said that the Core i7 had the same performance at a fraction of the cost, you're projecting the message of "over-priced".
Dell offers single-socket systems using desktop parts, including Core 2 Quad and Core i7.
Sure, Dell offers that today, but were they offering that in January 2008? Considering that the i7 was just released, the answer is "No". As such, you're trying to insert the non-issue of normal technological advancement product development leadtimes and making a "2009 Product vs 2008 Product" comparison.
And what is playing a part here is the Law of Diminishing Returns: to get the last 10% of available performance costs disproportionately more than the cost of going from 66% to 75% - - always. All technologies.
The Apple (& Dell) Xeon Workstations functionally represent last year's "95th percentile performance" product, which because of a year's worth of technological progress, is now equivalent to this year's 75th percentile performer. As such, we can rely on that the price for its 'equal performance' replacement will be progressively lower with each passing month, for as long as Moore's Law remains in effect.
FWIW, I can recall when 16K of memory cost $1000. Today, I can get 100,000x more for roughly 1/100th the cost.
That is why Dell can offer systems which match the performance of the Mac Pro (and Dell Precision) in some cases, and beat the price of the Imac.
There's yet another example of a "...they are over-priced..." inference statement.
I don't know why you want to argue that a Dell Xeon workstation price is wrong, that's just a distraction that doesn't follow from the arguments that have been made.
You find it 'distracting' because what it does is detracts from your message of "...they are over-priced..." that you claim to not be saying.
Once Dell Precision has Nehalem Xeons, the current situation where the desktop system can match the workstation will end - the much more expensive system will be more powerful than the desktop.
Once the Mac Pro has Nehalem Xeons, the huge gaping hole in the product line will get even larger. (if...)
If one speculatively assumes that the iMac will remain a relatively mundane (mobile-CPU based) performer then it can be reasonably argued that the performance gap is likely to become larger. However, it is also possible that the next iMac might very well get the i7, which changes the terrain and arguably closes this chasm. The answer won't really come until there's a reliable leak, or official product announcements. Granted, this is a rumor site, but there's a bit too much Chicken Little (sky is falling!) for my tastes.
Yes, it's a great deal for what it is, but even with the Pros, most do not want or need a xeon workstation. If you need a desktop and have to spend almost an extra grand beyond what you paid on you're last PowerMac, its not such a good deal. When your computer and operating system company forces the hardware to basically, two extreme, it's very difficult to make direct comparisons. A lot of the time you're caught in no man's land have to make relative comparisons because the machine that you need does not exist on this platform anymore.
I hear you, but with ~30 years of personal desktop computing under my belt, I'm going to disagree: the general fallacy is in assuming that users come in a virtually continuous gradation of demands for horsepower. The reality is that they don't: its not a continuous distribution, but grouped in clusters. One cluster is low end. Another is midrange. Another is high-end. As such, one doesn't really need 100 discrete products to cover 1%-100%, but merely a couple of products that generally address where the data is clustered: simplistically, visualize this as being at 33%, 66% and 99%.
What the crux of this whole 'xMac' debate is that it is being claimed that the 66% solution isn't good enough and that the 99% is too expensive, so you want a "75%" solution.
There is life below the super high professional end of the spectrum and above the low to middle end typical consumer level.
Sure there is. But that's not the only issue to be considered when running a business: the key question is if the "75% need" is a significant enough cluster to be worth providing a solution.
And do keep in mind that just because Dell chooses to sell a 75% doesn't make the market segment significance answer to automatically be 'Yes' for Apple too.
In fact, for most applications, 6 of those 8-cores are going to be doing nothing at all.
Which makes it certainly sound like there's a big cluster of users who mostly only needs 2 cores ... aka the iMac.
Nobody is saying that the MacPro isn't a great deal for a workstation. Workstations are inherently expensive. What they're saying is that it isn't a very good deal for someone who needs a desktop, something that does not exist with on Mac OS X currently.
Yes, but you've not successfully made the business case that there's enough of a cluster (demand for "75%") to be fiscally viable for Apple to cater to you.
It is not germane to Apple what Dell does, because Dell is in a different enough market ... and with different competitors ... such that Dell makes different business decisions.
To abuse YA automobile analogy, Porsche doesn't make motorcycles even though BMW does, and BMW doesn't make heavy trucks, even though Mercedes does. Since all three (cars, trucks, motorcycles) are transportation products, how come none of these three companies make all three?
-hh