[mitchell_pgh]

Again...

Quicksilver 733 Tower (My home system)
Original Price: $1100
Date Purchased. Jan. 2002

F1000's BROKEN VALUE TRACKER
- - - - - - - - - - - - - - - -
Jan. 2002 = $1100 (original price)
Jan. 2003 = $550 (1/2 of $1100)
Jan. 2004 = $275 (1/2 of $550)
Jan. 2005 = $137.50 (1/2 of $275)

Please show me where I can pick up a G4 733 tower for under $150. The average price on eBay is twice that.

My system, while not a screamer, is more then capable of designing web and print projects.

[misnomer]

Originally posted by The Godfather:
Your source please?

The Virginia Tech News website as well as their Big Mac site.

Nearly two years ago the project began and back in August of '03 they started publishing Linpack benchmark scores for the cluster. Dr. Varadarajan's original proposal and presentation had a relatively short timetable for their cluster to reach a production state as a functioning research cluster, where CPU time could be rented to outside researchers as well as used for research by the aerospace department at VT. Those goals passed by, unmet, and in January of '04 Big Mac I was dismantled before ever having reached a production state. Earlier, Dr. V had stated there were some issues caused by lack of ECC RAM in the original G5 PowerMacs and long after the cluster was torn down a press release on the Big Mac site mentioned much the same thing:

Well with the concept proven we now had to make sure we had a system capable of conducting scientific computation. We needed to upgrade the system to something with error correcting code (ECC) RAM. The Power Macs did not support it and the XServes were coming. So in January we tore the system down and started prepping for the XServes. And now they're here and we have our final system. The best is yet to come.

What's interesting is you can see in that press release Dr. V positioning the original PowerMac version of Big Mac as a multi-million dollar proof of concept, only meant to pave the way for the eventual switch to Xserve G5s, which could then be used for actual production work. That of course flew in the face of his original presentation that had proposed their PowerMac cluster being in a production state, selling CPU time, many months earlier. There's also the matter of the original facilities that were built around housing and cooling PowerMacs, and now have a lot of wasted capacity, at least until they add many more xserves in the future.

Unforeseen shipping delays with the xserve G5s left Big Mac dismantled for much of the rest of the year, until the very end of sept '04 and early oct, when the first Big Mac v2, pre-production Linpack benchmarks started popping up. It wasn't until January 10th of this year that Big Mac entered a production state and VTech finally began selling CPU time.

[f1000]

However, your argument still, IMO, refers mainly to consumers and not professionals who make their price calculations somewhat differently to yours. When a company buys its hardware, (let's assume that they prefer Macs for reasons of reliability, easier maintenance etc, or need it for the video or audio software) they will make take the obsolescence into account, which means that they know that their machines will probably need to be replaced in three to four years. If their productivity is important, they will buy machines that they feel will still be usable three years down the line, with software upgrades etc that will have come in that time.

My argument probably is more pointed at consumers, which I noted at the end of my first post, since most professionals can justify the purchase of a more expensive setup as providing extra CPU cycles that result in net profits despite increased depreciation. I also acknowledge that there�s a quantization of usability for some computer applications. A video editor, for example, needs to see full frame rate playback. She can�t work with a 50% slower computer that can only achieve half the frame rate; for her, the 50% slower computer might as well be a zero computer since it can�t get the job done at all. Buying a cheaper but useless computer is a false economy.

Still, even businesses have to consider the bottom line. If my argument is correct, then it�s better for businesses to buy that single processor PowerMac over the dual processor one if the only justification for the latter is futureproofing. Replacing the cheaper model biannually as opposed to replacing the more expensive model every four years isn�t going to put a big dent in productivity, and it probably will be financially more prudent.

This means, and I have concrete examples in the form of Migros, the largest supermarket chain in switzerland, which uses sunflower iMacs in their offices which were purchased about two years ago. They obviously didn't purchase PowerMacs because they would be overkill in an Office setting and would raise the purchase price enormously. They did, however, buy the fastest iMacs available at the time.

There are two issues here: 1) Did Migros need the fastest iMac to achieve an acceptable performance. If so, then Migros clearly didn�t buy the fastest iMac at the time to insure against obsolescence; 2) If Migros did buy the fastest iMac at the time to hedge against obsolescence, then did they base their decision on some rational, data-driven metric? According to the Fed Reserve paper, this may not be the case.

Another example would be a friend of mine who has a small software company. He two Macs, one an older Quicksilver PowerMac which serves as his file/web/mail/version control server, with maxxed out hard drives, and the second a 17" Powerbook, on which he does his development. Since he also does consulting he needed a portable to take with him to customers, and he needed a large screen since working all day on a 12" or 14" ibook would ruin his eyes.

Here you make a good case for need. Your friend needs a dedicated server, and a PowerMac with multiple internal hard drives suits this purpose better than any single drive Mac. Your friend also needs a laptop with a large screen because his eyes can�t handle anything smaller.

Since he's a small company, he will sell both machines eventually on ebay or Ricardo (more popular in Switzerland). Also , since he's a small company with no huge profits, he can't afford to buy new machines every two years. So he buys the machines that he can still sell at a decent price a few years down the road, but which also offer him the best bang for his buck right now.

If you show him this thread, he may reconsider his plan.

As I said, while price is a very important part of the equation, it isn't the only part.

The i-Series and mini are perfect for most home users, but your original post, stating that they are better economically is a bit too simplified. There are many home users who simply want a Powerbook with its 15" or 17" screen and fairly many who simply want the fastest G5 PowerMac available and have the money to buy one.

After all, playing Doom3 on a mini is no fun

I pretty much stated the same thing some posts before yours. Price/MIPS isn�t the bottom line for all users; if it were, none of us would ever buy laptops, or even Macs for that matter.

My argument remains that when we give advice to people, we should tailor our recommendations to their needs, but leave out any mention of futureproofing.

[f1000]

Originally posted by Eriamjh:
The original post would be true except for one thing: the true market value at the end of three or five years is not the same as the remaning value after depreciation.

Others have already pointed this out.

You didn't read my reference:

Our paper builds on the work of Geske, Ramey, and Shapiro to further narrow the knowledge gap surrounding depreciation rates for PCs. Like their work � and most of the earlier literature � we rely on prices of used assets to estimate depreciation. We construct a large dataset of prices and model characteristics for used PCs listed in bluebooks. This dataset includes nearly 13,000 observations and spans the period from 1985 to 2002, covering almost the entire era of personal computers.

As you can see, the authors back-calculated depreciation rates from the prices of used PCs listed in bluebooks. Their dataset spans almost 20 years.

Moreover, they conclude that PCs lose roughly half their remaining value, on average, with each additional year of use not just from depreciation, but from a combination of depreciation and revaluation.

EDIT: I misworded that last paragraph.

[f1000]

Originally posted by Big Mac:
Your initial posts were a bit confusing, f1000, because you seemingly implied that everyone would be better off with entry-level hardware. You clarified your position when you said that people should not recommend more hardware than individuals need.

Actually, I am going one step further. I�m saying that people should not recommend more hardware than individuals need at that particular moment, obsolescence be damned.

Originally posted by Big Mac:
I don't think anyone does that, at least not around here. If someone just wants to run basic applications, they should get either an iBook, a mini or an eMac. Higher end iBooks and the iMac are there for midrange customers, and professionals have Power Macs and PowerBooks. The ideal is to always buy that which suits the individual best. Only dishonest sales people will try to sell high-end computers to people who clearly do not need them. So you're essentially barking up the wrong tree and belaboring an obvious point.

I initiated this thread to serve as a point for future reference; it�s not meant to chastise anyone in particular.

I�ve seen enough people make the well-meaning recommendation to buy more computer to stave off obsolescence that I decided to post a definitive thread on the matter. I doubt that anyone who makes such a recommendation is being dishonest; they simply haven�t analyzed the economics before.

[ort888]

Originally posted by mitchell_pgh:
Again...

Quicksilver 733 Tower (My home system)
Original Price: $1100
Date Purchased. Jan. 2002

F1000's BROKEN VALUE TRACKER
- - - - - - - - - - - - - - - -
Jan. 2002 = $1100 (original price)
Jan. 2003 = $550 (1/2 of $1100)
Jan. 2004 = $275 (1/2 of $550)
Jan. 2005 = $137.50 (1/2 of $275)

Please show me where I can pick up a G4 733 tower for under $150. The average price on eBay is twice that.

My system, while not a screamer, is more then capable of designing web and print projects.

When was a Quicksilver $1,100 bucks? Did you buy it used or something?

They came out in mid 2001 for $1,700. Now they sell for $500 - $600 on ebay. That's about a 30% depriciation every year.

His math may be a little off, but his reasoning is sound.

[f1000]

Originally posted by goMac:
The other problem is Moore's law is quickly starting to not apply anymore. For people like us it is still very important, because we do Photoshop and games and very cpu intensive sutff. The average user only does Word or email. These days, any machine can handle these tasks well. The average user isn't going to be on a two year upgrade cycle. They'll probably keep their machine until it dies. They'll buy a machine with the functionality they need. Big screen, some upgrade slots if they want them, dvd burner, you get the idea. They don't care how fast the machine is, they just care about what it comes with. And these days thats really starting to matter the most.

You miss the full ramifications of Moore�s Law. Moore�s Law is not just about the fastest computer; it�s also about the cheapest computer. Due to Moore�s Law, the price of a new computer that can do a certain number of MIPS will decline over time. In that sense, Moore�s Law will continue to apply since future buyers will always be able to get a cheaper PC to do the exact same thing that they are doing today.

In any case, I believe that in the coming months HDTV will spur increasing numbers of PC and Mac users to upgrade their machines to even faster processors. Not everyone is a Photoshop guru or a gamer (certainly not me), but just about everyone loves listening to super high fidelity music (DVD-A) and watching HDTV.

Originally posted by goMac:
So when I recommend a machine, I just recommend what would be best based on what the user wants in a computer, and usually a little more I think they will need eventually. For a user who wants a machine "to last them a while" I will usually recommend a more expensive machine, simply because I think the build quality will probably be better.

But that�s the crux of the matter, goMac. It�s better for such a user to buy the cheapest system that�ll accomplish what he needs to do now (and then buy a new computer when he needs to do more a year or two from now) than it is for him to buy a little more now thinking that he will need it eventually.

Show me the calculations, if you will, that prove your approach is more cost effective. Like Spheric, I've taken a bath on expensive computer hardware.

[goMac]

Originally posted by f1000:
You miss the full ramifications of Moore�s Law. Moore�s Law is not just about the fastest computer; it�s also about the cheapest computer. Due to Moore�s Law, the price of a new computer that can do a certain number of MIPS will decline over time. In that sense, Moore�s Law will continue to apply since future buyers will always be able to get a cheaper PC to do the exact same thing that they are doing today.

In any case, I believe that in the coming months HDTV will spur increasing numbers of PC and Mac users to upgrade their machines to even faster processors. Not everyone is a Photoshop guru or a gamer (certainly not me), but just about everyone loves listening to super high fidelity music (DVD-A) and watching HDTV.



But that�s the crux of the matter, goMac. It�s better for such a user to buy the cheapest system that�ll accomplish what he needs to do now (and then buy a new computer when he needs to do more a year or two from now) than it is for him to buy a little more now thinking that he will need it eventually.

Show me the calculations, if you will, that prove your approach is more cost effective. Like Spheric, I've taken a bath on expensive computer hardware.

I think you miss my point. You're looking still at a 3 year upgrade cycle. The upgrade cycle these days is like 8 years due to Moores Law becoming no longer relevant. Users have no reason to upgrade, therefore instead of paying $750 every 2 years for a computer, they can justify paying $3000 every 8 years.

[f1000]

Originally posted by mitchell_pgh:
Quicksilver 733 Tower (My home system)
Original Price: $1100
Date Purchased. Jan. 2002

As ort888 pointed out, is this how much YOU paid for your Quicksilver in 2002? What is your original price after figuring in inflation?

Originally posted by mitchell_pgh:
F1000's BROKEN VALUE TRACKER
- - - - - - - - - - - - - - - -
Jan. 2002 = $1100 (original price)
Jan. 2003 = $550 (1/2 of $1100)
Jan. 2004 = $275 (1/2 of $550)
Jan. 2005 = $137.50 (1/2 of $275)

Please show me where I can pick up a G4 733 tower for under $150. The average price on eBay is twice that.

1. Are you sure that the eBay items are exact replicas of your model? Some of those sellers throw in AppleCare and illegal extras such as full versions of Photoshop.

2. Are those eBay prices of sold computers or actively listed auctions? Only prices for successfully completed auctions should count.

3. Some people say that eBay, because of its auction nature, often causes people to pay more for junk than they would have otherwise.

In any case, as ort888 also pointed out, you're still experiencing a 35% decline in value annually. I think it�s safe to assume that Macs probably hold their value a bit better than the average generic PC.

Originally posted by mitchell_pgh:
My system, while not a screamer, is more then capable of designing web and print projects.

I should hope so�sucker!

[Brass]

Originally posted by f1000:
You didn't read my reference:

As you can see, the authors back-calculated depreciation rates from the prices of used PCs listed in bluebooks. Their dataset almost spans 20 years.

Moreover, their conclusion that PCs lose roughly half their remaining value, on average, with each additional year of use is not based solely on depreciation, but on a combination of depreciation and revaluation.

I'm not entirely disagreeing with your original sentiments. But some of your arguments are not quite up to scratch. It is well known that Macintosh computers hold their resale value much better than Wintel PCs (and have a longer useful life span).

Therefore this calculation of depreciation may be good for Wintel PCs, but it's unlikely to be quite as accurate for Macs.

[f1000]

BARRETT: NO END IN SIGHT FOR MOORE'S LAW

By Stephen Shankland, CNET News.com
Published on ZDNet News: March 1, 2005, 11:55 AM PT

SAN FRANCISCO--Moore's Law will boost chip abilities for many years to come, Intel CEO Craig Barrett predicted on Tuesday.

The momentum will be kept up first through conventional manufacturing processes, then for many years after that by other technology, he said in a keynote speech at the Intel Developer Forum here.

Barrett predicted that traditional chipmaking technology will permit features as small as 5 nanometers--about the width of 50 hydrogen atoms--to be used on processors.

"We can see how to do this down into the 5-nanometer range," Barrett said. "Beyond that, lots of leakage currents and things like that get in the way. But every time we seem to get into a roadblock, the bright engineers...seem to circumvent that problem."

The future of Intel and the computing industry in general depends in large measure on the ability to pack more circuitry components, called transistors, ever more tightly onto a slice of silicon. To do that, the size of chip features must be shrunk.

Intel today is preparing to introduce processors with features measuring 65 nanometers, or billionths of a meter. Company engineers have forecast the feasibility of 5-nanometer manufacturing processes before. But the public declaration of the chipmaker's top executive carries more weight.

"He was willing to extend the planning horizon to 5 nanometers," Peter Glaskowsky, an analyst at Envisioneering Association, said. The rest of the computing industry will be able to factor that into their plans, he added. "That means everybody gets to stay on the treadmill."

Barrett showed photographs of transistor prototypes built with features measuring 65, 45, 32 and 22 nanometers. Glaskowsky said current expectations are for new processes with features of 15, 10, 7 and 5 nanometers. However, it is expected that the two-year pace of each new manufacturing process will slow, he said.

Conventional chip manufacturing processes use a technology called complimentary metal oxide semiconductors, or CMOS. It's not clear what technology will replace that to create even tinier transistors, but Barrett mentioned three options: quantum dot, polymer layer and nanotube technology.

"I don't think the world has decided what is the replacement yet," Barrett said. "Meanwhile, it's full blast ahead with the standard Moore's Law."

Moore's Law is the 1965 projection by Intel co-founder Gordon Moore that the number of transistors on a chip will double about once every two years. "I've told Gordon I plan to help him celebrate the 50th anniversary of Moore's Law in 2015, and hopefully the 60th anniversary in 2025," Barrett said.

http://news.zdnet.com/2100-9584_22-5594779.html

[f1000]

Originally posted by Brass:
I'm not entirely disagreeing with your original sentiments. But some of your arguments are not quite up to scratch. It is well known that Macintosh computers hold their resale value much better than Wintel PCs (and have a longer useful life span).

Therefore this calculation of depreciation may be good for Wintel PCs, but it's unlikely to be quite as accurate for Macs.

Based on the anecdotal evidence, I'd have to agree that Macs seem to depreciate more slowly than PC�s, but whether this depreciation rate is 35% or 50% the results will still be fairly similar. I�ll demonstrate this fact in my response to GoMac.

[connollyck]

noob Q:
Whats a "poweruser"?