…of course they’d also have been miniaturized down to two inches per side 
With Windows XP, one can often reinstall the OS from CD-ROM over the previous installation once the new motherboard, CPU, and RAM are in place, which has the effect of redetecting all of the new hardware while keeping installed programs and settings intact.
Well, you’d have to reformat it of course… I like to do that every once in a while anyhow. If replacing the guts of your computer is in any way a hobby, I’d assume reinstalling the OS would be part and parcel of that. It certainly is in my case. But it’s not like it’s really going to matter… Longhorn’s installer is somewhat more intuitive and a full install will take about 15 minutes. Who could have an excuse for not reformatting then?
Shame to hear about BTX. I don’t think there’s anything wrong with the ATX form factor that a large case doesn’t solve.
Well, maybe not now, but our computers are getting more powerful, faster, bigger and with more bells and whistles…that all produces heat, and the heat is going to keep increasing as we advance, without shifting to entirely new technologies to help alleviate that heat, and motherboards is just one of the possible ways.
BTX will also be quieter, and with some of the top-end gaming machine systems, they can certainly use it now let alone in the future.
I have 6 case fans in my computer at the moment, plus the fan on my 9800, plus a larger non-OEM processor fan, and there’s no getting around the fact I’m all for anything that will help my machines stay cool and healthy. 
Case fans and size of the case will only go so far, and not everyone wants a big full-size server tower just for their home computer.
Everyone’s saying 3 to 5 years for your computer to be truly obsolete, but the truth is, it’ll never really be obsolete beyond your expectations at the time of original purpose.
For example, a Mac Plus from 1984 will do today exactly what it did 20 years ago.
Computers don’t get obsolete; your expectations change.
Given that, I’m on my 4th PC-clone, second PowerBook, and 6th desktop Mac. Not because the other three PeeCees, other PowerBook, and other 5 desktop Macs went obsolete. But the way I use them and what I expect them to be able to do has changed.
Markxxx wrote: One person’s obsolete isn’t anothers.
I use the car analogy. My old Honda gets me to work and back. A new Honda has things I like better, but it won’t get me to work any better than the old one.
When is my old car obsolete? When I get tired of it.
I assume the OP means, "by the time I get a computer, is there already something in the same general product range with more capability?
Almost definitely. I have had close dealings with two tech companies, and both of them seemed to operate on a model where each version of a product was a separate profit center that had to prove itself independently.
In the early 1990’s I did some work for the Bose speaker plant near Boston. Their big product rollout at the time was the Acoustimass system, the first of the “tiny tweeters with hidden bass box” models to hit the market. They were rolling version 1 out the door with full marketing fanfare. Meanwhile, most of the executives had one of the many finalized, already manufactured, ready-to-ship superior version II speakers on their desks. These were not prototypes, but market ready product whose rollout was still months away while they shipped out the inferior version I’s.
In the late 1990s, I attended a FileMaker conference in Monterey, CA. FileMaker is a fairly close-knit community, and the corporate honchos were asking the users in an open forum about what features should be included in the future. The developers gave their feedback, along with many queries about whether the changes could be ready in time for Version 5.0 (version 4.0 was about to be released to market, and we all received an advance copy at the conference). There was some hemming and hawing, and some of found out why later on in the week. At a special seminar, certain people got to look at the final, ready-for-market but not yet to be released version 5.0.
So when you buy your fancy box from the computer store, it would not surprise me in the least to hear that its successor (that leaves it in the dust) has already been mass-produced.
Although that’s certainly true my home PC is called BEHEMOTH for a reason. I have nothing against a more discrete form factor (except for what it’s going to cost me to upgrade) but this big ATX bad boy would be a cast-iron bitch for burglars to pack away! (I bet they love LAN party cases… look, it even has a convenient carrying handle!) It also impresses all my friends. They’ve just never seen a computer so huge. (And no, not trying to compensate for any inadequecies here… 
)
Still, better cooling and less noise is always a good thing. From the links posted it looks like this is pretty exclusively Intel though… is it going to be a different form factor for boards that take AMD chips?
My own twist on the OP: will computers ever be fast enough that they will never be considered obsolete? In other words, computers that are so fast they can perform any programmed task instantly, no matter what chip and ram are on the board?
My first Windows based PC was a Pentium-166. It was about the fastest consumer model available, and I paid about $3400. I could have gotten a Pentium-75 for about $1500. Had I done so, by the time the 75mhz was obsolete, I could have purchased a Pentium 200MMX for about a grand. At that point I would have had an old machine to hand down to the kids, a faster primary machine, and cash left over to uprage again in a couple of years to another computer that was about a year behind the times.
As it was, my “State of the Art” machine had to last far longer than I would have liked.
Back in 2000, I was planning on buying a new motherboard and CPU, and I was looking at the newspaper where all the ads were saying the latest Pentiums were at 450Mhz. That weekend I went to a computer swapmeet and bought a P550!!! Wow! I’d superceded the ‘latest’ stuff!
Exactly seven days later, I’m not kidding, all the newspapers were saying the latest CPU speed was 650Mhz. In literally two weeks, minimum specs had leaped right past me.
The end result was that I was one of the few people in the world who had a PIII 550.
I’ve been pretty lucky with leapfrogging over standards; I usually only upgrade when I can quadruple my computer’s effective performance (e.g., not just clock speed but also RAM, HD volume-speed considerations, video RAM, etc.) for on the close order of $1,000. I build systems in old cases from bargain parts purchased at Newegg.com, re-use power supplies, CD-ROMs, and HDDs where possible, and keep a “hardware graveyard” of other useful widgets (I ran across a surplus vendor selling floppy drives for $5.00 each; I bought ten, so when one goes bad, I just swap). NICs and modems? I’ve got three or four spares of each. This keeps my costs down to
- video cards
- motherboards
- RAM
- CPUs
- cooling equipment
And as long as I buy a motherboard that uses the latest interfaces, the standard changes infrequently enough that I can adapt easily. This means I can usually get one major upgrade out of each system I build before I have to go with a new motherboard and a ground-up redesign (the switch from PCI graphics cards to AGP graphics cards caught me flat-footed; ditto the switch from PC-133-size to PC-2100-size RAM). $1,000 for the big buy, and then maybe $500 or $600 on the rebuild a few years later. Not too shabby, but not cutting-edge, either.
[QUOTE=jinty]
I wouldn’t say obsolete, which implies useless for its main purpose.
Moore’s Law: " the number of transistors on integrated circuits had doubled every [18 months]."
Can somebody please explain this to me? How can a mass-produced item “just happen” to double its efficiency and quality, on a regular schedule.? If Intel knows how to make a certain chip today,and knows how to improve it by next September, why don’t they -or their competitors–make the next September’s version already?
(Yes, I know, I’m being naive–the world doesnt work according to scientific advances, it runs according to stock market quarterly reports)
So, yes, a given chip, or a given version of software, may be good today.But if the makers already know how to make the next advanced version, how come none of their competitors do it first?