How do you collect? Popularity or Rarity?

[chipcollector]

Let's say you come across a chip and you check ebay but find nothing there. Then check usbid and find out there's about 10 or so of this chip on the obsolete IC market.

Would you buy a chip if it's rare, not caring about it's popularity? Or would you let it go? For example, I remember back in the day (1-2 years ago) when eproms were one of the worst things in existance, now look at how far it's gone, with some worth as much as $200 a piece or more.

[JAC]

Eye candy every time for me. I dont care if its rare or not. It is a annoying seeing idiots bid silly prices on some chips though. So I guess my answer would be popularity, as I prefer Intel chips generally. I do like other brands, but not as much.

[chipcollector]

[CPUShack]

if its a chip, I collect it (whic explains my wallet)

[JAC]

Museums of the future.

[RM68]

I don't care if it is rare or not. I'd like to have a complete collection. At the moment only of one chip manufacturer: Intel. This is because of reasons of economy and of space in my house

[southsails]

I really only collect rare chips or chips with historical significance. That being said, if someone is interesting, I'd buy it for a few bucks, even if there is nothing "rare" about it.

[alvaro84]

My present goal is to have a little home exhibition that gives a nice glimpse of the history of CPUs. I'd like to have every core version of popular line-ups, multiple variants that differ in clock speed or otherwise (different markings, different package) but I don't really care about SSPECs or batches.

On the top of that I'd like to show as many interesting pieces as I can. If it's rare, it's fine, I'm happy. But I like common chips too, especially when they look nice. A few hundred different CPUs make great eye candy too.

My main limit is the price: lately I only go for pieces I can get for free. One hand, working for an electronic waste recycling company I'm at a great position to do so. On the other hand, I can't really afford anything else, working for said company for hardly more than the Hungarian minimum wage.

I had much better payed job earlier, yet I wasn't this much into this hobby back then and I went for cheap pieces at online flea markets and I also kept what I got from friends. But I have nice pieces from that era too, like a Slot A Athlon, which is rare by my standards (I've only seen two in my life, both are in the collection). If I still had a salary like that now I'd surely pick up at least a few dozen pieces from this board now. Then a few more dozens, and so on

So I have a long and vague wish list (like any Pentium Pro above 256K cache, any 8087 (I had one in the early '90s), a Little Foot K6 (I had one too), a black cap K6, a silver cap 6x86, any Cyrix 5x86, Rise MP6, IDT WinChip, any Itanium, anything newer than the Wolfdale I'm typing this post on, and so on... I can only dream of a zebra 4004, I guess )

I also like to test the more popular pieces if they work. I keep a little pile of main boards in the cupboard: 486, Socket 7, Slot1, 370, 462, 478, 775, 939, with some ISA/PCI/AGP/PCIe VGAs and also a few GUS' because it's nice to see a DOS demo on the real thing. My fastest DOS demo rig is an Abit KT7A with a Palomino, this is the newest I could find with an ISA slot. A Tualatin P3 could be a nice one too, if I found a working Tualatin board with an ISA slot!

Back in the K7 days I wrote a 256-byte intro (it's called Retribution) that needs a K7 to run fluently but it's incredibly slow on any P4. It wasn't intentional even though I hated the P4 as I saw it as a bad concept for the way it worked and the way it was marketed Now that those are nice and humble pieces of my collection too I've made peace with them But they still run Retribution very slowly

(Hell, even my present Wolfdale can't run it any better than a barefooted <1GHz Duron. At least it's much faster at it than any P4 ever )

[Guest]

Mine started by simply not tossing chips I used in personal computers, then expanded into collecting any CPU that was visually interesting to me to make the wall art pictured here. But space is limited and I don't want a bunch of the same shape or color so I trade the rest. Anybody thats been in my office cares more about a shiny mobile Athlon than a 4004 sitting right next to it, so completeness, historical value or rarity don't carry much weight unfortunately.
And I like modern computer parts too much to spend much on something I only glance at once in a while. But don't get me wrong I'll spend 3 hours of blood and sweat trying to pull some $2 chip from a server I found in the trash...lol

[Glory_Cloud]

Nice wall display!!

[Guest]

Thanks, this is and older pic, at this point I think I need a bigger wall

[achtung049]

[Vlasta]

[rjluna2]

Nice display, c3h5

[Chook]

I collect in a rather haphazard fashion. My main collection in x86 only, non x86 chips are collected sporadically only if they look cool. For x86 I collect mainly from 386-Pentium 4 socket 423. Anything outside this range, I just get a few representative chips. Within the main collection range, I try to get all speeds, and all variations of markings. In most cases, I am not an sspec collector.

Recently I decided to limit the main collection to 386-P4 as it was getting too much to think about and keep track of. That is why I have sold a lot of P4 socket 478 and Xeons recently. I no longer actively collect anything newer than Socket 423 or Xeon 603 or anything older than 386. I am thinking about limiting the range further.

And on the topic question, for chips in the main collection range, I collect both rare chips and popular chips. I will often pay much more than I care to admit to myself to buy that last chip needed for a certain set.