Cool, never heard of the 337, got a 380, and since I've only seen those as Compaq surplus, or in Compaqs, I am gonna blame Compaq. I have the idea that one day the VP of marketting woke up and said "oh no, there's a gaping hole in the range between 366 and 400!" and got on the phone to AMD, who muted the phone to have a good laugh, then said "Certainly sir, minimum order is 200,000"

BitWrangler wrote on 2024-03-07, 18:36:

Cool, never heard of the 337, got a 380, and since I've only seen those as Compaq surplus, or in Compaqs, I am gonna blame Compaq. I have the idea that one day the VP of marketting woke up and said "oh no, there's a gaping hole in the range between 366 and 400!" and got on the phone to AMD, who muted the phone to have a good laugh, then said "Certainly sir, minimum order is 200,000"

Yes, well the 380 was marketed before the 400, it was a particular frequency, and it could easily be used at 375 with FSB 75, many have used them at 400, bringing the FSB to 100.
Regarding the 337, it was actually used if I remember correctly, on IBM PCs with FSB 75, perhaps it was a way to have an alternative to the corresponding Cyrix IBM 6X86MX, the peculiarity is that there is no 337 engraved on it, but 38L3054, precisely I don't know what that means, but you can see all the details here:
https://www.x86-guide.net/en/cpu/AMD-K6-2-337 … cpu-no1810.html

PC@LIVE wrote on 2024-03-07, 19:29:

Yes, well the 380 was marketed before the 400, it was a particular frequency, and it could easily be used at 375 with FSB 75, ma […]
Show full quote

BitWrangler wrote on 2024-03-07, 18:36:

Cool, never heard of the 337, got a 380, and since I've only seen those as Compaq surplus, or in Compaqs, I am gonna blame Compaq. I have the idea that one day the VP of marketting woke up and said "oh no, there's a gaping hole in the range between 366 and 400!" and got on the phone to AMD, who muted the phone to have a good laugh, then said "Certainly sir, minimum order is 200,000"

Yes, well the 380 was marketed before the 400, it was a particular frequency, and it could easily be used at 375 with FSB 75, many have used them at 400, bringing the FSB to 100.
Regarding the 337, it was actually used if I remember correctly, on IBM PCs with FSB 75, perhaps it was a way to have an alternative to the corresponding Cyrix IBM 6X86MX, the peculiarity is that there is no 337 engraved on it, but 38L3054, precisely I don't know what that means, but you can see all the details here:
https://www.x86-guide.net/en/cpu/AMD-K6-2-337 … cpu-no1810.html

Having worked with IBM hardware in the past, it looks a lot like one of their internal product numbers (FRU).

They all came out at once it appears, the 366, 380, 400... https://web.archive.org/web/20100315194219/ht … 554~815,00.html that was the CXT core, the whole point of which was to push past 400... well to push past 350, which only unicorn preCXT were doing. Though I had not seen that release before, only remembering that AMD didn't have 380 listed on their site for ages, maybe didn't show until 550 came out.

Hrmrmrmrmrm....

I picked up (what I thought was) an Acorp 6VIA85 with an Apollo Pro+ chipset for a pretty sweet price. Pic from TRW:

It hasn't arrived yet, but I looked at the listing pic again and I must have been a little buzzed when I dropped the coin because I missed seeing the intel chipset on the mobo: 🤦‍♂️

So, for some reason, Acorp decided to release a near identical motherboard, down to the silkscreened model name, but with a 440ZX chipset. Intel-based motherboard, but keep "VIA" in the name... sure, why not??

(Yes, yes pedants can say that the "/" in the silkscreened name means it covers more than one model, but jeez)

With these it's usually a safe bet to check if there's a sticker, and the naming convention goes like this in that case:

6VIA85P - Apollo Pro+
6ZX85 - 440ZX

And it's not a near identical mobo. It's the same PCB used for both of the boards! They reused the PCBs and just changed the chipsets at their free will - it appears the Apollo Pro Plus and the 440s likely share the same BGA pattern.

At one point I jokingly called these "the Apollo 440s", after the "Stop The Rock" song. Simply because they support both chipsets on the same PCB.

There's also the 6VIA81P and the 6BX81. You can tell which is which - former runs VIA693, latter runs the well known "Seattle" chipset.

vmunix wrote on 2024-03-07, 13:37:

badmojo wrote on 2024-03-06, 10:49:

Socket3 wrote on 2024-03-05, 08:37:

The roland card appears to be an SCC-1. I don't know much about it, apart from it's supposed to be sort of an internal version of the SC-55.

Internal SC-55 + intelligent mode MPU401... and it's worth a lot of money these days.

Same thought when I saw it, that, the PAS16 and the ESS with wavetable already worth whatever you have paid for the lot, imho.

I played a little over 200 euro for the whole lot. There was more stuff in that lot I did not post in the thread:

Apart from these there was also an SB16 vibra (the version with a real OPL3 chip), a Radeon 9000 PRO (working), an AWE64 value with a couple of damaged caps (10uf and 100uf, easy to source), a low profile 128bit geforce 2 MX400, a pixelview Geforce 4 MX440 with 3.6ns BGA ram (working, overclocks to MX460 clocks no problem) and a radeon 2600XT AGP (also working).

Out of all the cards I got, so far:

Gigabyte GA 7DPXDW-P dual socket A board - working. Missing northbridge heatsink. Had to replace a couple of 2200uf capacitors, they were bent quite a bit, partially exposing the plates. The board posts with anything up to Athlon XP 2600+. Unfortynatly I can't test dual CPU functionallity since I don't own any athlon MP's. This board alone was (to me) worth buying the lot, since it fulfils a highschool fantasy. Back then I dreamt about owning a dual Athlon MP system with 1GB of ram, dual athlon MP 2000+ and a Geforce 4 Ti4600. Now I'll have to source a pair of athlon MPs - luckily I've found a few sources.

Acorp MVP3 board - working.

SiS socket A board - missing some SMD components on the back. So far I've only given it a bath. In the pile of "to be repaired" it goes.

ECS P6S5AT - dead. Came with two 1500uf caps ripped completely off. Recapped and washed the board, re-flashed bios, no POST. There is activity on the board, the CPU and chipset get warm, but it refuses to POST. I also lost my post code card when I moved my stuff almost a year ago and have yet to order a replacement. I think it might be repairable, but it's beyond my current skills right now.

Leadtek Winfast FX 5600 - working. Noisy fan. Another card I've been looking for.

Leadtek Winfast FX 5500 - missing SMD's. Untested.

Pixelview FX 5600XT - missing SMD's. Untested.

Gecube? Radeon 9600XT - artifacts.

Powercolor Radeon 9000 PRO - missing a couple of SMD caps. Replaced missing parts, working.

Palit Radeon 9100 - 4 missing SMD's. Untested.

Club3d 6600GT AGP - several missing SMD's. Untested.

EVGA Geforce 6800GT AGP - posts, driver installs, freezes in 3D or black screen. Works fine in 2d.

Skywell Vodoo 1 - working

Voodoo 3 3000 AGP - working

SiS 305 - working

Geforce 2 MX400 low-profile 128 bit - working

Geforce 4 MX440 pixelview - working

Creative Vibra16 - working

Creative AWE64 value - 2 damaged capacitors. Untested

Roland SCC-1 - working

Pro Audio Spectrum 16 - working

BTC (behaviour tech) ESS ES1868F + Wavetable ISA card - working. At least the sound blaster part, can't get the wavetable going, but I haven't tried for more then 30 mins.

Radeon x1900 - working

Digital Reasearch pentium 1 PC - untested

K6-III 400 CPUs - working. Took forever to get all the pins streight, but it was worth it.

PC-CHIPS board with on board VIA C3 800 - 1 working, 1 dead.

Most of the time over half the parts I get from this recycling center are dead or beyond my ability to repair - but they are cheap. But other times, I get really really lucky and find rare stuff in working condition or repairable. It's a gamble I know, but it still beats ebay prices. Hell, the dual-skt A board alone is 200$ on ebay, not to mention the SCC-1. My voodoo 5's (witch are now long sold) and ATI Rage Fury MAXX came from this source as well. If you don't mind spending a whole day half burried in piles of e-waste, you can still find rare hardware for good prices, but recyclers are getting expensive too. 5 years ago I would have payed 5e per addon card and 10e per motherboard...

I might be wrong but if the "Acorp MVP3 board" is the one above that "Pentium MMX" case in the 3rd photo, that's a Luckytech board, not an Acorp.

PcBytes wrote on 2024-03-07, 22:57:

With these it's usually a safe bet to check if there's a sticker, and the naming convention goes like this in that case: […]
Show full quote

With these it's usually a safe bet to check if there's a sticker, and the naming convention goes like this in that case:

6VIA85P - Apollo Pro+
6ZX85 - 440ZX

And it's not a near identical mobo. It's the same PCB used for both of the boards! They reused the PCBs and just changed the chipsets at their free will - it appears the Apollo Pro Plus and the 440s likely share the same BGA pattern.

At one point I jokingly called these "the Apollo 440s", after the "Stop The Rock" song. Simply because they support both chipsets on the same PCB.

There's also the 6VIA81P and the 6BX81. You can tell which is which - former runs VIA693, latter runs the well known "Seattle" chipset.

Wehhhhhh... I was looking for a baby AT that has some kind of 133MHz FSB support. 😐 I don't know much about the ZX chipset other than "inferior to the almighty BX, and therefore best avoided"

Given that you'd improve the reliability of some BX boards by ripping the 3 and even 4th that some had RAM slots off, the ZX being capped at 2 slots 512MB didn't matter a bit. the ZX-66 however was "May as well have bought a 440LX"... There were some ZX boards though that were designed as celeron boards and had other stuff cheaped out and maybe were ppga instead of fcpga even, but mostly the BX boards with a ZX on instead were a smart buy. I never suspected before that the apollo and ZX were same footprint though.

PcBytes wrote on 2024-03-07, 23:16:

I might be wrong but if the "Acorp MVP3 board" is the one above that "Pentium MMX" case in the 3rd photo, that's a Luckytech board, not an Acorp.

I think you're right, it looks a lot like the P5MVP3 from lucky tech - https://theretroweb.com/motherboards/s/acorp-5via77

Acorp didn't make their own boards as far as I know, they rebranded PCChips and other motherboards as their own.

There are other boards using both, VIA and Intel chipsets, on one and the same layout.
MS-6156 for example. Mostly this one comes with a VIA chipset (maybe the (cheaper) OEM version) but sometimes it can be found with Intel.

https://theretroweb.com/motherboards/?page=1& … =0&name=MS-6156

Took me a while to realize that both chipsets share the same footprint. A bit hard to believe on first sight.

Socket3 wrote on 2024-03-08, 06:33:

PcBytes wrote on 2024-03-07, 23:16:

I might be wrong but if the "Acorp MVP3 board" is the one above that "Pentium MMX" case in the 3rd photo, that's a Luckytech board, not an Acorp.

I think you're right, it looks a lot like the P5MVP3 from lucky tech - https://theretroweb.com/motherboards/s/acorp-5via77

Acorp didn't make their own boards as far as I know, they rebranded PCChips and other motherboards as their own.

Nope, they did make their own boards. Look at the 5VIA77 - it's their own design of a MVP3 mobo.
Most of their P2/P3 designs are their own, with very few rebranded boards, and their rebrands were usually Fordlian boards, which were rather reliable IMO.
Acorp almost never rebranded PCChips boards, for that matter.

BitWrangler wrote on 2024-03-08, 02:18:

Given that you'd improve the reliability of some BX boards by ripping the 3 and even 4th that some had RAM slots off, the ZX being capped at 2 slots 512MB didn't matter a bit. the ZX-66 however was "May as well have bought a 440LX"... There were some ZX boards though that were designed as celeron boards and had other stuff cheaped out and maybe were ppga instead of fcpga even, but mostly the BX boards with a ZX on instead were a smart buy. I never suspected before that the apollo and ZX were same footprint though.

I'd kinda disagree with one bit on the ZX-66 - as far as I know, it's a good thing it's derived off BX - LX cannot handle SSE while ZX can, so as a added bonus a PGA370 would be able to support some of the earlier 66FSB Coppermine Celerons - namely 566, 600, 633, 700. Same chips would lock up on a LX due to it not knowing what to do with these instructions.

ChrisK wrote on 2024-03-08, 07:59:

There are other boards using both, VIA and Intel chipsets, on one and the same layout. MS-6156 for example. Mostly this one come […]
Show full quote

There are other boards using both, VIA and Intel chipsets, on one and the same layout.
MS-6156 for example. Mostly this one comes with a VIA chipset (maybe the (cheaper) OEM version) but sometimes it can be found with Intel.

https://theretroweb.com/motherboards/?page=1& … =0&name=MS-6156

Took me a while to realize that both chipsets share the same footprint. A bit hard to believe on first sight.

That was probably VIA trying to compete with Intel at the time. Shame the 693A wasn't the brightest Slot1/370 chipset you could get - it looks promising enough until you remember it's considerably slower than the same board on a 440BX chipset.

(e.g DTK's PRM-27IV and PRM-27I are the best examples of my theory - both use the same PCB, except one is VIA693/596 and the other is 440BX. Oh, and the VIA version omits the 4th SDR slot.)

Socket3 wrote on 2024-03-07, 23:03:

vmunix wrote on 2024-03-07, 13:37:

badmojo wrote on 2024-03-06, 10:49:

Internal SC-55 + intelligent mode MPU401... and it's worth a lot of money these days.

Same thought when I saw it, that, the PAS16 and the ESS with wavetable already worth whatever you have paid for the lot, imho.

dt.jpg

Dang it! you scored 2 Venturis !! it took me ages to score a single one and not exactly the prettiest.
Edit: Celebris not Venturis, mine is a Venturis not sure the difference, the case is the same

Today I got those for free:
-SONY CDU701 32x IDE CD-ROM drive from June 1998
-TEAC CD-W552E 52x24x52x IDE CD-RW drive from April 2003
Any opinions?

PC@LIVE wrote on 2024-03-07, 17:42:

I found a K6-2-533 for sale, for a MB Camaro, but if you want you can use it in other boards, as long as it has FSB 97, otherwis […]
Show full quote

I found a K6-2-533 for sale, for a MB Camaro, but if you want you can use it in other boards, as long as it has FSB 97, otherwise you have to use it at 95 or 100.
With this example, I would have all the K6-2s from 233 to 550, there are also other examples with strange frequencies, for example the 337, I'm missing that one.
The second purchase, not very old, is an ASROCK Micro MB with FM2 socket, I currently don't have any suitable CPUs, I think I'll have to get at least one in the near future.
But I have several CPUs arriving, today a Core2 Duo E4700 @2.60GHz arrived, I think I'll try raising the FSB from 800 to 1066, just to see how it goes.

Also, there was 380 (as mentioned by others here as well), and 475. Both as 95 MHz FSB. Furthermore, 333 was also released as a 95 MHz version (or maybe it was just the regular 333 supporting 3.5x multiplier as well?)

RetroPC_King wrote on 2024-03-08, 12:03:

Today I got those for free: -SONY CDU701 32x IDE CD-ROM drive from June 1998 -TEAC CD-W552E 52x24x52x IDE CD-RW drive from April […]
Show full quote

Today I got those for free:
-SONY CDU701 32x IDE CD-ROM drive from June 1998
-TEAC CD-W552E 52x24x52x IDE CD-RW drive from April 2003
Any opinions?

they're free anyway 😀

i'm sure they are fine, but maybe clean inside and out if they are working, to make sure they are in best condition

then they will make fine additions to any vintage machine

I was looking for a slot 1 BX board to replace the Via one i was using.
I found this lot for 9€+4€ shipping :

Abit BE6 and BH6. Boards are untested but said to be unused. Super clean, caps looks ok. I'm optimistic.

Will use one of them with my Celeron 300A@450.

PC@LIVE wrote on 2024-03-07, 17:42:

there are also other examples with strange frequencies, for example the 337, I'm missing that one.

That one came in two versions. The standard K6-2/337AFR and one with an IBM part number, K6-2/337 38L3054. Neither is very common. I used to have both, but have very few K6 chips these days.

H3nrik V! wrote on 2024-03-08, 12:23:

PC@LIVE wrote on 2024-03-07, 17:42:

I found a K6-2-533 for sale, for a MB Camaro, but if you want you can use it in other boards, as long as it has FSB 97, otherwis […]
Show full quote

I found a K6-2-533 for sale, for a MB Camaro, but if you want you can use it in other boards, as long as it has FSB 97, otherwise you have to use it at 95 or 100.
With this example, I would have all the K6-2s from 233 to 550, there are also other examples with strange frequencies, for example the 337, I'm missing that one.
The second purchase, not very old, is an ASROCK Micro MB with FM2 socket, I currently don't have any suitable CPUs, I think I'll have to get at least one in the near future.
But I have several CPUs arriving, today a Core2 Duo E4700 @2.60GHz arrived, I think I'll try raising the FSB from 800 to 1066, just to see how it goes.

Also, there was 380 (as mentioned by others here as well), and 475. Both as 95 MHz FSB. Furthermore, 333 was also released as a 95 MHz version (or maybe it was just the regular 333 supporting 3.5x multiplier as well?)

I'm aware of four 475MHz K6-2 and two K6-2+. I have a K6-2/475ACK (that actually a K6-2-P) in my collection. There are two different K6-2/380 (one is a K6-2-P in the manuals)