BitWrangler wrote on 2025-05-01, 04:19:

I had an HD4670 in 2009 and it seemed like more GPU than I knew what to do with, as I was still picking up cheaper/older games, not paying release prices. Also was at modest monitor res. So seemed like overkill just that. For XP/Vista/Early7 builds sticking with "office" resolution monitors, they still seem to cover a wide range adequately, as long as it's okay on DX10. Heck they still come into consideration for near modern, since they do as well as a 440GTX or something.

Well I'm not sure how to put this but the 440GT is 15 years old at this point and only a few years older than the 4670, neither is really near modern, the GTX 700 series might pass as near modern but it is EOL so the GTX900 series would fit as near modern since it still has driver support IIRC.

Though the GTX 10 series is apparently losing driver support next year and being made EOL.

I think we all know why nvidia has made that decision, they want to GTX 1080ti to die, they simply made that card far too good and even in modern games it still matches the RTX 3060.

750 and 750ti also belong to Maxwell and are supported by drivers like 900.

Trashbytes wrote on 2025-05-01, 04:39:

Well I'm not sure how to put this but the 440GT is 15 years old at this point and only a few years older than the 4670, neither […]
Show full quote

BitWrangler wrote on 2025-05-01, 04:19:

I had an HD4670 in 2009 and it seemed like more GPU than I knew what to do with, as I was still picking up cheaper/older games, not paying release prices. Also was at modest monitor res. So seemed like overkill just that. For XP/Vista/Early7 builds sticking with "office" resolution monitors, they still seem to cover a wide range adequately, as long as it's okay on DX10. Heck they still come into consideration for near modern, since they do as well as a 440GTX or something.

Well I'm not sure how to put this but the 440GT is 15 years old at this point and only a few years older than the 4670, neither is really near modern, the GTX 700 series might pass as near modern but it is EOL so the GTX900 series would fit as near modern since it still has driver support IIRC.

Though the GTX 10 series is apparently losing driver support next year and being made EOL.

I think we all know why nvidia has made that decision, they want to GTX 1080ti to die, they simply made that card far too good and even in modern games it still matches the RTX 3060.

Yeah I'm thinking near modern like playing older still popular stuff like Rocket League and CS:GO at 720p -ish res, 1280x900 office monitors.

Got these for free:

- ASRock 775i945GZ - AGI Express strikes again!
- Gigabyte FX5200 - needs a VGA header
- HIS Radeon 9250 256MB
- 3dfx Velocity 100 8MB SGRAM

PcBytes wrote on 2025-05-01, 18:31:

Got these for free: […]
Show full quote

Got these for free:

- ASRock 775i945GZ - AGI Express strikes again!
- Gigabyte FX5200 - needs a VGA header
- HIS Radeon 9250 256MB
- 3dfx Velocity 100 8MB SGRAM

These days if you manage to get anything 3dfx for free, it is an amazing win!

Couple of weeks ago I manged tp get Gainward Banshee in a lot with other stuff for 20€ all.

Finally bought a PowerVR PCX2. Its a Matrox M3D REV: A

I paired it with a Virge DX and ran benchmarks. It is smooth and had no issues in Final Reality, but is missing half of the textures in 3Dmark 99.

I havent played SGL capable games on it yet, but it came with an Evaluation CD that has about 10 game demos I can try later.

How about an ASUS P3B-F 1.03 with what is possibly a PIII 500 and 2x dual sided ram sticks for 35 USD ?

No idea if it works as its sold as is, but these are known to be stupidly sturdy motherboards so fingers crossed.

Its a gamble but I couldnt resist for such a low price and it looking in good condition.

Another board added to the lot today, a Gigabyte 6BXC rev 2.0.

Needs some work (missing BIOS and CMOS holder, as well as cleaning) but otherwise it's in good shape IMO.

Hi

I found this today and I have no idea what it is
looks like some fancy and rare graphics card
It says "Boca Vortek" and I am DEFINITELY not going to peal the label to see what chip is underneath

I checked online and there seems to ba a VLB card with similar name but thats as fas as I researched

There are some bent pins on the main chip so I will carefully repair them on the next days before attempting to turn it on

if someone can help me identify it that would be great
I promise I will upload all the possible information to the Retro Web

cheers

Boca Research's iteration of a IIT AGX014 card, according to a russian forum post from IXBT. There's a VLB variation of it too.

PcBytes wrote on 2025-05-02, 21:35:

Boca Research's iteration of a IIT AGX014 card, according to a russian forum post from IXBT. There's a VLB variation of it too.

thank you for the help 😀

Trashbytes wrote on 2025-05-02, 11:16:

How about an ASUS P3B-F 1.03 with what is possibly a PIII 500 and 2x dual sided ram sticks for 35 USD ? […]
Show full quote

How about an ASUS P3B-F 1.03 with what is possibly a PIII 500 and 2x dual sided ram sticks for 35 USD ?

The attachment ASUS P3B-F.jpg is no longer available

No idea if it works as its sold as is, but these are known to be stupidly sturdy motherboards so fingers crossed.

Its a gamble but I couldnt resist for such a low price and it looking in good condition.

Nice ! Hope you figure it out. Maybe something simple or my guess is a bad bios, bad caps or bad vrm....

Horun wrote on 2025-05-03, 02:38:

Trashbytes wrote on 2025-05-02, 11:16:

How about an ASUS P3B-F 1.03 with what is possibly a PIII 500 and 2x dual sided ram sticks for 35 USD ? […]
Show full quote

How about an ASUS P3B-F 1.03 with what is possibly a PIII 500 and 2x dual sided ram sticks for 35 USD ?

The attachment ASUS P3B-F.jpg is no longer available

No idea if it works as its sold as is, but these are known to be stupidly sturdy motherboards so fingers crossed.

Its a gamble but I couldnt resist for such a low price and it looking in good condition.

Nice ! Hope you figure it out. Maybe something simple or my guess is a bad bios, bad caps or bad vrm....

Might also work just fine, seller had no way of testing it. Had a good look at the other pics and no sign of bad caps or any board damage.

Doesn’t mean it ain't broken but it's got better odds than a lot of the untested gear I see on Evilbay. !

I got six motherboards for cheap ( all for less than $7). Two are dead, because one (not shown int the photos below) is missing a pair of PALs and another has the vram chips torn and a cache chip burst by an internal short-circuit, those are may be for spares... Also, all the cache chips that were in sockets are also missing. Hope those ones works without cache, as I have some chips but not the necessary amount nor all types.
Two are Socket 4, one is a Socket 5.
From 4 of them I already sent better images to theretroweb, and from 2 the bios. With one of them a new entry was created on the web, as "[Unknown] OPTI 386-16".
I have started restoring this 386. It is slightly damaged by the battery (easy compared to the one I restored before in which some lines had already disappeared eaten away). this and a MiTAC/Trigon PWA-PB5500C are the ones that interests me the most.

kinetix wrote on 2025-05-03, 03:05:

I got six motherboards for cheap. Two are liquidated, because one is missing a pair of pal and another has the vram chips torn and a cache chip burst by an internal short-circuit. Also, all the cache chips that were in sockets are also missing. Hope those ones works without cache, as I have some chips but not the necessary amount nor all types.
From 4 of them I already sent better images to theretroweb, and from 2 the bios. With one of them a new entry was created on the web, as "[Unknown] OPTI 386-16".
I have started restoring this 386. It is slightly damaged by the battery (easy compared to the one I restored before in which some lines had already disappeared eaten away). this and a MiTAC/Trigon PWA-PB5500C are the ones that interests me the most.

Interesting stuff...

Very early socket 5 board, unusually to see one with an oscillator instead of PLL. You might need an early stepping P90 to get that going as it might not understand the later multiplier standards.

LPX 4a board looks like IBM microelectronics had something to do with it, by it's grid markings. Extra power connector might be a drive pigtail takeoff rather than input, unsure.

LPXish 3a is marked Dell, so proceed with caution in assuming the PSU connector is standard pinout.

Nice PCI/VLB board, seems a bit Asus to me, due to style of markings.

386 could be an ECS... but NEC, AST, Packard Bell and others had those funky memory/expansion slots.

BitWrangler wrote on 2025-05-03, 03:30:

Interesting stuff... Very early socket 5 board, unusually to see one with an oscillator instead of PLL. You might need an early […]
Show full quote

Interesting stuff...
Very early socket 5 board, unusually to see one with an oscillator instead of PLL. You might need an early stepping P90 to get that going as it might not understand the later multiplier standards.
LPX 4a board looks like IBM microelectronics had something to do with it, by it's grid markings. Extra power connector might be a drive pigtail takeoff rather than input, unsure.
LPXish 3a is marked Dell, so proceed with caution in assuming the PSU connector is standard pinout.
Nice PCI/VLB board, seems a bit Asus to me, due to style of markings.
386 could be an ECS... but NEC, AST, Packard Bell and others had those funky memory/expansion slots.

Thanks for the recommendation regarding the Dell.
Yes, two are Socket 4, one is Socket 5. I have a P5-60 (SX835-C1) and some P54C (75 969-C2, 100 SX963-C2, I suppose those are too late and you refer to a B1 stepping) for testing S4 and S5.
But this S5 have a cache chip blown and ripped vrams. But I have another S5 still untested.
The one not in the photos I posted is an Acer 1120SX, missing two PALs, but marked as defective anyway.
The FIC-PN2000 have the IO chip with a couple of broken legs (and signs that someone tried to fix them ), I'll try too when I have some tools. If I have no luck, I hope I could use a multifunction IO card.
For the PCI/ISA/VLB I dont have that missing small cache chip (P4C187, 64K x 1bit). I hope the bios have option to disable the cache, if the system is able to boot without it first .

Bought a new MIDI module for my collection. It is a Yamaha DTXpress. Officially this is a drum trigger module but it is also GM compatible, exactly the reason why I wanted it.

Salient wrote on 2025-05-03, 13:22:

Bought a new MIDI module for my collection. It is a Yamaha DTXpress. Officially this is a drum trigger module but it is also GM compatible, exactly the reason why I wanted it.

Looks pretty cool. Would love to hear some game samples from it.

zuldan wrote on 2025-05-03, 19:00:

Salient wrote on 2025-05-03, 13:22:

Bought a new MIDI module for my collection. It is a Yamaha DTXpress. Officially this is a drum trigger module but it is also GM compatible, exactly the reason why I wanted it.

Looks pretty cool. Would love to hear some game samples from it.

Check out my website (www.wavetable.nl). By now I updated it with the recordings from this module.

ps: forgive me about the double posts, the site gave me some internal server error 500 a couple of times but it turned out to have posted anyway.

Salient wrote on 2025-05-03, 13:22:

Bought a new MIDI module for my collection. It is a Yamaha DTXpress. Officially this is a drum trigger module but it is also GM compatible, exactly the reason why I wanted it.

And you can switch the drum sets while in use.