• Please review our updated Terms and Rules here

ID brand for me: CVGA-V3-256, cirrus logic ISA VGA

kishy

Veteran Member
Joined
Aug 22, 2009
Messages
1,065
Location
Windsor, ON Canada
Quick question, perhaps not-so-quick answer.

I have a 16-bit ISA VGA video card. It has a Cirrus Logic GD5320 chip. It functions in both 8 and 16 bit slots without jumper changes, though it gets pretty glitchy in an 8 bit slot and I suspect one of the jumpers may help fix this but I need to know what they do.

On the back it is marked "CVGA-V3-256". The card's colour is sort of yellow, similar to two I/O cards I have from a company called JPN. However, Googling for this particular model finds one relevant result and it's useless.

Worth noting: the sticker on the BIOS chip says Cirrus Logic, so if it was made by a "rebrander", they sure didn't DO much to it.

Anyone recognize the naming pattern or happen to know much about JPN's product lines?



Better yet, anyone think they know what the settings are?
4 jumpers with 2 positions each (1-2, 2-3). JP6, JP7, JP8 and JP11.
There's also a 2-pin jumper, JP10. Neither JP11 or JP10 have shunts on them and the card has been working OK like this.
 
I've got some old CL ISA cards, but they're pretty much all GD54xx ones. There are two common jumpers--one for "turbo mode" (to match the CPU/bus speed, I'm guessing) and one for the IRQ2/9 frame end interrupt.
 
Quick question, perhaps not-so-quick answer.

I have a 16-bit ISA VGA video card. It has a Cirrus Logic GD5320 chip. It functions in both 8 and 16 bit slots without jumper changes, though it gets pretty glitchy in an 8 bit slot and I suspect one of the jumpers may help fix this but I need to know what they do.

On the back it is marked "CVGA-V3-256". The card's colour is sort of yellow, similar to two I/O cards I have from a company called JPN. However, Googling for this particular model finds one relevant result and it's useless.

Worth noting: the sticker on the BIOS chip says Cirrus Logic, so if it was made by a "rebrander", they sure didn't DO much to it.

Anyone recognize the naming pattern or happen to know much about JPN's product lines?



Better yet, anyone think they know what the settings are?
4 jumpers with 2 positions each (1-2, 2-3). JP6, JP7, JP8 and JP11.
There's also a 2-pin jumper, JP10. Neither JP11 or JP10 have shunts on them and the card has been working OK like this.

I´ve got a similiar adapter with the same chip components, but a different layout, and amazingly also without a back bracket. Mine only has a 2-pin jumper header however, currently open, and with the circuit trace leading to the CL-GD5320. There is also a 3-pin jumper solder pad position, hard-wired to a setting. This BIOS is by Award.

FWIW, mine is of a Chinese manufacture, with the FCCID of IATVP44. Warranty sticker is marked as November of 1991. On the front is the silkscreen ¨VP44B¨.

The Cirrus Logic chip designs commonly tied in directly to the bus, so even if other companies used the chipset, the wiring and components didn´t vary by much. IBM used a Cirrus Logic 5428 chip on one microchannel adapter (http://www.ibmmuseum.com/OhlandL/video/1MB_Short_SVGA.html), which had a noticible flicker in some environments. Peter Wendt (the same as the Dallas module patch, and whom works for IBM in Germany) studied the wiring and found a capacitor reversed (replaced with one in the correct polarity, and the flicker goes away).

Based on that, I checked the same video chipset used on the planar of the IBM Server 720. The capacitor (on the underside of the planar, so you have to tear the system down to get at it) there was also reversed. I guess IBM got the wiring diagrams for that one wrong somehow.
 
I've got some old CL ISA cards, but they're pretty much all GD54xx ones. There are two common jumpers--one for "turbo mode" (to match the CPU/bus speed, I'm guessing) and one for the IRQ2/9 frame end interrupt.

Turbo mode...lol.
Every time I see the word 'turbo' associated with a computer it kind of tickles because it's such a ridiculously bad name for what's being done. It's not like the computer's exhaust heat is being used to spin a turbine which somehow increases a frequency (though I'm tempted to investigate the possibility of doing that now so I can say 'turbo' and actually mean it)

If I described where the traces from the jumpers go, do you think you could throw out some guesses of what they may do?

About 54xx, I have a 5422 "Evaluation Board". Wondering if that's at all rare...

I´ve got a similiar adapter with the same chip components, but a different layout, and amazingly also without a back bracket. Mine only has a 2-pin jumper header however, currently open, and with the circuit trace leading to the CL-GD5320. There is also a 3-pin jumper solder pad position, hard-wired to a setting. This BIOS is by Award.

FWIW, mine is of a Chinese manufacture, with the FCCID of IATVP44. Warranty sticker is marked as November of 1991. On the front is the silkscreen ¨VP44B¨.

The Cirrus Logic chip designs commonly tied in directly to the bus, so even if other companies used the chipset, the wiring and components didn´t vary by much. IBM used a Cirrus Logic 5428 chip on one microchannel adapter (http://www.ibmmuseum.com/OhlandL/video/1MB_Short_SVGA.html), which had a noticible flicker in some environments. Peter Wendt (the same as the Dallas module patch, and whom works for IBM in Germany) studied the wiring and found a capacitor reversed (replaced with one in the correct polarity, and the flicker goes away).

Based on that, I checked the same video chipset used on the planar of the IBM Server 720. The capacitor (on the underside of the planar, so you have to tear the system down to get at it) there was also reversed. I guess IBM got the wiring diagrams for that one wrong somehow.

Bracket - yeah, I removed it a while back because I use the card for testing purposes. Generally that means the motherboard is on a flat surface so the bracket gets in the way. Funny coincidence though.

Mine's BIOS is also by Award, I should have clarified that. It is by Award but branded as Cirrus Logic on the same sticker.

Sounds like you have fewer configurable options on your card. From looking at various ISA video cards in TH99 I see a few (of no relation to this one, just random ones) that have jumpers to select 8-bit operation. Since not all (it looks like a minority, actually) cards support working in 8-bit slots it might make sense that the extra fuss on my card relates to that.

That's kind of...well, dumb on the behalf of IBM to have shipped cards with such a relatively serious problem. Did the quality control people just think it was supposed to flicker? I've noticed some inconsistencies and quality concerns with some IBM peripherals (the famous Model M included) and I'm inclined to believe the quality they were known for may have just been luck in a lot of cases...they seem to have been rather careless at times.
 
...That's kind of...well, dumb on the behalf of IBM to have shipped cards with such a relatively serious problem. Did the quality control people just think it was supposed to flicker? I've noticed some inconsistencies and quality concerns with some IBM peripherals (the famous Model M included) and I'm inclined to believe the quality they were known for may have just been luck in a lot of cases...they seem to have been rather careless at times.

They seemed to have documented it, but never figured it out (I´m inclined to believe that the incorrect wiring diagram was received by two different part designs being affected, but I don´t have enough samples of that chipset on clone boards to confirm the theory better)...

The famous ¨EC¨ wiring seems to be somewhat common from IBM in the PC line and early PS/2 (there is an early PS/2 Model 55SX called the ¨cockroach planar¨, because an EC upside-down chip glued to the board, with patch wires run to different places) system and adapters. With later PS/2s the EC patches aren´t that common. I don´t know if it was that they felt quality had improved, or stopped caring at that point.

I´ve seen the GeekHack debate, but had always thought that the black/blue-badged Model Ms were that way depending on which system they were sold with. The later 95xx-series PS/2s had blue badges, so the keyboard badge would typically match what the system had. As we discussed, those later PS/2s treated keyboards differently, but ironically the Host-Connect and Space-Saver keyboards only had black badges.
 
I recall some of the major Asian manufacturers of the late 80's and early 90's that made generic VGA cards were - Pine Technology and Oak - yes they're both named after trees. Anyway, Oak made cards with their own graphics chips, so it might not be Oak. I do recall Pine making a lot of generics.

I went thru this whole list looking for your layout but still couldn't find it...but I might have missed it.

http://stason.org/TULARC/pc/graphics-cards/
 
They seemed to have documented it, but never figured it out (I´m inclined to believe that the incorrect wiring diagram was received by two different part designs being affected, but I don´t have enough samples of that chipset on clone boards to confirm the theory better)...

Funny that they never figured it out but observant and 'techy' end users did. Oh well, can't fault the big corporation for not properly investigating such a malfunction lol

I´ve seen the GeekHack debate, but had always thought that the black/blue-badged Model Ms were that way depending on which system they were sold with. The later 95xx-series PS/2s had blue badges, so the keyboard badge would typically match what the system had. As we discussed, those later PS/2s treated keyboards differently, but ironically the Host-Connect and Space-Saver keyboards only had black badges.

It would seem that the label colour was based on time...which coincided with the machines of course, but there were probably some instances where an older keyboard got paired with a newer machine or vice versa resulting in mixed up logo colours. Ultimately there are minor internal differences between the years but nothing hugely significant hinging on the logo colour.

I recall some of the major Asian manufacturers of the late 80's and early 90's that made generic VGA cards were - Pine Technology and Oak - yes they're both named after trees. Anyway, Oak made cards with their own graphics chips, so it might not be Oak. I do recall Pine making a lot of generics.

I went thru this whole list looking for your layout but still couldn't find it...but I might have missed it.

http://stason.org/TULARC/pc/graphics-cards/

Pine usually throws a model number starting in "PT-" on anything...I have a Realtek video card they made which is labeled PT-505S and a Pine motherboard PT-428. Of course, they do make or at least resell generic products (up until recently you could still find Pine-branded boxes with 56k modems, simple sound cards, etc in the local Walmart...the cards in the box would clearly be by another manufacturer and would use drivers from another manufacturer)

Oak...I think they put OTIVGA on most of their video products. I don't know where it went but I had an Oak video card.

It may sound crazy but I think the card colour is a major help in identifying it. Cards in this yellow colour seem to be quite a bit less common than green.

I have searched TH99 before so that makes two of us who didn't find it there.
 
I've got some old CL ISA cards, but they're pretty much all GD54xx ones. There are two common jumpers--one for "turbo mode" (to match the CPU/bus speed, I'm guessing) and one for the IRQ2/9 frame end interrupt.

Do you have the Windows 3.11 driver for any of these, maybe the 54222 Chip?
 
Back
Top