Anyone know what this NuBUS card might be?

[Phipli]

This is what I've thought as well from my research. My card came with 16-bit working via its original 341-0266 ROM and even after PRAM reset, all I could get with 832 x 624 sense pins was a black screen when 341-0812 was inserted into the same card. Not even mono or 8-bit.

It sounds like someone pinched the Rev B ROM out of your card?

I think mg.man's ROM is also Rev A to be honest, excuse it is copyright 1990, but it could be from later in the year .

 

[jeremywork]

Does anyone know what Brooktree part that Bt RAMDAC really is, underneath its Apple disguise?

I don't know the answer to this, and the following might be of no help, but Dale Adams had a bit of information about the 8•24's chunky planar scheme which might at least narrow the search?

[of the Quadra 700/900]

In a number of cases the design was optimized for high performance over low cost. A good example of this is 32 bpp operation on Apple's standard 13-inch RGB monitor at 640 x 480 resolution (and this also applies to VGA and NTSC), which is probably the most common color monitor in use on the Macintosh. The actual number of memory bytes needed to support 24 bpp is 640 x 480 x 3 = 921,600. This would seem to fit within 1 MByte of memory (as is the case with the Apple 8*24 video card), but the Quadras actually require 2 MBytes of VRAM for this mode. The 8*24 card supports 24 bpp at 640 x 480 by using a storage mode called 'chunky planar' to fully utilize all its 1 MByte of VRAM. However, this results in having to perform 3 separate memory accesses for each 24-bit pixel read from or written to the frame buffer. (This is done in hardware so software only performs a single read or write.) On a NuBus video card, this inefficiency is partially masked by the synchronization delays which occur at the processor-bus/NuBus interface. However, when frame buffer memory is placed directly on the processor bus, this approach results in a nearly 3X performance degradation. This was judged unacceptable for the Quadras. Each 24-bit pixel occupies one longword (4-bytes) in VRAM, so the Quadras actually provide 32 bpp for the 640 x 480 resolution. This pushes the memory requirement for this mode over the 1 MByte boundary (640 x 480 x 4 = 1,228,800 bytes)

 

[jeremywork]

It sounds like someone pinched the Rev B ROM out of your card?

Except with 341-0266 I have all of the features indicated to be available only on Rev B ROMs and Rev B cards. 832 x 624 works fine, and 16-bit is enabled on both 832 x 624 and 640 x 480.

 

[Phipli]

information about the 8•24's chunky planar scheme

Careful, it probably doesn't make a difference in this instance, but that quote is talking about the 8•24, not the 8•24 GC.

 

[Phipli]

all I could get with 832 x 624 sense pins was a black screen when 341-0812 was inserted

Except with 341-0266 I have all of the features indicated to be available only on Rev B ROMs and Rev B cards. 832 x 624 works fine, and 16-bit is enabled on both 832 x 624 and 640 x 4480.

Sorry, misread. I thought you meant 832x624 never worked.

 

[mg.man]

It sounds like someone pinched the Rev B ROM out of your card?

I think we're getting wires-crossed... according to : https://support.apple.com/kb/TA48238?locale=en_US - the -0266 *is* the Rev.B ROM?

 

[jeremywork]

Careful, it probably doesn't make a difference in this instance, but that quote is talking about the 8•24, not the 8•24 GC.

That brings up an interesting point though...

I have an early example of a 4•8 (1986-90) and a later 8•24 (1986-91) which both use Bt 357S0010-A, same as mg.man's 8•24 GC. My GC however, has an AMD 343S1075, which probably is the hardware difference required to support 832 x 624.

It still stands to reason that if research is correct, inserting 341-0622 into the Rev A card should enable 16-bit color in 640 x 480 mode.

Edit: I've been backwards.

It's 834 x 624 which becomes enabled via ROM. The revision B hardware (revised RAMDAC) is needed for 16-bit color support.

 

[Phipli]

I think we're getting wires-crossed... according to : https://support.apple.com/kb/TA48238?locale=en_US - the -0266 *is* the Rev.B ROM?

That's what I thought we thought . Given the copyright is newer.

 

[Phipli]

That brings up an interesting point though...

I have an early example of a 4•8 (1986-90) and a later 8•24 (1986-91) which both use Bt 357S0010-A, same as mg.man's 8•24 GC. My GC however, has an AMD 343S1075, which probably is the hardware difference required to support 832 x 624.

It still stands to reason that if research is correct, inserting 341-0622 into the Rev A card should enable 16-bit color in 640 x 480 mode.

As long as nothing else changed.

 

[jeremywork]

As long as nothing else changed.

I already just realized I had it backward

Upgrading the ROM should add 832 x 624, but only the Rev B hardware gets 16-bit.

 

[Phipli]

@MrFahrenheit's is the same 91 copy right ROM and RAMDAC pair :

https://68kmla.org/bb/index.php?attachments/ddaef46c-903e-47c2-bf86-916a49134938-jpeg.47801/

 

[MrFahrenheit]

@MrFahrenheit what ROM does your board have and what silkscreen dates on the board?

This one ?

 

[Phipli]

I already just realized I had it backward

Upgrading the ROM should add 832 x 624, but only the Rev B hardware gets 16-bit.

Sorry, I wasn't paying enough attention and didn't notice

Wish that they'd done 16bit on the Rev B vanilla 8•24s too. They added 832x624, but not 16bit.

 

[mg.man]

The serial number of my card does start with CA2... Still, give it a go.

OK... so... Round 1 == my card as received :



And... it's working!! For me, that's a result given it took ~3mths by boat to get here!

Now... it does look like a Rev. A (Rev ... 1.0), there's no 16bit (Thousands of Colors, right?), and setting my adapter to 832x624 just gives a blank screen.

I have tracked down a 27C512, I hope a -150 is fast enough. Just waiting for the UV eraser to do its thing and will report back.

 

[Phipli]

OK... so... Round 1 == my card as received :

View attachment 52856 View attachment 52857

And... it's working!! For me, that's a result given it took ~3mths by boat to get here!

Now... it does look like a Rev. A (Rev ... 1.0), there's no 16bit (Thousands of Colors, right?), and setting my adapter to 832x624 just gives a blank screen.

I have tracked down a 27C512, I hope a -150 is fast enough. Just waiting for the UV eraser to do its thing and will report back.

If you don't have them all, drivers in an old post here :

https://68kmla.org/bb/index.php?threads/wtb-apple-8-24-gc-nubus-graphics-card.42222/post-458537

Note the card is technically only supported up to 7.1 and is buggy in newer systems.

 

[mg.man]

OK... so... Round 2 :



The latter pic is running at 832 x 624!! But... no 16-bit / Thousands...

Oh well, you can't win them all! Thanks for the ride... and for the ROM @jeremywork!

 

[Phipli]

OK... so... Round 2 :

View attachment 52859 View attachment 52860

The latter pic is running at 832 x 624!! But... no 16-bit / Thousands...

Oh well, you can't win them all! Thanks for the ride... and for the ROM @jeremywork!

Still more graphics card than most Mac IIs can dream of

 

[jeremywork]

OK... so... Round 2 :

View attachment 52859 View attachment 52860

The latter pic is running at 832 x 624!! But... no 16-bit / Thousands...

Oh well, you can't win them all! Thanks for the ride... and for the ROM @jeremywork!

It did begin to make perfect sense that it'd be a Rev A, but I'm glad the ROM swap gave you something new!
On the bright side, the 8•24 GC was optimized for 8-bit and Millions before Thousands existed, so even with 16-bit color is enabled it's usually not visually faster than 24-bit. Mostly a bummer if you intend to run 832 x 624 primarily, but I wouldn't miss it that much in 640 x 480.

Also appreciate your help confirming 341-0812 is the Rev A ROM and 341-0266 is the Rev B.
I'm not sure there ever was a third ROM; was just poking LEM's statement that both 0812 and 0266 provided Rev B features.

 

[jeremywork]

Does anyone know what Brooktree part that Bt RAMDAC really is, underneath its Apple disguise?

I've done some more guess-searching on this...

The 8•24 has thicker traces for each of the color channels reaching the DA-15 plug. They're each routed behind L3 on the backside, and reach pins 27, 28, and 29 of the PLCC68. Pin 26 connects to the ground plane and Pin 20 connects through R22 on the backside to C45, along with D7 on the front near the ROM.



The pathing on the earlier 4•8 is a bit more comvoluted but corroborates the 8•24.



I did a quick image search for 'plcc 68 ramdac brooktree' and found two PLCC68 RAMDACs, Bt473 and Bt485.

To disclaim, I'm out of my depth here; Bt485 seems to support more features than the Nubus cards were sold for, though Bt473 seems more likely to fit the description. Neither fit the pinout required by the 4•8 and 8•24, though. Maybe still a good jumping-off point?

 

[Melkhior]

To disclaim, I'm out of my depth here; Bt485 seems to support more features than the Nubus cards were sold for, though Bt473 seems more likely to fit the description. Neither fit the pinout required by the 4•8 and 8•24, though. Maybe still a good jumping-off point?

The BT484/485 are unlikely candidates, they are probably newer than the 8*24 GC - they're not listed in the '89 or '91 databooks, but appear in the '93 handbook (see on bitsavers) with a 84-pins PLCC package. BT473 is the only one I find in the databooks with a 68-pins PLCC package, but as you say that's the wrong pins on the board for the analog outputs of the BT473.

OTH, Brooktree made a lot of products, and not all variants may have made it to the databooks - I've seen devices with higher frequency labeling than the databooks say are available. It's also possible Apple was using a custom package for otherwise 'standard' BT silicon. Ot hey could have had the silicon customized.

Theoretically, it should be possible to reverse-engineering the ROM to figure out how the RAMDAC is set up (color palette, ...) and so to which family of BT RAMDACs it belongs. But that would be a lot of work.