IIsiColorPivotII_PDS_Card_HackProject™

[bigmessowires]

That makes sense, but is probably a little crazier than I'm likely to attempt any time soon. Maybe you want to try it? Those cards are probably just going to gather dust if I hold on to them.

 

[Trash80toHP_Mini]

My cards will be splayed across the basement underneath the MoBo because the Radius Rocket/SCSI II Daughtercard will be in the NuBus Adapter topside. Like I said, your hack would be almost as radical as the Rocketized SuperIIsi™ [}] ]'> so I've got several of the RCPII/IIsi cards and an extra wrong angle connector or two on hand for my project.

Any help you can give on the cable building project will be much appreciated, I wouldn't know a diode from my elbow . . .

. . . meanwhile I'll practice my soldering on one of the Radius cards for you. [] ]'>

 

[Bunsen]

I anted up some fairly serious chips on an eBay Card Lot just to get that particular SE'30 version of the card/cable combo

Wait ... you have a working example?

 

[Bunsen]

after reading through this whole thread, I'm / confused.

You and me both. This mess is unreadable.

 

[Trash80toHP_Mini]

Wait ... you have a working example?

Of the IIsi version, several, they all work great with the SE/30 card's cable. The SE/30 card itself is untested, no SE/30 ATM, but it's not really an interesting card anyway, it's basically a full height PDS card with no passthru connector, you can't even install it on top of a NIC.

One day I'd like to get an SE/30, mainly just to try lopping off the connector section of the IIsi Card. By jumpering the few components on that section and wiring the cable connections to vias nearer the source, I think I might be able to get it to fit and maybe even to work.

I also have NuBus and LC PDS varieties in different revisions of the Pivot Cards, but I haven't spent any quality time with any of them as yet.

As for the thread being confusing, like I said when bigmess came on board: the first page covers my tests to establish the card's capabilities, on the second page zuiko chimed in and we started trying to figure out how to build his clone of my SE/30 cable and I said something about wanting to do the VGA conversion In-Cable.

Now we're discussing some of the installation options for the card(s) on this third page.

That's about all there is to it so far, the devil is in the details I presume. :-/

 

[Trash80toHP_Mini]

Surprise the thread/auction link from almost two years ago is gone, but I found the main pic and blew it up a bit, it was a tad tiny at half this size.



The tall "L shaped" card sitting next to my precious cable is the SE/30 card, but I didn't find it in the the Magic Shoe Boxen just now so I can't confirm which version it is. mcd said it was color in the trading post thread. ISTR it being an RCPII and I have a vague memory of testing it as good sticking straight up out of the IIsi.

Maybe this thread seems a bit disjointed because the project has been creeping along since July of 2011? It has pretty much been conversations between Z and myself, it'd be nice if the two of you stick around for some fun. I can get back on this part of the overall project now that the IIsi has successfully been Rocketized! [] ]'>

 

[bigmessowires]

Got my IIsi Color Pivot card today. It came in what appeared to be the original anti-static bag, with serial number stickers attached. I plugged it in and booted the IIsi, but nothing new appears in the Monitors control panel. I guess that's expected, because I don't actually have a monitor connected to the Pivot card? What's the normal behavior in a Mac with two video cards, where there's only a monitor plugged into one of them?

The monitor connector is just a plain 15-pin 0.1 inch header, so it should be easy to build a cable to match. It's actually got some kind of keying so the connector can only be inserted one way, I'm not sure if that's a standard of some kind, or something custom. At any rate it should be possible to plug in a plain 0.1 inch female header and ignore the keying, but it would be cool to use the correct keyed header if I can find it at a parts supplier.

I assume there's a 1-to-1 correspondence between the pins on that 15-pin internal header, and the pins on a Mac DB-15 monitor connector. Exactly which pins go to which is probably described in those docs you linked - haven't looked at those yet. But I'm assuming it's just a straight-through connection of wires (a big mess of wires, if you will) with no other electronics needed.

I just realized I don't actually have two monitors I can use with the IIsi, only one. So testing this may be interesting!

Here's a close-up of the video connector. It's got two narrow grooves at either end of one side, and a large notch in the middle of the other side.

 

[bigmessowires]

At a quick glance, I'd guess this DB-15 breakout board plus 15 female-to-female jumper wires should make an easy no-solder solution for $6. Then you can play "musical pins" with the jumper wires until you figure out which one goes where. I'm going to order these, and see what I can do.

http://www.ebay.com/itm/like/121110063236?lpid=82

http://www.ebay.com/itm/like/171106109778?lpid=82

 

[Trash80toHP_Mini]

I like that prototyping hardware, especially the extra wires available on that jumper harness for alternate ground lines. As far as swapping anything around goes, we've pretty much got the cable figured out except for the diode.

Now that I've got Illustrator up and running under OS9 on the Pismo, I'll see if I can pull the pinouts and diagrams together into a graphic for when your ProtoTypin'Toys arrive. We can mark it up for revision as we make progress.

If you know about diodes and crystals, please check into what we'll need for the cable build and the underclocking project, respectively.

Very glad to have you aboard Bigmess! :approve:

 

[bigmessowires]

I think the diode is only necessary if you're trying to build a cable that simultaneously converts from the Pivot connector to Mac video, *and* from Mac video to VGA video. IMHO that's not a great idea, since it means you could never use it with a vintage Mac monitor that doesn't have a VGA connector. It also means you wouldn't have any switches to change the sense lines, in order to pretend to be different monitor resolutions. Mac to VGA monitor adapters are cheap and plentiful, so why not use one?

But if you want to try it, pretty much any diode should work. Here's one: http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_36038_-1 The little stripe on the diode body shows you which way to orient it. The diode's triangle in the schematic diagram points in the direction of the stripe. I don't really understand the theory behind why there's a diode - it seems to contradict the sense code table below the diagram, on the page you linked to. But I'll assume whoever made the diagram knew what he was talking about.

Unfortunately I doubt it's possible to "underclock" the card with a crystal swap, even if you could find the right crystal and swap it with a slower one. I don't know about other modes, but at least for standard 640 x 480, the 75 Hz standard isn't just a 25% faster version of the 60 Hz standard. Compare http://tinyvga.com/vga-timing/640x480@75Hz with http://tinyvga.com/vga-timing/640x480@60Hz. The number of lines per frame is different, and some of the timing parameters aren't just a factor of 25% different between the two.

I'm not sure you could find the right speed crystal anyway. You've got a 14.316 MHz crystal now, it looks like a HC-49 package. Here's a list of potential replacements from DigiKey:

http://www.digikey.com/product-search/en?pv69=3&pv16=10079&pv16=318&pv16=319&FV=fff4000d%2Cfff8016d&mnonly=0&newproducts=0&ColumnSort=0&page=1&stock=1&quantity=0&ptm=0&fid=0&pageSize=25

If 14.316 MHz is giving you 75 Hz vertical, then to get 60 Hz vertical you'd need 11.4544 MHz. That's not a standard crystal frequency. The closest I see in that list from DigiKey is 11.52 MHz.

 

[bigmessowires]

Ah, I understand the diode now. http://support.apple.com/kb/TA21618 has a good explanation. By building a cable with a diode in it, oriented as per the diagram, you're sending the sense code for an Apple Multiple Scan 21 monitor, good for resolutions up to 1152 x 870. Reversing the diode gives you the sense code for an Apple Multiple Scan 16 Monitor (1024 x 768), and replacing the diode with a plain wire gives you the code for an Apple Multiple Scan 14 (832 x 624).

 

[Trash80toHP_Mini]

< WAG mode >

I think diode juju probably has something to do with Mac Monitor ID from sense line indications, which are modified in hardware switches on the VGA adapters.

As I understand it; the Mac's resolution "switching" is done by toolbox routines, the Monitors Control Panel would be the UI for doing so.

The reason z only sees "activity" on one sense line would be the indicator for the Pivot's orientation. Sense lines are otherwise static configurations. Multiple resolutions on the Pivot would be handled in the Monitors Control Panel.

< /WAG mode >

If underclocking the card ain't gonna work, all that means is that we're stuck with using MultiScan VGA CRTs and LCDs because "standard" 60Hz LCDs won't sync. Available Pivot Card resolutions will work fine with fixed scan rate Mac Monitors at said resolutions set on the monitor's cable/cable connector. I'll re-test that with my 12" RGB and 15" Hi-Res Mac CRTs.

edit: HEH! you beat me to it!

 

[bigmessowires]

Unsuccessful test #1. I put a jumper between pins 10 (sense line 2) and 11 (GND), which should make the Pivot card think there's an 1152 x 870 Apple 2-page display connected. But in the monitors control panel, it still only shows one monitor. Jt, does yours "just work" when you plug it in to your IIsi, or did you have to install some Radius software? Does your monitors control panel show two monitors, and let you position them relative to each other on the desktop?

Edit: also tried a jumper between 6 and 7, which should identify as either 832 x 624 or 640 x 870, depending on which source you believe. Same result, monitors control panel gives no hint of a second monitor.

 

[bigmessowires]

I just noticed you appear to have two different Pivot cards, and they're different!

Look at the area between that rectangular yellow capacitor C4 and the video connector. This post shows a card with CR8 and R14 between the capacitor and the connector, but this post shows a card with only CR6 in that space. Those aren't the same card.

Mine looks like the first example, with CR8 and R14. I don't know if that's an important difference or just a minor board rev. Do both of your cards actually work?

 

[bigmessowires]

Another unsuccessful test. Based on Arkku's report that his card only worked with the sense codes for a 832x624 16" color display, I tried another test with pins 4 and 10 connected together (sense 0 and sense 2). According to this Apple doc, that's the correct sense code for the 16" color display. Got the same result, nothing in the monitors control panel.

 

[bigmessowires]

Third unsuccessful test, and now I'm doubting that this card works at all. I bit the bullet, and hand-made a complete connector from the Pivot card to VGA, with the sense code hard-wired to appear as an 832 x 624 Apple 16" color monitor (pins 4 and 10 tied together).



I got nothing. No response from the monitor connected to the Pivot card, nothing in the Monitors control panel, just a big fat zero all around.

 

[bigmessowires]

It looks like the two different versions of this card have the 15-pin video connector reversed.

I used your photo to determine which side was pin 1, but after a closer look, I realized my connector appears to be flipped from yours. Testing with a multimeter seems to confirm this. But when I connect the NEW pins 4 and 10, I still get nothing. I fear I may have fried the card by running it with the wrong pins shorted, while I thought pin 1 was on the wrong end.

 

[Trash80toHP_Mini]

I doubt you killed anything on the card, there's no appreciable voltage (any?) coming from the Monitor to the card or vice versa and mixing up the wires shouldn't do it either AFAIK. Dunno much though, I leave that stuff to you electron pushers, I'm just a lowly electron plumber.

RCPII/IIsi Playtime is up next, I'm going to try one (all?) of my cards on all four Apple CRTs in the collection. It's been a long time . . .

. . . 8-o exactly TWO friggin' years and four months ago tomorrow, since I started this hack!

My card(s?) just work when I hook them up. I've had a slew of Radius CRTs, now LCDs and almost a drawer full of Radius VidCards, VidCap cards, Accelerators, etc. IIRC I've never installed any software, everything just worked.

Sometimes I wonder if I'm a really Macintosh collector or a radius junkie? :?:

I'll report back real soon now, just gotta clean up some space so I can fill it back up again with this stuff after one or two more tests on the current project setups.

 

[bigmessowires]

Oh snap, I got it to work! I had to connect a real monitor to the Pivot card before I could see anything - simply putting jumpers on the sense lines and then looking in the monitors control panel didn't seem to be enough. I haven't been able to test a dual-head setup yet, since I only have one monitor, but I was able to boot and run at 832 x 624 x 256 colors @ 75 Hz with the Pivot card as the sole video card.

For future reference, pin 1 on the Pivot card is the pin adjacent to the "J2" text on the board's silkscreen. The pins are:

1        RED GND
2        RED
3        C SYNC
4        SENSE[0]
5        GREEN
6        GREEN GND
7        SENSE[1]
8        N.C.
9        BLUE
10       SENSE[2]
11       C & V SYNC GND
12       V SYNC
13       BLUE GND
14       H SYNC GND
15       H SYNC

If you want to build a video cable to connect to a Mac-standard DB-15 monitor, then just wire those pins straight to the DB-15, matching them up pin for pin. Pivot card pin 1 connects to DB-15 pin 1, Pivot pin 2 connects to DB-15 pin 2, etc.

If you want to build a combo video cable and VGA adapter with a hard-coded 832 x 624 sense code (which seems to be the highest possible Pivot resolution that's also supported by modern LCD monitors), here's the pin mapping:

PIVOT   VGA

1,6,13  4,5,6,7,8  VIDEO GND
2       1          RED VIDEO
4,10    NC         PINS 4 & 10 JUMPERED TOGETHER ON PIVOT END OF CABLE
5       2          GREEN VIDEO
9       3          BLUE VIDEO
11     11          C & V SYNC GND
12     14          V SYNC
14     10          H SYNC GND
15     13          H SYNC

 

[Trash80toHP_Mini]

Dynamite! Way to go bigmess! :approve:

You'll not want to use the onboard video, if the IIsi senses that no monitor is hooked up to its DA-15, all of Bank A's lifeblood stays in place, none of it gets sucked into the Vampire Video buffering setup. Better performance, more Pixels a/o higher bit levels.

Now I don't have to pull the Rocket out of the IIsi or haul out all the Mac Monitors! [] ]'>