Compact Mac Video Timings Needed for SuperVideo Input Parameters

[Trash80toHP_Mini]

Trying to get Compact Mac Video up and running on a Multisync CRT TPD.  Really! :approve:

Compact Mac Video Timings

Horizontal Frequency__________?

Horizontal period______________?

Horizontal back porch_________?

Horizontal front porch_________?
Horizontal sync_________________?

Horizontal active_______________?

Horizontal blanking____________?

Pixels visible____________________512

Pixel clock_______________________?

Vertical frequency______________?

Vertical back porch____________?

Vertical front porch____________?

Vertical Sync____________________?

Vertical active___________________342 lines

Total lines_______________________?

@techknight lil' help here? I got two right off! []

 

[Trash80toHP_Mini]

Here's the screenshot from waybackwhen.

View attachment 16458

@Gorgonops finally got it through my thick skull that this card can't output more pixels than it already is, but just maybe I can coax a lot less pixels out of it? If it will output 512x342 to my TPD and the CRT can handle the freqs, it might be interesting to see what else this card might drive? Lots of calculations to do to get the right numbers into those eight inputs there. :-/

 

[Trash80toHP_Mini]

Here's what video timings look like over on tinyvga:



I was able to harvest a bit of data from the RCPII/IIsi Grayscale in SE/30 discussion:

Compact Mac Video Timings

Horizontal Frequency__________ ?

Horizontal period______________?

Horizontal back porch_________?

Horizontal front porch_________?
Horizontal sync_________________ 22.254 kHz

Horizontal active______________  15.37 ms

Horizontal blanking____________?

Pixels visible____________________  512

Pixel clock_______________________?

Vertical frequency______________?

Vertical back porch____________?

Vertical front porch____________?

Vertical Sync____________________  60 Hz

Vertical active___________________ 342 lines

Total lines_______________________?

Dot Clock of SE/30 is the same as that of the 12" RGB  .  .  .   whatever that is?

Might be a third of the way there, 5 specs down, 10 more to go. Lil' help, gang?

edit: gotta scan the manual section on setting up the card's output for a custom monitor. Maybe that will spark some interest?

 

[Bolle]

What is that tool in the screenshot that shows the oscillator frequency and all those values?

Is this part of the SuperVideo panel? Never seen it before.

 

[Trash80toHP_Mini]

That's probably because it's specific to only this rev of the Spectrum/24. The card can be modified to run customized resolutions by changing the crystal on board to the value calculated when those inputs are specified.

edit: oopsie, yes, it's part of the SuperVideo Control Panel.

 

[Trash80toHP_Mini]

Manual and card are from 1989, so I think this is the first rev of the Spectrum/24. It shipped with a 64 MHz crystal for its max res of 1024x768 @ 24bit to drive SuperMac's 16" and 19" monitors. It also shipped with a 30.24 MHz crystal to drive Apple's 12" Monochrome and 13" AppleColor monitors.

Given the proper input specs for Compact Mac 9" resolution it should give us a lower frequency spec. Hopefully I can get it to sync up on one of my TPDs. If not it'll be a real shot in the dark when hooking it up to A/B and GS neck board in a Compact. Dunno if it will lead to a new internal GS Card for the SE/30, but stranger things have happened. If nothing else it'd be sweet to have a Hobbit lookin' little 9" Grayscale monitor sitting atop the pet IIfx. There's really not room for a third 21" CRT on that workstation. []

 

[Trash80toHP_Mini]

@techknight had posted this earlier in the thread, gotta find his GS Card development thread for more info.

The SE/30 is hampered by that darned video hardware, as its clocked off the main system bus and it doesn't use its own clock.

Once you get it off the dependence of that on-board video, than anythings possible. The Simasimac we are all known and accustomed to is generated from the video RAM at its initial power-up state, before the VRAM is ever initialized by the mac itself.

I did notice the simasimac is different between the 2 different RAMs used.

uniserver's response of five years past:

1 - this IIsi radius video <Color Pivot II/IIsi> card works fine with stock SE/30 ROM.

2 - pulling the SE/30 onboard video rom does not give you any worth-while results… the computer still boots using the internal CRT anyways, even if the rom is not installed.

so the internal video would have to be disabled in other ways

This tidbit appears to affect two concurrent projects, this GS Card Windmill and proposed testing testing of the WireWrap1 LC PDS NIC prototype adapter. When the IIsi boots without detecting sense line encoding, buffering of the first Megabyte of system memory as blood sucked VRAM does not take place. I imagine the IIsi might be run headless? At any rate, a PDS card can then become the startup screen in a clean break from Slot $E operations as opposed the the bastardized setup of the SE/30 hybrid?*****

The SE/30 appears to be hitting on internal video to display the  :huh: icon at startup before INITs load and even before the Video ROM is polled? That makes sense regarding its odd designation. From my read, it's not a Declaration ROM per se in terms of the Slot Manager, being either a subset (superset?) of DeclROM function?

Available GS Card is an either/or proposition in the SE/30 for internal/external display support. WAG: Control Panel component of the GS Card driver would set state of INIT component of the driver, disabling normal ops of the SE/30 startup sequence to divert process of writing to VRAM for the startup screen to writing to GS Card for same? Is there a blip in video output for startup screen or any other obvious indication that this would be the case?

Logic Analyzer time again? Looks to me like the GS Card's Neck Board would be the A$$ end of this hack with its hacked INIT on the other? 

< Tangent >

*****This all makes some sense to me as the SE/30 sits at the fork in the road of Compact Mac development: The SE/30 can be seen as a Color QuickDraw/NuBus architecture PDS hybrid from IIcs Sire and SE QuickdDaw PDS dam. The Classic, which is assumed to be the replacement the SE/30 in hardware comparisons, can then be seen as its purebred sibling, being a direct developed of the Compact Mac line that lived on for a bit in the Classic II. The Color Classic in its bloated Compact form factor would be the SE/30's badly malformed offspring.

< /Tangent >

 

[BadGoldEagle]

Just stumbled across this thread. Can this be of any use?

https://nerdhut.de/2016/06/26/macintosh-classic-crt-1/

 

[Gorgonops]

From a link in the last link:

(Unfortunately the corruption that wipes out a few of the values appears to be in the original PDF for the Classic II Developer Note, at least the first version I stumbled across. But you can actually copy-paste the text out of that area if you have the PDF and see some of what's cut out.. )

 

[Trash80toHP_Mini]

This is what I got doing that:

Horizontal Ti

Active vid

Horizontal sync = 2

Vide
/HBLAN
/HSYN
192 do

512 do
704 do
Horizontal sync = 2
Vide
28 lin 342 lin
370 lin
Vertical sync = 4 lines
Back porch = 24 lines
/VBLAN
/VSYNC
Front porch = 0 lines
Dot clock = 15.6672 MHz
Dot = 63.83 ns
Full line = 49.93 ms
Line rate = 22.25 kHz
Full frame = 16.626 ms
Frame rate = 60.15 Hz
Black
White
Black
White
Active vid
178 do 110 do
Front porch = 14
Active vid

Seems like they're specifying time intervals as multiples of the 6.383 ns Dot? How helpful is that? :

Came up with bopkes in the DevNote for the Classic after doing some serious minus work in the Boolean search string.  Took a wild guess it'd be: mac_classic.pdf

Finally made some headway. Looked over to the bookcase at arms length and pulled Technical Introduction to the Macintosh Family off the shelf. Du'Oh!



Too tired to set up the scanner so I leaned one side of the book up against a sneaker and piled wrenches on it to hold the wood pulp based storage media flat-ish.

Macintosh Family Hardware Reference has been sitting next to it this whole time. It's out for a morning skim if the eyeballs start tracking together again. Sheesh! [:I]

 

[Trash80toHP_Mini]

LC_II DevNote nomenclature: fiddled around with the numbers for a horizontal scan line  as specified in dot intervals I got 511.90098 for the active area. That's probably cumulative roundoff error?

Recall that a program may use a much larger area of memory for graphic operations, using QuickDraw to create off-screen imagesn images so that theycan be displayed more rapidly.

LC_II DevNote nomenclature: fiddled around with the numbers for a horizontal scan line  as specified in dot intervals I got 511.90098 for the active area. That's probably cumulative roundoff error?

 

[Trash80toHP_Mini]

Ran out the edit clock, definitely asleep here:

Recall that a program may use a much larger area of memory for graphic operations, using QuickDraw to create off-screen imagesn images so that theycan be displayed more rapidly.

In a fatigue confuzzled state, that tidbit had me thinking we gave up on using the RCPII/IIsi too soon five years ago. Delusion was that extra lines of grayscale pixels could be handled as a panable off-screen image. Mixed up a Quickdraw operation in software with timing limitations of the Compact hardware.

Still interesting: does the Xceed GS card do panable 640x480(?) on the internal side? Got manual linkage?

LC_II DevNote nomenclature: fiddled around with the numbers for a horizontal scan line as specified in dot intervals getting 511.90098 after subtracting 192 do (do = 63.83 ns) for the active area. That's probably cumulative roundoff error? Makes me wonder about number of significant digits required for calculations? TinyVGA appears to use 15?

Good news: the LC DevNote expression of Vertical Sync in lines might be plugged directly into the SuperVideo spreadsheed formulas/Inputs?

Compact Mac Video Timings  -  as expressed in SuperVideo oscillator calculation formula inputs and maybe its Control Panel inputs?

Horizontal Frequency__________ ?

Horizontal period______________?

Horizontal back porch_________?

Horizontal front porch_________?
Horizontal sync_________________ 22.254 kHz

Horizontal active______________  15.37 ms

Horizontal blanking____________?

Pixels visible____________________  512

Pixel clock_______________________?

Vertical frequency______________  60.1 Hz  - 

Vertical back porch____________  24 lines

Vertical front porch____________  0 lines

Vertical Sync____________________  4 lines

Vertical active___________________ 342 lines

Total lines_______________________?

Eight of fifteen input calculation variables so far, not too bad  .  .  .  if an accurate assessment? Need more sleep and then coffee, but if the Classic II Devnote's "do"  totals =  (VIDCLK periods) I'm farther along here than I've been thinking?

Compact Mac Video Timings  -  as expressed in SuperVideo Control Panel inputs:

Horizontal End Sync (VIDCLK periods) ___________ ?

Horizontal End Blank (VIDCLK periods) __________ ?

Horizontal Start Balnk (VIDCLK periods) _________ ?

Horizontal Total (VIDCLK periods) ________________ ?

Vertical End Sync (Lines) ___________________________ ?

Vertical End Blank (Lines) __________________________ ?

Vertical Start Blank (Lines) _________________________ ?

Vertical Total (Lines) _______________________________  ?



Am I on the right track or is there an error in my logic in equating the different terms/measurements? Wondering SuperVideo inputs relate to Front porch/Back porch =

 

[Gorgonops]

I'm kind of confused as to why you're looking in the LC II devnote for anything related to this if your goal is to set up a Compact Mac-compatible display mode? The Compact Mac screen does *not* run at the same scan rates as the 12" 512x384 monitor.

If you're serious about trying to pull this off, IE, converting a B&W Mac monitor to run off of a more conventional video card, I have a suggestion for you: put this SuperMac video card out of your head for a while because it's actually a pretty lousy thing to be starting with. Instead of having a programmable timing generator like you'll find on more modern cards it requires swapping oscillators to implement new video modes, and of course is tool to set up said modes isn't by any means a standard thing so you're going to have a lot of trouble finding anyone able to help you translate specs into its lingo.
 

I'd suggest starting with a rotgut PC with a simple VGA card (almost anything will do, ATI cards have always been a good choice from a compatibility perspective), slap Linux or FreeBSD on it, and once you've determined how you're electrically going to interface the VGA card to your compact Mac monitor (an exercise for the reader, remember that in addition to either needing to swap the neck board out or radically hacking the analog board if you want to accept an analog grayscale signal you'll have to add some circuitry to make it accept standard sync signals, that has been discussed in a few other threads) you can try using a standard Modeline calculator to come up with a video mode that works.

Based on the values in the "important numbers" box on this page I used this modeline calculator to come up with the following Xfree86 modeline:

Modeline "512x342"   15.6672   512 526 529 704   342 342 346 370  -hsync

Here's what I input to come up with that; some of the values arrived at by trial-and-error:



And here's what the calculator says they add up to:
 



Which looks like a pretty good fit for what we know.

Once you have this working with something like a standard video card then perhaps you can figure out how to translate a modeline into what your SuperMac's control panel wants and also know what crystal you need to buy.

 

[Trash80toHP_Mini]

I'm kind of confused as to why you're looking in the LC II devnote for anything related to this if your goal is to set up a Compact Mac-compatible display mode? The Compact Mac screen does *not* run at the same scan rates as the 12" 512x384 monitor. 

Confuzzlement there, I meant Classic II Devnote. This is the first time I've approached a wakeful state and skirted the realm of rationality over my last three posts. Habit of referencing the LCII DevNote in the NIC adaptation thread is somewhat problematic. Got most of my next post in the can, coming up soon.

Thanks for the great work, I'll try to parse your last after another mug of java. Trying to decid what's worse, chores or Scanning a half dozen pages from a pair of Apple Tomes and something lik 15 from the Spectrum/24 User's Manual, and reformatting to post here or some of my cleaning and organization chores? CarrotSTICK on both ends of that one.

 

[Trash80toHP_Mini]

Morning reading of Macintosh Family Hardware Reference was fruitful, bleary, but it's definitely a new, expanded, 4.5 page take on Classic Mac Video. Gotta scan it, the only link to a downloadable copy found so far was on sutraplay.com. The signup for download access looks suspiciously like a credit card phishing expedition. If the place is legit, I'd love to hear of it. Site said the PDF had been downloaded in the last 58 minutes, anybody here do that? It's not available on vintageapple.org

Anyway, rough overviews outline how buffering of video/main memory takes place in the PALs and reference "Classical Macintosh VIA" info in Chapter 2. Flicker Free screen redraw techniques by using the screen buffers independently are outlined and breaks off with a reference to using Segment Loader for this in https://vintageapple.org/macbooks/. break off That tidbit breaks in a reference to  Never realized VIA was an off the shelf Rockwell part. Always figured they were Apple ACICs?

Most interesting tidbit was dead-on last:

__________________________________________________________________________________________________________
Important
The Vertical Blanking interrupt is guaranteed to be coordinated with the screen's vertical blanking interval only in Macintosh computers that use a built-in video display. Computers with separate video cards-such as a Macintosh II or a Macintosh SE with a video card in theSE bus slot-may have a video display that is independent of the Vertical Blanking interrupt.
__________________________________________________________________________________________________________

Dunno, bleary confuzzlement of the AM would be that we've got PALs and VIA for a Vertical Blanking interrupt wedge. Mechanism might be found in QuickDraw Segment Loader routines in Volume II of  Inside Macintosh?

WAG: does the Xceed driver disable both? "Two areas of memory are reserved for use by the video circuitry: the main screen buffer and alternate screen buffer. A bit in VIA data register A determines which screen buffer is read by video circuitry."

Dunno where the wedge might be, but we're not worrying about smoking my SE/30. After (if and when ) I have a spare magic smoke powered Plus set up as a cute little 512x314 GS display atop the pet IIfx next to the 21" CRT testbeds in a lopsided three screen desktop setup, then SE/30 hijinks could begin. I'll look over your PC/rotgut VGA card/Linux suggestion if my brain comes online at any time on this day off from work. First response is that the numbers seem to be falling into place rather nicely for the calculation formulas in the Spectrum/24 manual I've not scanned as yet. Maybe today?

Upside to your suggestion is that the Intel D525MX is all set up, ready to rock with an ubuntu install IIRC. It's at the base of the tower of towers in between AppleDisplay half rack and the pet IIfx workstation and within cable reach of its monster monitors. Ten year old ATOM board has a single PCI slot, so linkage to an appropriate rotgut VGA card on eBay would be appreciated should we pursue that route. Caveat: I'll be totally at your mercy for VGA/Linux antics. Is a remote desktop setup possible or desirable from your end? Hardware I can handle/mangle. Wanna risk your free from me (via uni) 512k on this in your lab?

Ya gotta creep before you can crawl. A baby step would aiming too high for this initial project thread.

 

[Trash80toHP_Mini]

At this point developing the Compact Mac Video Specs required in the IP appears to be developing a Rosetta Stone translation table between the various representations of the documented timings to verify the numbers? There may be little need of employing the four pages of formulas to come up with the necessary Plug-ins:

 1. Decide how many horizontal and vertical pixels the monitor should display - GIVEN

 2. Find the horizontal period, and the horizontal blanking period. - SPECS

 3. Determine the required pixel rate and oscillator frequency. - SPECS and Oscillator TBT by SuperVideo Control Panel

 4. Determine the pixel rate divider value. - ???

 5. Determine the time required to display one horizontal unit of pixels. - SPECS

 6. Determine the value for the Horizontal End Sync Field. - SPECS

 7. Determine the value for the Horizontal End Blank Field. - SPECS

 8. Determine the value for the Horizontal Start Blank Field. - SPECS

 9. Determine the value for the Horizontal Total Field. - SPECS

10. Determine the value for the Vertical End Sync Field. - SPECS

11. Determine the value for the Vertical End Blank Field. - SPECS

12. Determine the value for the Vertical Start Blank Field. - SPECS

13. Determine the value for the Vertical Total Field. - SPECS

14. Set "Interlace" Check Box if applicable -  NA

Some of my "SPECS" notations might not be valid, still confuzzled here and remain in caffeine deprivation mode.

Close to a done deal by my read, so we're bound to get an unsupported frequency error error, not an oscillator spec. Then its time to take it back to learning to roll over as creeping progress would be nullified.

 

[Trash80toHP_Mini]

Breakdown of best specs unearthed to date: from Macintosh Family Hardware Reference expressed as equation results for formulas in the SuperVideo manual.

 1. Decide how many horizontal pixels and vertical pixels the monitor should display 512 x 342

 2. Find the horizontal period - 512 pixels = 32.68 µs. (MFHR below)

     horizontal blanking period. 192 pixels, or 12.25 µs. (MFHR below)

 3. Determine the required pixel rate - 15.6672 MHz or about 0.064 µs per pixel. (MFHR)

     The pixel clock rate (the frequency at which pixels are displayed) is 15.6772 MHz or about 0.064 µs per pixel.

     For each scan line, 512 pixels are drawn on the screen, requiring 32.68 µs.

     The horizontal blanking interval takes the time of an additional 192 pixels, or 12.25 µs.

     Thus, each full scan line takes 44.93 µs, which means the horizontal scan rate is 22.24 kHz.

     Determine the oscillator frequency. TBD by SuperVideo Control Pane?

 4. Determine the pixel rate divider value. - ?

 5. Determine the time required to display one horizontal unit of pixels. 15367.65 µs (MFHR above)

     A full screen display consists of 243 visible horizontal scan lines, occupying 15367.65 µs.

 6. Determine the value for the Horizontal End Sync Field. - ?

 7. Determine the value for the Horizontal End Blank Field. - 28 scan lines = 1258.17 µs (MFHR)

 8. Determine the value for the Horizontal Start Blank Field. - SPECS imply 0 value for front porch

 9. Determine the value for the Horizontal Total Field. - ?

10. Determine the value for the Vertical End Sync Field. - ?

11. Determine the value for the Vertical End Blank Field. - 28 lines = 1258.17 µs.

     During the vertical blanking interval, the turned-off beam-while moving from the bottom of the screen to the top-

     invisibly traces an additional 28 scan lines, taking 1258.17 µs. or about1.26 ms to do so. This means the full frame is re-displayed

     every 370 scan lines, or once every 16625.8 µs. That's about 16.6 ms per frame, which means

     the vertical scan rate (the full screen display frequency) is 60.15 Hz.

12. Determine the value for the Vertical Start Blank Field. - SPECS imply 0 value for front porch

13. Determine the value for the Vertical Total Field. - 370 lines

14. Set "Interlace" Check Box if applicable -  NA

That's what I could wring out of the page in one sitting. Need to proof it. I've struck thru any approximations expressed in microseconds from hard numbers expressed in milliseconds. ZERO room for or tolerance foof such slop in the numbers for this project.

Summary: Certainly missed one data point and there are three more to noodle out, but it looks like we've got hard, accurate data to several significant digits from Apple Docs for at least 10 of what appear to be fifteen of the calculations outlined in the Spectrum/24 manual for the process of setting up a custom monitor. "Interlace" is definitely NA

Compact Mac Video Timings  -  as expressed in SuperVideo Control Panel inputs:

View attachment 27847

Horizontal End Sync (VIDCLK periods) ___________   ?

Horizontal End Blank (VIDCLK periods) __________   ?

Horizontal Start Blank (VIDCLK periods) _________   Listed as 14 do(?) in Classic_II DevNote? VIDCLK Period = ???  Horizontal Start Blank = Front Porch?

Horizontal Total (VIDCLK periods) ________________   ?

Vertical End Sync (Lines) ___________________________   ?

Vertical End Blank (Lines) __________________________   28

Vertical Start Blank (Lines) _________________________     0 of anything? that's a hard number from Classic_II DevNote - IF - Vertical Start Blank = Front Porch?

Vertical Total (Lines) _______________________________  370

Nagging questions about nomenclature and units of measure:

VIDCLK Period = ?????

Horizontal End Sync?

Vertical End Sync?

Numbers in the 640x480 value boxes above don't make a lot of sense to me yet.

edit: corrected/annotated withdata from Classic II Devnote:

 

[Trash80toHP_Mini]

Ran the edit clock out again!

From that diagram, it now looks to me like the H&V Sync numbers amount to whatever pixels remain after active area pixels and porch pixels are subtracted from total horizontal "do" on that axis and total lines on the vertical axis?

Does this make sense to anyone out there? @Gorgonops The values I'm getting from Macintosh Family Hardware Reference are looking a lot like the numbers you posted above. Which numbers did you come up with by trial and error and are they any different from mine?

 

[Trash80toHP_Mini]

LOL! Now none of this makes sense to me! But I've learned enough about the comings and goings of the video process to start working through the series of calculations necessary to come up with the proper plug-ins for the control panel  .  .  .  maybe. :blink: