Latest acquisition: //c plus THREE IIgs and more!

[Anonymous Freak]

Ah, what $60 will get you...

I just picked up a //c with power supply, plus matching 5.25" external floppy drive and green-mono CRT. Includes original manual and two set-up guides. Sadly, the //c doesn't power on. The power switch feels like the green part you push has disconnected from the actual switch inside. Turns out the power supply is dead. This one works (although you have to jiggle the power button to make is stay in the 'on' position,) but the PSU doesn't work on another //c. A copy of MousePaint was in the drive.

Along with that are three IIgs systems.

One is a ROM 1, stock.

One is a ROM 1 with memory expansion (AppleWorks says 1149 K available.) The memory card has all of its sockets full.

The third is a ROM 3 with memory expansion (AppleWorks says 1277 K available.) The memory card has all of its sockets empty.

I swapped the memory expansion cards, to give the ROM 3 machine 2 MB, and the ROM 1 machine 384 K.

Also included are a boxed copy (new-style packaging with the black bar) of AppleWorks, along with a second copy on old-style (colorful writing on beige) disks.

An ADB keyboard (the small style that came with the later 68k and early PPC machines,) and mouse (the first ADB mouse,) were given as well.

Two 800K floppy drives were thrown in.

Finally, two 3.5" disks were in the two 800K drives.....

One is rather boring, an Apple IIgs copy of Math Rabbit. (Maybe I'll have my five-year-old daughter play it...)

But, the bonus disk is a "Macintosh System Disk" (complete with colorful 'Picasso Mac' label) containing a pristine copy of System 2.0/Finder 4.1! It is now my oldest official-from-Apple Macintosh system disk. And it works like a charm on my SE. (Can't wait to get a 128K to use it on.) Yes, I promptly made a disk image of it.

The dead //c, along with the two ROM 1 IIgs systems, are going up for sale, look for the thread in the Marketplace. (Along with two more //c's I just found in storage that both work perfectly.)

 

[MrMacPlus]

Nice score, I got a ][gs back in june but haven't been able to use it (TVs make terrible composite monitors). The one time I did use it, the power supply was making noises.

 

[Anonymous Freak]

Update: My daughter LOVES Math Rabbit. I think I'll have to find some other vintage age-appropriate games for her. Maybe I'll get her started on LOGO Turtle, and get her programming by age 8.

 

[Nathan]

Nice score, I got a ][gs back in june but haven't been able to use it (TVs make terrible composite monitors). The one time I did use it, the power supply was making noises.

Would it be fair to say that the TV is not at fault but rather the messy output of the computer? Maybe using an RF converter would produce better results.

 

[Paralel]

I remember someone told me that modern comb filters on the last generation or two of tube TV's was to blame for the useless image that is generated by older computer systems that output to composite. I have absolutely no idea whether this is true or not.

 

[Nathan]

Number 018 http://www.faqs.org/faqs/apple2/faq/part16/

From: Michael Pender and Rubywand
018- What is the usual way to connect an Apple II to a display?

For a IIgs, you connect to the IIgs RGB monitor. For other Apple II's, you use a video quality 'RCA cable' to connect to a composite color monitor, such as one of the Apple II monitors or a third party monitor like the Amdek Color-1. (The IIgs has a composite color output which can go to a composite monitor; but, for IIgs super-res, the results are unsatisfactory.)

An alternative to a composite monitor is to connect to a color TV directly or through a VCR. Most VCRs and many modern televisions have a composite video input port. Rather than buying a new monitor, a person can usually use an RCA cable to connect the Apple to a spare VCR or television. The stabilization logic built into even a cheap VCR can go a long way toward cleaning up an ugly picture.

Read more: http://www.faqs.org/faqs/apple2/faq/part16/#ixzz0mEvFBnO0

The fact that they suggest passing it through a VCR to use its "stablilization logic" suggests that the Apple II's composite video output is less than perfect for input to modern TVs. It is possible that the video circuit was designed around/optimized for the composite monitors they sold. I don't know enough to say, but I would not be surprised if what you say might be true about the filters is true. I think it might also be true that there is RF interference between your computer and the TV. Any chance you could describe what the problem is like or link to some pictures?

 

[H3NRY]

There are a couple of things that can make a TV mess up on an Apple II or II Plus. First, composite NTSC video is 525i, 30 frames of 525 lines each second interlaced in 2 fields of 262.5 lines. An Apple II generates 262p at 60 frames per second, which is pretty close to NTSC, but each half-frame is the same, and there's a half-line missing. Analog sync circuits mostly lock to this without a problem. Later digital sync circuits get messed up by the missing half-line.

Color TV sets have filters designed to roll off video response before 3.58 MHz, which is the color subcarrier frequency. This is pretty low video bandwidth, corresponding to about 200 lines horizontal resolution. Some sets use comb filters which allow video bandwidth up to 4.2 MHz, but which again don't like the un-interlaced Apple signal. They count on the color burst phase changing 180 degrees every line, and the Apple II doesn't.

So a Sony or other top line TV from 1977 will look pretty good. A 1990s TV with digital counters for sync won't lock very happily. A composite monitor like one of Apple's or Commodore's is next best to RGB, and RGB adapters for the Apple II are rare. LCD TVs are digital and most won't lock to an Apple II. Even Apple's own blue & white Studio LCD, which has a composite input, will not display Apple II text, although it will display graphics.

The IIgs with a gen-lock card is a different matter. I understand it is capable of true NTSC output.

 

[Nathan]

Is it possible to easily modify or add circuitry to make some kind of digital sync signal for the monitor? Any chance an IC analog to digital (ADC) converter might suffice?

 

[Osgeld]

oh I have been thinking about it sorta half hart, I am using a tv card in my computer, and it makes 80 col text impossible to read, growing up with the apple monitor II green screen I have never seen anything that quite compares to that heavy shadow mask and tweaked timing, even running it though a vcr (actually the pc tv tuner card is quite possibly the best i have seen an apple on a "tv")

the problem I see is that to adjust any analog video timing you pretty much have to capture it and then process it then output it, I dunno ill toss it on the o-scope tonight and see how different it really is, and with hard numbers in hand it would be easier to answer that question

no I am not a tv engineer, but I have been designing my own rs170 (B&W ntsc) system to use with microcontrollers and am saturated with video timing data (heck even made a vga sync signal out of a 556 timer)

http://www.cheesefactory.us/filecenter/vga/PDR_0192.JPG

 

[Osgeld]

ps i am interested in that "dead" //c also

 

[Osgeld]

ok few things

262.5 lines

most of the text I casually looked up suggested 262 lines, with the time of the "extra half line" being the gun sweeping back to the top time or never mention the half line so some clarification on that would be appreciated (as I have a few working rs170 examples using 262 and they behave fine on new sets)

Color TV sets have filters designed to roll off video response before 3.58 MHz, which is the color subcarrier frequency. This is pretty low video bandwidth, corresponding to about 200 lines horizontal resolution. Some sets use comb filters which allow video bandwidth up to 4.2 MHz, but which again don't like the un-interlaced Apple signal. They count on the color burst phase changing 180 degrees every line, and the Apple II doesn't.

as I have not even looked at colour burst yet so I will take your word, but it also makes me wonder how other old computers and video game systems do the same trick and still function on new screens

ok my findings on my apple //c doing a line scan, is that everything about the video output of my computer is short in time

vertical sync lasts only 3.0us (4.7 is the norm)

the visible video signal is 48 us vs 51.5

and the whole line takes 52us vs 63.5

I wasnt able to grab a good back porch and front porch time, but it doesn't matter, its short by roughly 11us, which is pretty good considering the age of the design BUT as tv's become more (hd super lux 3d truvision virtualreality media centers) anal retentive I would suspect this timing discrepancy to be the major culprit

easy solution? not off hand other than using a pretty powerful mcu (or whatever your into) with a decent chunk of ram to sample the image per frame and re-output it to whatever format you want, and someone has already done this for the II and the IIc (not sure about GS) with a fpga, but I dont think they are currently in production

 

[Nathan]

Making an add-in card to replace the onboard video out and having it read the video portion of the ram makes more sense than trying to capture the output frame by frame. Perhaps it could be done by combining this idea on the web: http://neil.franklin.ch/Projects/SoftVGA/ with a way to read the Apple IIs video ram.

 

[Osgeld]

if you want to spider-web your apple with 30 gauge jumpers maybe you could only do low rez then yes, but your still sampling the video data and then post processing it to output, so you need memory or the screen will flicker

I am using a ATmega 328, which is the same as he is using, at a high res image monochrome, 1 bit per pixel your looking at about 8k of ram at 256x256, the apple does 290x192 in colour, either way the mcu has 2k of ram, adding extra ram slows down the mcu for communications, and its already using assembly language to barley put out vga sync signals and whatever is left is used for 40 col graphics in colour if your slick on code

so you need more cpu and more ram and it snowballs

you could do it line by line, but its the same deal but you need a ton more speed, and stuff that has more speed usually has enough ram to store a sample, and at those speeds its just easier to decode the analog signal and resample it, plug n play style

its not expensive or anything, it just a large project for a small audience

 

[Nathan]

Can the expansion cards access all of the normal system memory? If so, you could certainly access the video ram directly. Otherwise, why not just use an sram chip with another avr microcontroller to copy video memory into and just have the ATmega 328 read the sram for video data?

 

[Dog Cow]

Can the expansion cards access all of the normal system memory?

Yes.

 

[Osgeld]

why not just use an sram chip with another avr microcontroller to copy video memory into and just have the ATmega 328 read the sram for video data?

I guess you could, all I can say is have fun, and I would not use a atmega (or similar) , as stated before your going to be doing some seriously finicky coding as just driving a sync signal out of something like that (even with rs170), I would look at at least a 16 bit mcu

 

[Nathan]

Any reason you couldn't just use it over the serial port as though your modern desktop were a terminal connected by serial to the machine? Then you could have whatever font you want. It's not like there's anything special it couldn't transmit except maybe graphical characters, but that should be possible with the right terminal program and maybe a copy of those graphical characters.

 

[H3NRY]

Any reason you couldn't just use it over the serial port as though your modern desktop were a terminal connected by serial to the machine?

If all you're looking at is text, hook up a serial line, run a terminal program on your other system, type PR#[serial card slot], and voila! Graphics, no way.

There were various 3rd party video cards and Apple's genlock cards which put out honest-to-gosh NTSC properly timed broadcast-able video. Otherwise, it was usual to aim a TV camera at the Apple's monitor. There are a few TBCs which will accept Apple's funny video and correct it.

80 column text requires about 7 or 8 MHz video response, and that's beyond NTSC color composite compatibility, though RS170 monochrome video is compatible. For 80 columns, you need a monochrome or an RGB monitor.

@Osgeld, think about interlace. One field starts at the upper left margin of the screen and ends at the lower right. The next field starts in the top middle of the screen and ends in the bottom middle. So there is a half-line of nothing at the start and end of the second field in each frame. So you could say 525 lines is made up of a 262 line field + a 263 line field, or two 262.5 line fields. Or whatever. Yes, the half-line time is also vertical retrace time. No wonder video engineers called it "Never Twice Same Color" and other less printable names. Apple video is shorter than standard (longer front & back porches) because it is designed to be in the visible part of a TV screen, whereas broadcast video is normally overscanned (extends past the visible area on both sides).

Other systems like Atari and Commodore used custom 4-bit DAC / timing chips to generate video, which produced a more standard video signal, but still usually not broadcast-able, because they also used 262p video to avoid flicker from interlace. Don't know if that makes the subject any clearer. . . :-/

 

[Osgeld]

there is some graphical characters on a apple II but most of the graphics is real graphics, maybe there could be a driver to output graphics to serial made, and for a ton of programs that outline and flood fill a bitmap (trolls tale, Carmen San Diego, etc) it might be tolerable, but for others like choplifter or night crawler or probally gsos, who are doing high speed stuff it would not work

text applications would function