Can WB3.0 autodetect and truncate for 7-bit printers?

[Hodgkinson]

Hi folkes,
At the moment im in the process of adapting a old *cough splutter* printer for use with my A1200 on the printer port.
An important point about the printer is that it supports only 7 bits, thus, it can only work with ASCII characters from 0 to 127. There is physically no 8th data bit input line.
I've noticed that when I send characters in the extended region of the ASCII table to the printer, that the output of the printer bears more than a passing resemblance to the characters that were sent to print; rather than printing their truncated character from the first half of the ASCII table. Furthermore, I can see no activity of the 8th data bit line.


For instance (The characters here are only guesses as to what was on the screen at the time):
庩®þ
Prints
a*crt
Rather than
(Invalid)x8)g

Is the system (WB3.0, generic printer using the standard system printing routine, printed from CygnusEd in the standard default font) somehow detecting that the printer is only seven bit and automaticially substituting in characters similar in appearance?

Hope someone can help me with this mystery,
Hodgkinson.

[RW222]

I can see how it would be set up like that for a daisywheel printer, or a laser emulating daisywheel, but if it's a dot matrix it seems a little odd, unless that also has a daisywheel emulation mode.

[Hodgkinson]

I can see how it would be set up like that for a daisywheel printer, or a laser emulating daisywheel, but if it's a dot matrix it seems a little odd, unless that also has a daisywheel emulation mode.

Heh, its none of the above. It’s actually a 20-Coloum thermal printer. Oh, and it pre-dates the A1000.

Like I say, the system seems to of detected that this printer; without any special driver; is 7-bit only and the system is automatically modifying each character before its' sent to the printer. I went through all of the different characters on the keyboard last night and sent them to print - Every single one was modified before it got to the printer.
Its got to be the system thats doing this since there is no 8th data line to the printer.

I wouldn't mind knowing if there is a way to fool the system into thinking that it is an 8-bit printer, so that I can actually test the circuit that I spent a week building to detect the 8th data bits. I'd just like to know whether the circuit *would* of worked.

Hodgkinson.

[Hodgkinson]

:bump:

[Zac67]

It all depends on the printer driver you've selected. The generic line printer style one will do exactly that kind of conversion (because ASCII is only defined for 7 bit data). Most printer command sets (ESC/P, PCL, ...) use escape sequences to switch the character set and only use the full data width when printing dot graphics.

I don't think it's technically possible to detect a missing data bit on a Centronics i/f.

[Hodgkinson]

Hmmm. That sounds interesting - So ASCII doesn't normally include the "Extended" part of the character range from 128 to 255?

I don't think it's technically possible to detect a missing data bit on a Centronics i/f.

Extra (8th) bit :-D
I’m just looking for a high logic level present when /STROBE is active low - Just a few NAND gates are used, one wired as an inverter, one used as a NAND, and the other two as a NAND latch.

Thanks for the info,
Hodgkinson.