Did anyone here ever try building the Morse Code decoder described in the October 1976 issue of BYTE magazine, by Robert Grappel and Jack Hemenway, entitled "Add This 6800 MORSER to Your Amateur Radio Station"?
By July of 1977 I had built and tested the tone detector depicted in the article, and I had copied the source code onto the Georgia Tech main computer. The only changes I made were to the address of the character output subroutine, and to the address of the PIA port, as advised by the article. But when I tried it all out, it didn't work, and I couldn't figure out why. I abandoned the effort, but I kept my cross-assembler output listing and the MIKBUG-format object code listing. I used the tone detector circuit for another project and it worked fine. Now, almost 40 years later, I have revisited that BYTE magazine article.
The authors stated on page 31; "The user of MORSER must provide a suitable [character output] routine for his or her own system. For example, the OUTEEE routine in Motorola's MIKBUG will work".
Then on page 35 of the article they stated; "The system subroutine used to drive the [output] terminal is called in this section on line 270. It should not use interrupts or any timing loops, since these will upset the timing in MORSER". In 1977 I did not understand that this statement contradicted the earlier statement regarding the OUTEEE subroutine.
The MIKBUG subroutine OUTEEE calls a software/hardware timing loop subroutine while serially transmitting the ASCII character from one data line of a parallel port (Motorola MC6820) to the RS-232C driver circuits. The return from subroutine does not occur until the entire character has been serially transmitted, which takes 33.3 milliseconds at 300 baud rate. So OUTEEE does not work for this application.
CONCLUSION: The MORSER character output interface must use a UART circuit, such as the General Instruments AY-5-1013A, or Motorola's MC6850 ACIA chip.
For example, MORSER would not work on an SWTPC 6800 with MIKBUG, but should work if the SWTPC has SWTBUG and uses an MP-S card for the console terminal.
-73
By July of 1977 I had built and tested the tone detector depicted in the article, and I had copied the source code onto the Georgia Tech main computer. The only changes I made were to the address of the character output subroutine, and to the address of the PIA port, as advised by the article. But when I tried it all out, it didn't work, and I couldn't figure out why. I abandoned the effort, but I kept my cross-assembler output listing and the MIKBUG-format object code listing. I used the tone detector circuit for another project and it worked fine. Now, almost 40 years later, I have revisited that BYTE magazine article.
The authors stated on page 31; "The user of MORSER must provide a suitable [character output] routine for his or her own system. For example, the OUTEEE routine in Motorola's MIKBUG will work".
Then on page 35 of the article they stated; "The system subroutine used to drive the [output] terminal is called in this section on line 270. It should not use interrupts or any timing loops, since these will upset the timing in MORSER". In 1977 I did not understand that this statement contradicted the earlier statement regarding the OUTEEE subroutine.
The MIKBUG subroutine OUTEEE calls a software/hardware timing loop subroutine while serially transmitting the ASCII character from one data line of a parallel port (Motorola MC6820) to the RS-232C driver circuits. The return from subroutine does not occur until the entire character has been serially transmitted, which takes 33.3 milliseconds at 300 baud rate. So OUTEEE does not work for this application.
CONCLUSION: The MORSER character output interface must use a UART circuit, such as the General Instruments AY-5-1013A, or Motorola's MC6850 ACIA chip.
For example, MORSER would not work on an SWTPC 6800 with MIKBUG, but should work if the SWTPC has SWTBUG and uses an MP-S card for the console terminal.
-73