Forum comments in chronological order

Anonymous
Wed 5-Jun-2019 06:01

Haha, this article IS the summary

Anonymous
Wed 5-Jun-2019 22:21

I can't seem to build this. backend.c and backend_z.h seem to be missing from the tarball.

Anonymous
Wed 5-Jun-2019 22:55

Cool stuff!! :)

lft
Linus Åkesson
Wed 5-Jun-2019 23:16

Whoops! Should be fixed now in 0g/02, which is otherwise identical.

Anonymous
Fri 7-Jun-2019 00:01

How do a program should propperly handle a "/dev/pts"? I mean, I am able to open that device for read and write, the program is effectively writing and reading stuff, but when I want to close the program gracefully and close the file descriptor used to open the device, the program crashes with no clue on what has happened.
Thanks for the excellent work!

Anonymous
Sat 8-Jun-2019 23:57

Having read your wonderful article, am feeling obliged to give my thanks. BIG thanks! :)

Anonymous
Sun 9-Jun-2019 23:50

I love it! Some great leaps / lateral thinking. Spectacular code Linus!

Anonymous
Thu 13-Jun-2019 06:58

What you are doing is simply awesome. Keep on going.

bertjerred
Bert Jerred
Sun 16-Jun-2019 06:45

Fascinating and fun. Thank you for sharing this!

Anonymous
Wed 19-Jun-2019 08:45

Great article! I have integrated the code into my disk utility and it allows me to scan the entire disk in 21 seconds using an interleave of 2 sectors!! I only send the first 4B and an error byte from each sector back to the C64, so that's why I can hit interleave of 2. I used the short checksum snipet from Shards of Fancy. I also noticed in Shards of Fancy that you use the "nop" for density 0 and 1 (Tracks 25-35), not just the slowest ones (31-35).

Now for some improvements!!
1) instead of storing $0100,$01FF->$0101, I store $01FF->$0100 so the data does not have a wrap. This is accomplished by starting the stack at $FF instead of $00, and also changing the final few instructions in the zero-page from pha before tsx to tsx before pha. This works because pha does not change the zero flag.
2) if the first byte of data is not $55, I give up after 3 tries and report error 04 for that sector and move on. also, if parity fails, give up after 3 tries and report error 05. This is because my project is a disk utility and not a loader :) I cannot "hang" trying forever.
3) save/restore the last 8B of stack, and SP, just before/after this routine, so I can still jsr/rts everywhere else in the code.
4) save/restore a portion of ZP before/after reading all the sectors I wanted, so I can return to KERNAL when done. Returning to KERNAL is important for my utility project. I am using $86-$E5 for the ZP code and only save/restore $99-$B4 -- the rest is just zero'd out.

Cheers!

Anonymous
Wed 19-Jun-2019 21:18

this is very helpful indeed. thanks a lot!

fieserwolf

Anonymous
Thu 20-Jun-2019 21:03

Amazing stuff! Thank you!

Anonymous
Sun 23-Jun-2019 16:52

That got me going, thanks. Have you ever considered putting this in a public repository so that others can contribute, too?


I really like the idea of this system: it's a nice sweet spot between writing and programming.

Anonymous
Tue 25-Jun-2019 01:00

http://www.linusakesson.net/games/autosokoban/?v=1&seed=1783561528&level=100

Writing this down just to remember it, a nice little level worthy of playing. Some interesting tricks in it.

Anonymous
Thu 27-Jun-2019 06:53

This is amazing... well done!

z/DFM

Anonymous
Sun 30-Jun-2019 08:53

Hi! I have a "?" signs for non-Latin characters in the debugger. Is there some way to get around this problem?

IBelanszky
Sun 30-Jun-2019 18:03

Hi Linus, tried to catch up with you on IRC to recommend this sweet two part BBC docuseries on contemporary minimalist composers:
https://www.imdb.com/title/tt8095518/

Hope you enjoy it, cheers!