Prog8 language and compiler topic

All aspects of programming on the Commander X16.
User avatar
desertfish
Posts: 1097
Joined: Tue Aug 25, 2020 8:27 pm
Location: Netherlands

Re: Prog8 language and compiler topic

Post by desertfish »

are you using uninitialized memory somewhere?
note that memory is not filled with $00 at startup.

or something else is changing the ram bank.
yock1960
Posts: 136
Joined: Tue Nov 16, 2021 8:42 pm

Re: Prog8 language and compiler topic

Post by yock1960 »

desertfish wrote: Fri Jun 23, 2023 3:33 pm are you using uninitialized memory somewhere?
note that memory is not filled with $00 at startup.

or something else is changing the ram bank.
I left the rambank(0) command out, it doesn't seem to matter, in the emulator anyway. I fixed the missing files issue in the WIP post.
yock1960
Posts: 136
Joined: Tue Nov 16, 2021 8:42 pm

Re: Prog8 language and compiler topic

Post by yock1960 »

Another oddity, I mention only to be....informative...

While experimenting with the .wav files for my project, specifically to see if I could get one to fit in a single bank of hiram...which I did, by chopping it down a bit, I discovered that while the shortened .wav file (time wise) and also encoded at 8000hz, all performed in Audacity, would play fine using streaming (stream-wave), but not using play-adpcm. I haven't pored over the code to see if there's a difference, but there must be. I haven't tried using the 'approved' conversion programs, but figured compatibility in one would carry over to the other program. Well, I'm still happy being able to stream .wav files, period!
User avatar
desertfish
Posts: 1097
Joined: Tue Aug 25, 2020 8:27 pm
Location: Netherlands

Re: Prog8 language and compiler topic

Post by desertfish »

The play-adpcm code is very picky about the encoding format, maybe Audacity encoded it in a different format.
The decoder understands ONLY ima-adpcm encoded files with a block size of 256.
In the code comments of the adpcm.p8 file you can find a couple of commands on how to produce such a file.
yock1960
Posts: 136
Joined: Tue Nov 16, 2021 8:42 pm

Re: Prog8 language and compiler topic

Post by yock1960 »

desertfish wrote: Sun Jun 25, 2023 6:00 pm The play-adpcm code is very picky about the encoding format, maybe Audacity encoded it in a different format.
The decoder understands ONLY ima-adpcm encoded files with a block size of 256.
In the code comments of the adpcm.p8 file you can find a couple of commands on how to produce such a file.
Cool! Being a 'soft' gui type user, I tend to shy away from command line only tools, but I just can't decipher Audacity's myriad of export, menu choices....none of them specifying anything as helpful as 'ima-adpcm'. After several attempts failed, I downloaded sox. First attempt success! Any tips on converting files for use with the zsound demo? I tried coverting the same source .wav file, using the sox command line example on the zsound Github docs page and then recompiling play-zsound to use this file. Silence and a page filling raster effect, when activating the digi-sample trigger. I tried with and without adding a dummy 2-byte header as also suggested, but deemed unnecessary. Although, if I recall correctly, it didn't load until I added the dummy header.
yock1960
Posts: 136
Joined: Tue Nov 16, 2021 8:42 pm

Re: Prog8 language and compiler topic

Post by yock1960 »

Irmen,

I know I'm not the sharpest guy, but for the life of me, I can't figure out why this code doesn't work correctly, using version 9.0 on R43. I read a list of top scores (10 scores, 0 to 9) and player initials from a file, into to string arrays, this works fine, printing them gives the expected results. Then I pass a new score to this subroutine. the 'top_score' value is actually the lowest high score, so if score is lower, we exit. Otherwise, find where it fits, move everything down 1 slot from the insertion point, bumping out the lowest hi score. I get a double entry with the new score, and lose the score that was to be the first score moved down, everything else is correct, the old lowest score is gone. I've tried this with both a for loop and a while loop...same result. I'm stumped. Could be bad logic on my part....wouldn't be the first time, or even the 10,000th! Even stranger, there is no double entry in the scores array in memory, although the lowest score is still there....I'm confused!

sub check_score(uword score) -> ubyte {
ubyte n
ubyte i

if score < top_score {
return(10) }
else {
for n in 0 to 9 {
if score > conv.str2uword(scores[n]) {
for i in 8 to n step -1 {
names[i+1] = names
scores[i+1] = scores }
string.copy(p_str,names[n])
conv.str_uw0(score)
string.copy(conv.string_out,scores[n])
return(n)
}
}
}
}
User avatar
desertfish
Posts: 1097
Joined: Tue Aug 25, 2020 8:27 pm
Location: Netherlands

Re: Prog8 language and compiler topic

Post by desertfish »

can you format the code in mono or pre to fix the indenting? It's kinda hard to read right now
yock1960
Posts: 136
Joined: Tue Nov 16, 2021 8:42 pm

Re: Prog8 language and compiler topic

Post by yock1960 »

desertfish wrote: Sun Jul 09, 2023 3:57 pm can you format the code in mono or pre to fix the indenting? It's kinda hard to read right now
I came up with a version that works...second snippet. I still don't understand why the first snippet doesn't work though.

non-working:
snippet1.jpg
snippet1.jpg (242.42 KiB) Viewed 9360 times
working:
snippet2.jpg
snippet2.jpg (245.33 KiB) Viewed 9360 times
User avatar
desertfish
Posts: 1097
Joined: Tue Aug 25, 2020 8:27 pm
Location: Netherlands

Re: Prog8 language and compiler topic

Post by desertfish »

I think I understand what you're trying to do here. In the first code snippet, is p_str the pointer to the name for the new high score entry?
At first glance I don't see anything wrong with the first snippet. I will have to compile it myself in a small test program to see how it behaves, but this will take a while to figure out.
Maybe you're triggering a prog8 compiler bug! I have fixed a few compiler bugs since the 9.0 version of the compiler came out, so perhaps the code works fine with the current development version of the compiler. If you feel comfortable you can get it from a CI build artefact on the github actions.
(I'm planning to release version 9.1 next week btw.)

P.S.
copying code from a screenshot is ... kinda hard ;)
you can paste the code here and surround it with the mono or pre tags or use those markup buttons above the text entry, this way we can simply copy and paste the program text
yock1960
Posts: 136
Joined: Tue Nov 16, 2021 8:42 pm

Re: Prog8 language and compiler topic

Post by yock1960 »

desertfish wrote: Sun Jul 09, 2023 8:00 pm I think I understand what you're trying to do here. In the first code snippet, is p_str the pointer to the name for the new high score entry?
At first glance I don't see anything wrong with the first snippet. I will have to compile it myself in a small test program to see how it behaves, but this will take a while to figure out.
Maybe you're triggering a prog8 compiler bug! I have fixed a few compiler bugs since the 9.0 version of the compiler came out, so perhaps the code works fine with the current development version of the compiler. If you feel comfortable you can get it from a CI build artefact on the github actions.
(I'm planning to release version 9.1 next week btw.)

P.S.
copying code from a screenshot is ... kinda hard ;)
you can paste the code here and surround it with the mono or pre tags or use those markup buttons above the text entry, this way we can simply copy and paste the program text
Here's a zip file with a couple of demo programs.....you'll have to rename them to .p8 from .asm...(my editor only allows .asm files.....I know :roll: ).

hiscore doesn't work correctly, hiscore4 does, so far as inserting a new high score into the 'top 10', but I'm working on writing the updated data to a file...which seems to work, but when I go back and try and re-read the updated file, it fails and I haven't figured out why yet. On re-running the emulator and program, it loads/displays fine again. Technically, I might never need re-read the data file during a program run (thinking 'out loud' now), strictly working on already filled arrays....but it seems like this should still work.
Attachments
hiscores.zip
(2.91 KiB) Downloaded 451 times
Post Reply