Program Issues?

[TKD_01]

NOTE: I have solved the problem mentioned in this post. My most recent problem is listed in my fourth posting starting with "Sure enough, it was putAndSprite....".

Recently I have been growing in my proficiency with the z80 programming tasks and methods. Upon gradually increasing the complexity of my programs, I decided to make a simple program to make a sprite move around on screen. Bam! Done! Next, having tackled that, I decided to make the sprite shoot projectiles or bullets, as in Phoenix or Galaxian. Then my problem revealed itself: I made the program just fine. I double checked, triple checked, heck, whatever 20-check is. However, when I compile it, the 'call' and 'ret' functions do not seem to work properly...I attempted switching from Ion shell to Mirage shell, but...to no avail. So now, I'm hoping that someone may be able to help me through this post. (By the way, I'm running on Windows Vista home Premium Edition...if that makes any difference.)
Also, what's interesting is that jumps within the code work fine...meaning that there's nothing wrong with pc. Maybe something I overlooked with the stack?

For my compiling, I used Andys Assembler Compiler package from ticalc.org.
Here's the link:
http://www.ticalc.org/archives/files/fileinfo/179/17967.html

This is only the movement aspect of the sprite code, but it still doesn't compile:

[WikiGuru]

What error are you getting?
I copied the code and it seems to work fine.
I did define group0 = %11111110, maybe that's what you're missing?

[TKD_01]

Really..? Ya, I have the group0 defined in my keys.inc file. Oh, and there's actually a mistype that I just realized. I can fully 'compile' the program, but when I run it on my emulator (Tilem), the error presents itself. When I said "error with compiling", I was meaning that there may be something wrong with the compiler. Not sure though. It may be an issue with Vista..

Three quick questions to help me out, though:
1) Are you running Windows Vista?
2) Are you actually running the program? or just compiling?
3) If you run the program on your computer, what emulator are you using?

Thanks again for taking time to help me out. I really appreciate it!

[WikiGuru]

1. Yes, though it doesn't matter
2. Yes, compiling and running works (MOS)
3. I'm emulating it in PindurTI (fairly accurate to the hardware except for a few areas you're not using)

I've never heard of the compiler you're using, but I know that Brass works fine (one I use), TASM works fine (though is extremely out of date), and SPASM works fine.

Oh, incidentally, I did make another change:
for exit, I used call quittoshell instead of ret (only works with MOS)

[TKD_01]

Hey, thanks for your help. After fiddling around with changing my program header to Ion, and getting a different compiler, I got the program to work again. I appreciate the help! However, now, I've run into an actual issue within my program, and am eager to get it solved. I'm sure you've heard of the game Galaxian for the Atari (a great game, by the way!), and I'm also sure that you've heard of the game Phoenix for the TI-calcs, programmed by Patrick Davis. Well, I'm making a game similar to those two, and have hit a road block in the section of code which handles the on screen explosions. The issue is as follows:

Whenever a bullet makes contact with the player ship, certain bits are set to indicate that an explosion is going on (looking at the program may help to understand). Regardless, whenever a bullet makes contact and an explosion is initiated, a mere blank space is displayed, which is obviously not the kind of explosion I want. Wanting to get on with the programming, I'm open to any suggestions or solutions you or any fellow programmer may present. Thanks ahead of time!

NOTE: The code pertaining mainly to explosions is displayed in red.

[WikiGuru]

The problem I think is in your putandsprite. It's not a good idea to use and for a sprite. instead, i'd use xor sprites and for your last explosion sprite have it erase the result of the other sprites xor'd together. Also, I don't quite understand what's happening with handle explosiondata. If you want to offset to an image, simply multiply by 8. Lastly, I'm a little confused of why you're keeping track of the status, x, and y for the explosiondata 5 times. The explosion isn't moving (when i tested), so you only need 2 bytes, and then 1 byte for the explosion sprite offset, cutting out 12 bytes. This also allows you to put more sprites for the explosion.

ps. I don't know if this is pindur or not, but I couldn't trigger anything until i had left the boundaries, and then it seemed like everything was being checked normally (boundary detection, bullet collision).

[TKD_01]

Sure enough, it was putAndSprite...I changed it to XOR the images to the screen, and that works fine. Plus, I dont know what I was doing with offsetting by 9...I instead changed it to offset by 8, like you suggested. And the one last item regarding why I put the explosion 5 times: I made the list as follows:

[WikiGuru]

For some reason when I tried running your test code, it crashes at dispenemy1. I commented it out, and now it does the ghosting thing you were talking about. I think your error might be somewhere in dispenemy1. I'll examine it later when i have more time.

[calc84maniac]

In dispenemy1 I believe you are calling putenemy instead of putenemy1. Could that be the problem?

[TKD_01]

calc84maniac, you were right. I was in fact calling putenemy, instead of putenemy1. Thanks so much for your help!

WikiGuru, I believe that there may be something wrong with PindurTI, because the call dispenemy1 works perfectly fine on my emulator. Also,the boundary problem you said you had, back a couple of posts, never occurred when I tried to run the program. Is there any particular reason you use PindurTI? If not, have you checked out Wabbitemu? It perfectly emulates the TI83,TI83+,TI83+SE, and the same for the TI84 family. I'm pretty sure it runs TI82s, also.

[WikiGuru]

Yeah, i think you're right. It might just be a pindurti issue. The only reason I use pindur is because it's partially integrated with latenite (lame, i know, but i'm kind of lazy)

[tr1p1ea]

Neither WabbitEmu or PTI 'perfectly' emulate any model of calc however they are very close. Using either is fine. If you believe you have narrowed down a bug, it would be wise to contact the author about it.

[WikiGuru]

To be quite honest, I don't know what the bug is. I think latenite came with a rather old version of pindurti, so it might have been fixed and I just haven't replace the executable yet.

[TKD_01]

oh, ok. WikiGuru, do your other programs run fine on PTI?

tr1p1ea, when I said "perfectly emulates [those calculators]", it was a figure of speech. For what I've run and tested on the WabbitEmu, it emulates fairly close to perfect. Haha, I know that WabbitEmu will always have some flaw or something that is impossible to emulate on the computer, but for the purposes of beginning, intermediate, and even many advanced programmers, it runs flawlessly enough for them (and me). I do appreciate, however, your clarification! Sorry for any misconceptions!

[WikiGuru]

most of them due, but there are a few every now and then that have funny issues (like this one)

[TKD_01]

Alright, that's not so bad. Ya, I tried downloading Pindur at one point. I downloaded it, but after reading the README file, I realized that the person who made the emulator wasn't making any improvements to it, and didn't have any plans to do so in the future. So after a bit more searching, I found WabbitEmu, which has active support and constant updates. Have you ever used Tilem, out of curiousity?

[WikiGuru]

Yeah, but i stopped when i got my laptop. It was kind of a hassle to use, and kept crashing on me. I haven't done much calc programming lately, and usually when I do, pindurti suites my tests. I do have wabbitemu and use it for homework when i'm too lazy to pull out my real calc, or when i want to test something that works only on the 83+se/84+ series (basically anything but the 83+ regular).