This has been made impossible by recent TI hardware changes. But who reads my posts anyway.

Well, I read them, I must have missed that one, then. Sorry

Well, I don't care as I have an old hardware revision 0x5

Good news! I added GetPixel( in the ASM version!
It's still a bit primitive, the syntax should look like this:
Code:
. You do need the closing parenthesis, otherwise it won't work.

Download: https://www.cemetech.net/programs/index.php?mode=file&path=/84pce/asm/programs/ICECompiler.zip

Just a suggestion, I think it'd be nice if you could create and edit BASIC programs, by appending tokens and/or inserting tokens.

That will be added sooner or later to the C version, when pointers are implemented as well.

Will the C version include custom color palettes? This would fix one major flaw with CalcMeister's ICE sprite creator since XlibC doesn't have any grey colors!

PT_, you are hope for a weary soul.

So I downloaded the latest version of ICE(from cemetech) about an hour ago, and it has broken one of my programs.
So I stripped down my program to make it easier to do this bug report, and tested it in CEMU.
I originally was using the older version of ICE I downloaded from TiPlanet after I mentioned and you fixed the bug with getkey(KEY), so that version of ICE is from about a month ago.
Compiling the test program with the older version of ICE and running it yields exactly what it should:

A main menu, labeled "Temporary Main Menu", with four options below it.
Using the up and down keys moves the two side arrows up and down along the options.
After pressing Enter on any of the options, you get a screen filled with numbers, letters, and symbols, as it should:

Pressing Enter one more time exits the program.
So with the older version of ICE, everything works fine.
But with the newer version of ICE, 1.5.2, the program starts to fall apart a bit.
Compiling the test program with 1.5.2 and running it gives the main menu, but with a difference:

The fourth option, "4 for SCHED", is repeated twice.
Also, the arrow on the left is facing the wrong way and is slightly deformed.
Then, if you move down one selection with the down arrow key, the left arrow is now duplicated and facing different directions.
Moving down one more, the left arrow looks similar to what it looked like on the first option.
Then, with the fourth option selected, the left arrow is facing the entirely wrong way.
Now here it gets worse, hitting Enter on any of the options, you get a screen filled with number, letters, and symbols, but after it, a rectangle of garbled colored static:

And then hitting Enter once more exits the program, so it doesn't crash, freeze, or lock the calculator.
So I have no clue as to why that is happening, but I tried this on both my physical calc and CEMU, both running Ti 84 Plus CE OS version 5.3.0.0037, and with all of the latest CLIBs (Version 7.5, released on GitHub on Jul 25)

Here I provided a link to a Google Drive folder with the files:
https://drive.google.com/open?id=0B36Z6j9NdZndTE9iQ3RxWVdMNTg

"INPUTPRGM" is my stripped down program
"OLDICE" is the old version of ICE that I originally made the program with, that still works perfectly.
"NEWICE" is the (currently) newest version of ICE from Cemtech, that breaks "INPUTPRGM".
and the rest of the files are the CLIBS I used.

(Also, the test program has a bunch of "defineSprite" at the beginning, those are character sprites for all the letters, numbers, and symbols I use in that program. Yes, I know ICE has it's own text functions, I just created my own text character sprites and used those because I was testing out what I could do with ICE.)

There is currently a bug in which for loops execute one more time than specified. This may contribute to your problems. Would you mind posting source code so we can look for what might be causing this?

That glitch is quite beautiful, but very ugly at the same time.

800 posts

I guess I didnt make it clear, but the INPUTPRGM file I included in the Google Drive link is the uncompiled program. It can just be dropped into SourceCoder, CEMU, or edited in an actual calculator. It's not locked or anything.

Excuse the mess of code that it is, The full program is a huge conglomeration of many various programs I toyed with to see the capabilities of ICE.

It does seem like it is For loops that are causing the problem, however. So I guess I'll have to wait for that to be fixed.

It's a good thing to notify me of that, such that I can fix things. Anyway, I've fixed this bug, and your program works fine now.

This will be a bit of a problem the way that I had to work around this bug Back to the grind!

Okay, a question for y'all: what would you like to see implemented?
I'm almost done with the functions I wanted to add, I'm done with adding pointers, I've fixed strings, added sub( and length( and what more.
What do you want to see added? If it's not a common routine/function, please explain what it does and why it's useful. Thanks!

I think it would be great to have a createappvar( function , or something similar, because the highscore and large data saving isn't very easy using lists. Of course that's only a suggestion .

You know; with pointer support now you could integrate the fileioc lib

So it seems I had a bunch of free time and integrated the fileioc lib!



I've decided to use "sum(', but I can easily switch if necessary.

I would like to see trig functions, or at least the ability to use them in functions directly (which is hopefully easier than implementing them entirely)

Degrees is fine, I guess, makes it easier to implement. (probably)

If someone has some nice sin/cos(/tan) routines, sure. Sadly I don't have such routines, and I can't create my own one.

With all the amazing features you've added, trig functions will be the only thing between me and some of my older incomplete projects. I'm not an asm genius, so I don't know how hard this'd be, but is it possible to find a way to use the OS trig functions?