Im behind you all of the way. I have always looked for something like this, now i just hope you finish it.

Wow. Now maybe I can make an "asm" program real fast and impress you unsuspecting friends. Heh heh.

hmm....i dont like the sound of that evil laugh....BTW the email at the top is my real one....heh heh...

By the way, am I the only person having trouble claiming the programs I've uploaded to the server here? I get a "SLQ command failed" or something like that (im not sure about the letters)

It's probably an SQL error. E-mail the entire error (copy and paste it) to the address above in my comment and it will get resolved ASAP.

anyone know about how the speed of this compares to the speed of an asm program...is this gonna mean the end of asm programmers...C made asm unnesesary for the most part...i'll give this thing a try...
...i'm a little confused, how does this compare to small C being developed for the 86?..
very interesting!

The converted programs should be faster than their Basic counterparts. However, I would assume that at this stage, the compiler isn't particularly great at optimization, and therefore the converted object code wouldn't be as fast as an ASM program built from the ground up in Assembly.

With time and work, this can get better, but I highly doubt it means the end of Assembly programmers! While C did *lessen* the usage of Assembly, C is built to be far more efficient than TI-Basic, not to mention far more powerful. If anybody adapted a Z80 C compiler to compile for specific TI calculators (I assume something like what the small-C project is, but I'm not sure), that would be far more powerful (due to C's nature) than this TI-Basic compiler.

Don't get me wrong! I think the converter's a great idea!

Now if only someone will write a Lisp interpretor for the TI-85 ;-)

A Small C port to Z80 and Usgard TI-85 has been done and works wonderuflly. the only thing that is not so good about this (and the basic compiler, i am sure) is size. Speed is a small problem, but not that small. It is the size that is the biggy. on the computers, where we have 10gig drives with 128megs ram, size isnt a problem, or not much of one anyway, and so the compilers are acceptable. on the 27.5K total space for both saving programs and executing them, (on the ti-85) the c compiler makes code that is simpley too big for anything big like a good game. a c complier would be more feasible spacewise on the ti-86/89/92 systems, with more ram.

Are you referring to C as the compiler is on the calculator, or code is compiled on the computer then transfered? It wasn't clear...

I was talking about a compiler on the computer, then transfered to the calculator.

Now as for the Lisp interpreter, however ;-)

on the pc then transfered. not really enoguh room on calcs for a compiler hehe.
i wish

a lisp interpretter would be kinda cool
i havnt written lisp ion a long time or even looked at any code :) hehe
but the ease...

I'm missing out on a joke or something. I'm new in this ti-world, what the hack are you talking about lisp? I though he was talking about people who can't pronounce "s", but I believe I'm mistaken. could someone explain?
Ps that's not my real name

I think LISP is a programming language. Just pretend you never heard of it.

I did a quick BASIC/ASM comparison with the same prog. It was very simple, just a neverending Lbl-Goto loop where 1 was added to a variable, recorded as that variable, and then displayed. The compiled ASM prog was roughly 2.5 times faster than the basic program. I would have made a faster and better Basic comparison, but the compiler had trouble recognizing my "For" loops.

Yoy need somthing like this for the TI-83!

2 words: Kick Ass!

~Jake B

I want to know one thing: Will this support numbers bigger than 65535?

It will, but if you want very fast variable manipulation, you must tell TI-BASIC Compiler to treat certain variables as integer variables, which have a maximum value of 65535. If you tell it to treat a variable as a normal floating point variable, it's value is virtually unlimited (as with a normal calculator variable).

uh oh, Am I the only one to forsee floods of test based assembly programs on ticalc.org in the upcoming days?

I hope the server can handle it!

doleboy

This is awesome!!! Does the running speed increase when it is tokenized?

There seems to be a bit of confusion here. Tokenizing a TI-BASIC program converts it from pure text format to a series of tokens (where each token describes a number, variable, command, etc). All TI-BASIC programs are automatically tokenized when they are executed on the calculator. After they have been run, they are stored in the calculator's memory in tokenized format. Only when programs are in this tokenized format can they be compiled (at present).

Maybe i'm missing something here, but if it only does it on the calc, then how do you get the progs on the comp to be tokenised. or are they already that way?

Ignore my last comment. I had yet to read the help file.