This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - Halifax
Pages: 1 2 [3] 4 5 ... 90
31
« on: May 26, 2009, 06:59:26 pm »
BASIC: Length of Integer
1+iPart(log(N
This will return the number of the digits that make up the number. Note the use of Integer, meaning that it will not work on floating-point values.
Ex: N = 1234, Output = 4
z80 ASM: Pow2 Multiplication Trick
;a = input
ld d,a
ld e,0 ; de = a * 256
srl d \ rr e ; de = a * 128
srl d \ rr e ; de = a * 64
etc.
z80 ASM: Raw Key Input Optimization
ld a,$FF
out (1),a
ld a,$FE
out (1),a
nop \ nop
in a,(1)
becomes
ld a,$FF
out (1),a
dec a
out (1),a
ld a,(de)
in a,(1)
As you can see, the use of the first optimization can only occur when analyzing the key group $FE. The second optimization takes into account the fact that a will be destroyed by the port read, but the instruction will still delay for the same amount of time that two nop's will.
32
« on: May 26, 2009, 06:41:41 pm »
What example programs are you looking for exactly? Do you just want to be able to compile the code he provides, or what? Try to be a little more specific.
Also, spend more than one day on a lesson that is proving to be difficult. You want to learn everything inside-out. Don't just gloss over anything. To be a good assembly programmer you really need to know the internals; you can't just fudge your way through because that will just add to your frustration.
I really recommend SPASM to you, that's your best bet for an assembler. TASM really doesn't cut it. I realize RevSoft crashed, so I contacted someone and retrieved a copy of the source. If you need a SPASM binary, or the source, then just PM me.
You could get by if you skipped Week 3, but I don't recommend it because it is good stuff to know.
33
« on: May 25, 2009, 10:06:43 am »
yeah but im just pointing out that when garbage collecting or defragmenting, the indicator is slower
Okay, fair enough.
34
« on: May 25, 2009, 01:53:38 am »
DJ: just because the runindicator is slow, doesnt mean the calc version is faster. when my calculator defrags or garbagecollects, it is really slow, but it might be faster on a calculator anyways, but i just wanted to point that out.
Well if the emulator is correct, then I believe it should indicate that the calc version is faster. The run indicator is executed just like any other assembly that the OS contains, meaning that it is interpreted and performed at the same speed any other code is. So I believe that it is a good indicator.
35
« on: May 19, 2009, 09:55:39 pm »
Wow {AP}, that looks sick. Congrats on getting most of it done, and keep up the good work!
36
« on: May 18, 2009, 07:52:20 pm »
I hope no one rips off anyone...
If somneone did rip another person's special idea, then I'm almost sure that they would be called on it. Maybe I'm wrong, but that's at least how I would hope for it to be.
37
« on: May 18, 2009, 07:49:17 pm »
If I can get the time, I might enter as well..
I agree. To be honest, I really do like these contests. It helps to spark a motivation in me which is usually hard to find.
38
« on: May 18, 2009, 01:48:01 pm »
thanks everybody (mostly DJ and Halifax) but everyone else too, and now my problem is solved
Yeah, no problem.
39
« on: May 18, 2009, 01:55:02 am »
aww, rat's oh well. This was more of a platform engine then a game though
Rules are rules Eeems. If they made the exception for you, and accepted it just because it's more of an engine than a game, then they would have to do it for everyone else. Meaning that they could release the engine behind their game and retrieve help from the community. The whole point of the contest is to work independently and to show what you yourself can do.
40
« on: May 17, 2009, 11:16:07 pm »
Aah ok, that's cool then. I saw the link I was just unsure ^^
I don't know what other game used Bin2Var though, because it was so long ago when this problem was being talked about. I think United-TI, Calcgames (site, not forums) and TI-Freakware didn't even existed. Almost all info regarding this was either scattered all around the web on forgotten sites/comment pages or completly lost
Ah, I see. You know, if I ever have some time one day, I could always go back through the ticalc.org archives and compile a list of what games did use Bin2Var, and patch them if necessary. And yeah, I can tell it must have been a long time ago, because I've never heard or seen Bin2Var. I think I'm going to do a little digging for it though. EDIT: Well, I found Bin2Var on ticalc.org: http://www.ticalc.org/archives/files/fileinfo/143/14394.html. (By the way, that link is to Bin2Var 1.2) It was released in 2000. The porter of Joltima is also the author of Bin2Var, and he generated the files for Joltima with Bin2Var 1.1. He released the C source code for both Bin2Var 1.1 & 1.2, so I had a look at it. As I read through the source I found where he made his mistakes. He takes a binary dump (the binary representation of assembly code which can be generated by almost all assemblers such as SPASM, TASM, etc.) and he retrieves the size of it. He never adds the necessary 2-byte number of tokens section to the front of the data and also does all size calculations on the raw data. So that's where all the problems stem from. Well, that should clear everything up.
41
« on: May 17, 2009, 11:09:33 pm »
Oooh yeah Bin2Var, now I remember. This is what caused the transfer problem.
Should I update the RPG archives on Omnimaga with the new version? (when they are finished)
Yes you should. And the new versions are finished, I posted the link above: http://www.mediafire.com/download.php?imqr3wtkdmj
42
« on: May 17, 2009, 11:07:17 pm »
Alright, just as I thought. The header was clearly chopped and screwed by some application named Bin2Var, and the fact that anyone even got TiLP to send it to the calculator is amazing. Even more amazing is the fact that anyone even got it to run after being sent to the calculator because the header wasn't just incorrect, but the variable format too. Also, you were right. TI-Connect reports errors about the version because the file was specifying a value that isn't even in use. Anyways, here is a patched version of Joltima that works: http://www.mediafire.com/download.php?imqr3wtkdmSo now, just send those two files to your calculator with TI-Connect, un-archive JOLTIMA and run it with "Asm(prgmJOLTIMA)", and the game should start up no problem. I have verified this on my end with my TI-84+ BE. (By the way, make a note that you need 22 KB of RAM free to even be able to un-archive Joltima.) So maybe it has something to do with what Iambian said about ZJLOAD? I don't know, I tried using ZJLOAD directly, but it didn't do anything, so I had to un-archive Joltima itself. Anyways, if you have anymore problems with my patched files, just ask! Otherwise, have fun with the game. (From the little bit I played, it seems pretty fun.  )
43
« on: May 17, 2009, 07:52:16 pm »
First thing, never trust extensions. I can rename any test.8xp file to test.exe, and then have people wondering why they can't run the executable file on their computer. It's all about what lies under the hood in the binary.
Upload you exact file, that you're trying to send, and I will have a look at it. It's probably just something minor regarding the file header.
44
« on: May 17, 2009, 07:47:41 pm »
you mean that they were intentional
Haha. Either way you say it makes sense. "Just making sure [the numbers] weren't intentional" is the same as "Just making sure [the numbers] were intentional". By the way, this is a dual-layer platformer I'm guessing? And are those screenshots with 6 MHz?
45
« on: May 13, 2009, 08:30:48 am »
Yeah, it's not supposed to look better (as I said in my previous post, it will be of reduced quality). At any rate, are you sure you disabled everything, such as the grayscale interrupt (if you're using one) and your just doing straight one-layer black-and-white for the background with no overhead other than your tilemapper? For some reason I was thinking that you should at least get a minimal decrease in the number of milliseconds it takes to render. Even 1 to 2 milliseconds will increase the FPS by 1 in 6 MHz mode.
It was simply a switch from alternating the mask every frame to leaving the mask the same every frame. There's no fancy interrupts or anything.
Ah alright, that explains it. Thanks calc84maniac.
Pages: 1 2 [3] 4 5 ... 90
|
Show Posts