XB256 revised

[senior_falcon]

Attached is a zipped folder containing the revised version of XB256. XB256 and 256DEMO are XB programs. The docs are in a PDF file and there is a Win994a disk image containing XB256 and 256DEMO. This fixes the problems that some were having with disk access. It works with all the simulators except PC99. I had made some changes to code that I wrote 20+ years ago thinking that I am smarter now than I was then. Not so!

 

To get you started, load and run XB256, then load and run 256DEMO.

 

(Updated XB256 April 10, 2013)

XB256.zip

Edited April 11, 2013 by senior_falcon

[1980gamer]

I have just taken the time to run the demo! AWESOME!

The pixel left/right is exactly what I am looking for, I started a demo of space invaders with full char moving 8 pixels at a time.

I didn't like it enough to create graphics or colors etc. But if I can use pixel movement and also have collision detection of the chars...

I'll pick it back up.

 

DEMO before XB256 attached.

Gene

INV3-C.zip

[1980gamer]

Oh, your ship is the letter A. keys S and D move you left and right the E key fires.

I did not put an end in the demo, if you let the invaders hit the ground... or below, it will corrupt memory!

Gene

[sometimes99er]

The pixel left/right is exactly what I am looking for, I started a demo of space invaders with full char moving 8 pixels at a time.

I didn't like it enough to create graphics or colors etc. But if I can use pixel movement and also have collision detection of the chars...

I'll pick it back up.

 

Ha. Very nice.

[jacquesg]

Harry,

 

I have downloaded the 2 program files using MagicFM and run XB256. I now get 'I/O error 04 in 10' while with the previous version I would get a lockup with the DC light staying on. I understand that 04 means an OPEN error caused by a disk full or too many files opened. I tried changing the XB256 parameter to 3 and 5 and I get the same error message. I get the same error message if I RUN 256DEMO from the command line.

 

As with the previous version I can move the exact same two program files to PC99 and they run except that the Starwars like scroll locks up after displaying the first line.

 

Puzzled in Bathurst,

 

Jacques

[senior_falcon]

Puzzled here also. Magic FM makes me believe that you are transferring the two files to a real T.I. Is there anything different about your hardware - it seems like you mentioned a different disk controller? Does Missing Link run with this equipment?

 

PC99 seems to be unhappy with this. Lucien tested it with all the emulators he had and it worked fine with all except PC99.

[unhuman]

Well, there goes one of my ideas! I was going to try to make Inaccurate Invaders more Accurate. Anyhow, you should be able to do some logic to track where in the characters the invader is so you could do some simple logic to know where the invaders are offset by x pixels and compare that to the bullet location.

 

I have just taken the time to run the demo! AWESOME!

The pixel left/right is exactly what I am looking for, I started a demo of space invaders with full char moving 8 pixels at a time.

I didn't like it enough to create graphics or colors etc. But if I can use pixel movement and also have collision detection of the chars...

I'll pick it back up.

 

DEMO before XB256 attached.

Gene

[jacquesg]

Puzzled here also. Magic FM makes me believe that you are transferring the two files to a real T.I. Is there anything different about your hardware - it seems like you mentioned a different disk controller? Does Missing Link run with this equipment?

 

PC99 seems to be unhappy with this. Lucien tested it with all the emulators he had and it worked fine with all except PC99.

 

Yes, I am trying to run 256DEMO on my real TI. XB256 will run fine with the cursor returning after running the program. However my attempt to do any disk function such as RUN or OLD of any file returns the error message 'I/O error 04'. I can run TML with no problem. My system uses a SNUG BwG DC which I understand is similar to a CorComp DC. The BwG DC also has a CLOCK device which perhaps plays havoc with the number of open files.

 

Lucien is running PC99 with a TI DC while I am using a Myarc DC. I can run XB256 and the DEMO file on PC99 iust fine except for the hangup that occurs when running the third option (vertical scroll).

 

Jacques

[1980gamer]

Well, there goes one of my ideas! I was going to try to make Inaccurate Invaders more Accurate. Anyhow, you should be able to do some logic to track where in the characters the invader is so you could do some simple logic to know where the invaders are offset by x pixels and compare that to the bullet location.

 

No, knock yourself out! I was not a real big space invader fan to be honest. However, I was always very impressed with all of the conversions that came out.

Not for arcade perfect work, but just for pushing systems to do something they shouldn't be able to do.

 

The 2600 has all those invaders and only has 2 sprites, a ball and ?? can't remember, but even tripling the sprites, how do they do it?

So I was curious, and put that little demo of the chars, changing and moving to the edge, dropping a row and moving back.

 

That was all done in arrays... in my demo if you were wondering.

 

I have a scrolling driving game, I'd rather try (maybe) can I layout a MAP of the world and then scroll through it?

AUTO RACING l complete rip-off of the Intellivision version. @ 10 years old, I was amazed!

 

Actually, I would love to see parsec! on the intellivision... HINT!

 

Cannot wait to see how XB256 develops.

Gene

[senior_falcon]

Jacques, I have looked carefully at the differences between XB256 and TML. I made a couple of tiny changes to the program and have attached it to this post. It is very hard for me to see how this will make any difference. If it doesn't then I'm at a loss to understand how one can work and the other not work. It all works fine on my equipment - TI99/4A, Ti controller, fast memory, horizon ramdisk. If it works, let me know and I'll break out the champagne. If not try changing line 10 to open just 1 disk file and see if that makes a difference.

 

Harry

XB256.zip

[jacquesg]

Jacques, I have looked carefully at the differences between XB256 and TML. I made a couple of tiny changes to the program and have attached it to this post. It is very hard for me to see how this will make any difference. If it doesn't then I'm at a loss to understand how one can work and the other not work. It all works fine on my equipment - TI99/4A, Ti controller, fast memory, horizon ramdisk. If it works, let me know and I'll break out the champagne. If not try changing line 10 to open just 1 disk file and see if that makes a difference.

 

Harry

 

This version gives me the same result as the original version of XB256. which is, XB256 loads and cursor returns but my attempt to run 256DEMO lights the DC card but the disk light does not come on. The sysem just hangs and I have to turn if off.

 

With the last revised version I had a similar result but an 'I/O Error 04' message was produced and the system did not hang. A cursor returned but any attempt to do a disk function from the command line would repeat the I/O error message.

 

I know one other person who uses a BwG DC. I will ask him if he gets the same results as I do. We both run Funnelweb except that he runs an 80 column version.

 

I think that the BwG DC card is the culprit but I do not know why.

 

Jacques

[lucien2]

With MESS and the BwG controller, the version from post 1 runs fine and the version from post 10 hangs during OLD 256DEMO.

 

Maybe there is more than one version of the BwG DSR? Or MESS does not emulate it 100% correctly?

[Tursi]

The 2600 has all those invaders and only has 2 sprites, a ball and ?? can't remember, but even tripling the sprites, how do they do it?

 

OT, but: 2600 has two sprites (8 pixels wide and as tall as you want them), a ball (an 8 by X block) and two missiles (8x1) (plus a 20 pixel background that you can mirror or duplicate to get 40 pixels across). But it also has hardware to copy the sprites horizontally, and you can set up everything every scanline, as well as changing things mid-scanline with careful cycle counting. A combination of these tricks lets them pull off the seemingly higher numbers of sprites.

 

Reading the 2600 bases here at AtariAge will give you more technical information than you ever wanted on how these, and other tricks, work.

[1980gamer]

2600 even with copying horizontally... 123123123 I shoot the right most 2 in the copies... How do I not remove ALL 2's...?

 

Actually I am not much of a 2600 guy. How would Colecovision pull this off? Same as TI would be my guess!

 

Don't want to hi-jack this thread from XB256.

Gene

[Tursi]

2600 even with copying horizontally... 123123123 I shoot the right most 2 in the copies... How do I not remove ALL 2's...?

 

It's even harder, it's 121212 - but with the interleaving there is enough time to change the next '2' while '1' is being drawn, and vice versa. They have to time it carefully, but it's stable. And.. yeah, I will un-hijack this thread too, there's a whole forum here for that.

 

[senior_falcon]

I've been doing some revisions to XB256. First off, disk access now works properly with a real TI99/4A. Also the screen changes color in the command mode so you know it's loaded and active.

 

There are a couple extra subroutines. CALL LINK("DELAY",number) will pause program execution for the specified amount of time. CALL LINK("CWRITE",string) will write a compressed string to VDP RAM. Where do the compressed strings come from? There is a utility called COMPRESS that can be merged into an XB program and called up once the graphics, screen, colors, sound tables etc. have been created by your XB program. COMPRESS can embed a flag in a sound list so that CWRITE knows to start playing the sound.

 

Start it up with RUN "DSKn.XB256" then RUN "DSKn.256DEMO" This is the same demo as before except that it plays "Hell in Texas" while it is doing the screen graphics. Then, just to give a taste of what COMPRESS and CWRITE can do, I took the agonizingly slow 256 character and 28 sprite demo and converted the graphics into data statements. 256DEMOCW reads the data statements and writes them to VDP - you will notice a huge increase in speed.

XB256.zip

[Opry99er]

Really looking forward to start work with this. Very very cool stuff, Falcon... Thank you for your work.

[unhuman]

I am going to really appreciate the delay!!! Thanks!

 

One of these days! Got my ideas. Now need the time.

[Opry99er]

Okay guys, here's something awesome for you...

 

http://youtu.be/C8hZbV1DrU8

 

Riding for the Brand with a new and improved intro. Thanks falcon!!!

 

I'm looking forward to getting deeper into this thing and really making some cool improvements as I wind down my baseball game efforts. This is a very impressive program!

[Opry99er]

Falcon, could you describe in a little more detail precisely how this effect is accomplished? I'm aware that it is a LINKed assembly routine, but I was wondering if you could expound on it some... Or, if I've missed something, please link me to the proper post so that I can read up.

 

Thanks, and awesome program. =)

[Opry99er]

Check that.... RTFM. Got it now.

[senior_falcon]

Yep, you should RTFM before you RFTB.

[Opry99er]

Ha!! =) Yea... reading it now.

 

It might be in this manual, but can XB256 handshake with the compiler? By that I mean, can assembly subs like this one be compiled in with an XB program?

 

I'm assuming the answer is no, but I thought I'd ask the author before I make assumptions...

[senior_falcon]

There is no easy way to add CALL LINK to the compiler. I can see ways that might work OK but probably not super fast and there is a slight possibility of that in the future, but don't hold your breath waiting. The intention with XB256 is to add the A/L subs to the compiler so that you can at least use these subroutines with it. A few of these are included with the compiler already - DISPLY, VREAD, VWRITE, EARLY. The rest of XB256 won't requiire too many modification - mainly changes to how values get passed which is different in the compiler as compared with XB.

[Opry99er]

Thanks... I was just a bit curious. =) I'm going to try to do some work on Honeycomb Rapture this weekend if I can finish the other project I'm working on. Once again, sir... great work here. Seeing that title screen scroll like that makes me smile every time. =) Maybe if that was in place during the RDGC, I might have stood a chance. =)

 

The other games I was going against were pretty awesome though. =)