Bouncy

[Iwantgames:)]

Playing with Magellan is fun

 

metamaps.mag

[TheMole]

I tried adding a few sound effects but it doesn't work well with this music - it's breaking up too much.

 

Alas, but the speech is a great alternative! Have you considered encoding a sound effect as speech?

[ti99iuc]

I tried adding a few sound effects but it doesn't work well with this music - it's breaking up too much.

 

Rasmus, you could also think for a possibility to "enable/disable" Music option in the game ? so who prefer to listen Sound Effects will be able to choice. it's just an idea

[Asmusr]

Playing with Magellan is fun

 

metamaps.mag

 

Great, but the last level has too many character transitions (36). Your first level also had one too many, but that was easy to fix by moving the coin at position 2,23 one position up.

 

In the analysis tool you can click on a transition to see where on the map it appears. Start with the ones with the fewest occurrences.

metamaps.mag

[Iwantgames:)]

Darn coulda sworn I checked musta forgot. I'll fix it after work

[Asmusr]

 

Alas, but the speech is a great alternative! Have you considered encoding a sound effect as speech?

 

I thought lpc speech encoding was a lost art. Does any tools exist for current operating systems?

[Asmusr]

Darn coulda sworn I checked musta forgot. I'll fix it after work

 

This version has the first two of your levels included and starts on level 5 with the first one.

 

I have also included the source file for one of the levels if you're interested in how it's coded in assembly. It's really small and simple because all the transformations into a scrollable character set/map is done by the TI program.

BOUNCY.dsk

level5.a99

[Iwantgames:)]

Wow I got game over 1 go around on 1st map and I designed it 2nd maps pretty easy tho.

 

I fixed the 3rd map:

 

metamaps.mag

[Tursi]

 

I thought lpc speech encoding was a lost art. Does any tools exist for current operating systems?

 

Someone was working on a tool here on the forum -- I've actually been meaning to dig up that thread recently and ping for progress.

 

Otherwise, I think the MAME emulation is close enough now that we might be able to try some organic brute force conversions (which is essentially what that other tool was going for ).

 

Otherwise the art is Mark Wills' to own, IIRC he's got QBox down to a science.

[Iwantgames:)]

I found my 2nd map to be too easy so I moved a couple enemies and added a couple.

 

metamaps.mag

[Asmusr]

I found my 2nd map to be too easy so I moved a couple enemies and added a couple.

 

metamaps.mag

 

Here's a version with your three maps, starting at level 6 with the second map. I had to make some minor adjustments to the last map because transitions that appear by switching on tiles have to be present elsewhere in the map. This is why I added the lava tiles to the beginning again. The last map is pretty hard, and I haven't completed it yet, but I think it can be done. There are 8 levels now and only a few hundred bytes left.

BOUNCY.dsk

metamaps.mag

[Iwantgames:)]

Sweet! Can't wait to get home and try it out

[Iwantgames:)]

It is possible to beat my last level, just hard I game overed 6 times before making it

[Asmusr]

Here's the latest beta version with 8 levels. I'm out of space, so the game would have to move to cartridge in order to add anything else.

 

Thanks to Iwantgames:) for maps and testing, and to Tursi for VGM tools and XB loader.

BOUNCY.dsk

[Iwantgames:)]

Cool I'll give it a go tonight don't know what else you'd want to add. Id think it'll be tough enough to get through all 8 levels

[+InsaneMultitasker]

I don't know that you would want to go this route, but what about adding a level loader.

[Iwantgames:)]

I don't know that you would want to go this route, but what about adding a level loader.

Like a Bouncy Cartridge that could load custom levels from floppy disc? Now that would be cool

[+InsaneMultitasker]

Like a Bouncy Cartridge that could load custom levels from floppy disc? Now that would be cool

Yea, that's what I was referring to. Sort of like Tunnels of Doom

[Asmusr]

Yea, that's what I was referring to. Sort of like Tunnels of Doom

 

An interesting idea, but I wonder if anyone would use it? A cartridge version could, in theory, contain hundreds of levels, multiple tile sets, new tunes, etc., but I don't think the gameplay has enough variation to make such an effort worthwhile.

 

I have attached the source code if anyone would like to play around with modifying the levels. You need xas99 to assemble it.

 

Edit: There's something for Tursi to look into in Bouncy.a99 at line 712 .

bouncy-beta-source.zip

[sometimes99er]

*      Then copy remaining sprites starting from a changing position
       MOV  @SPR5TH,R4          * Get number of 5th sprite on a line
       JNE  FLICK1
*      No 5th sprite - just start at the beginning

Nice one.

 

Of course this is added for a reason. Why (or was this discussed somewhere else) ?

 

I am/was going to use barrel shift of 4 sprites through 12 sprites (for my Restless II game).

 

[Tursi]

Edit: There's something for Tursi to look into in Bouncy.a99 at line 712 .

 

Classic99 fakes the sprite rendering because it draws the entire screen at once, not per scanline... so currently it can't really return a meaningful value there.

[Asmusr]

 

Classic99 fakes the sprite rendering because it draws the entire screen at once, not per scanline... so currently it can't really return a meaningful value there.

 

Sorry for being cryptic. The issue I found is that Classic99 sometimes set the 5th sprite flag bit with the number of the 5th sprite reported as 31 even though there were far fewer sprites on the screen. This is by polling the VDP interrupt through CRU and reading the status reg at vsync. This made the program crash before I added a check that the number of the 5th sprite made sense. I verified that the program did not crash on the hardware and in MESS/JS99er. I can create a crashing version for you if you're interested.

 

Edit: It looks like the problems appears only once the first time I read the status flag after polling the interrupt from CRU. I get a status value of >FF.

 

Edit 2: Add a breakpoint for W833A=001F in the latest version. This will break at line 688 in the very first call to VSYNC.

[Asmusr]

*      Then copy remaining sprites starting from a changing position
       MOV  @SPR5TH,R4          * Get number of 5th sprite on a line
       JNE  FLICK1
*      No 5th sprite - just start at the beginning

Nice one.

 

Of course this is added for a reason. Why (or was this discussed somewhere else) ?

 

I am/was going to use barrel shift of 4 sprites through 12 sprites (for my Restless II game).

 

 

 

If there are more than 4 sprites on a line I give the 5th sprite highest priority next frame (highest after the ball, that is). This seems to create less flicker than other rotation schemes I tried.

[sometimes99er]

If there are more than 4 sprites on a line I give the 5th sprite highest priority next frame (highest after the ball, that is). This seems to create less flicker than other rotation schemes I tried.

Okay.

 

As sprites usually move around, I guess there's no system to guarantee a sprite not being left out in the cold.

 

Edited October 30, 2015 by sometimes99er

[artrag]

I did the same on msx.

The solution of moving on top the sprite reported by the status register is absolutely optimal if there is only one sprite disappearing on the screen.

It is a bit disappointing if there are two lines with 5 sprites or if there are more than 5 sprites on the same line...

On this regard many emulators fail to report the correct value of the 5th sprite when there is more than one disappearing sprite.

IIRC the tms9918 locks the first 5th sprite value in its status register until you read it.

Many emulators keep the value running so you get usually the last occurrence of the 5th sprite situation.

Edited October 30, 2015 by artrag