Daimakaimura (Ghouls 'n Ghosts) for the Atari STE

[Anima]

https://www.youtube.com/watch?v=YDwfwaBBMrw

[CiH]

So getting closer to complete gameplay. Nice work!

[catmando]

https://www.youtube.com/watch?v=YDwfwaBBMrw

Looking smooth

What are those glitchy horizontal lines that appear every now and again, will they be cleaned from the final release?

[calimero]

this is 25 fps constant?

[wongck]

I am blown away by the speed of the moving sprites.
great work !!!

[marss]

simply superb

[Anima]

So getting closer to complete gameplay. Nice work!

Well, actually the gameplay is complete; Yet the emulation of the CPS1 custom chips isn't.

What are those glitchy horizontal lines that appear every now and again, will they be cleaned from the final release?

It's due to the Blitter sprite routines and the fact that the screen memory areas (display, work and background) are stored in 64 kB pages. The screen display wraps at the line where the Shifter proceeds displaying the succeeding 64 kB page. A Timer B interrupt restores the high byte of the video counter to "jump" back to the previous memory page. So (at least some of) the lines are artifacts from the non drawing areas. So yes, that should be cleaned up in the final version.

this is 25 fps constant?

No. The tile update and sprite drawing is limited for each frame to provide enough time for the gameplay logic to finish. So I guess that the frame rate varies from 50 fps to 12.5 fps most of the time and will drop to a lower value at certain places. The goal is to conserve the gameplay as good as possible on the Atari STE.

[dlfrsilver]

that's great

In fact, the main problem of the CPS system, it's the huge amount of ram needed for the games, otherwise, it's perfectly fine.

[Anima]

In fact, the main problem of the CPS system, it's the huge amount of ram needed for the games, otherwise, it's perfectly fine.

I think I have a good solution for the memory problem thanks to the HD file detection routines by Peter (AtariZoll). This solution is actually the reason why I mentioned that an Atari STE with 2 MB is the target for a port. I have tile/sprite streaming from the HD in mind. However, I'll address this after the CPS graphics emulation is being cleaned up of the glitches.

[Anima]

Short update: the graphics line glitch has been eleminated.

[FedePede04]

some of you atari guys, are truly computer wizard, Keep up the good work

[calimero]

@anima it is Pera Putnik (aka Petari, AtariZoll)
It is really nice to see how different discoveries/technologies are combined into new achievements!

[rj1]

Let us download

[jotego]

Could you please share the download link? I cannot find it.

[Brume]

It's there:
http://atari-forum.com/viewtopic.php?f= ... 50#p329758

[Ragstaff]

I hardly engage in the ST scene or use my Atari's these days, not as much as I'd like to... this project is one of the things that keeps me coming back. Subscribed to the thread

[jotego]

It's there:
http://atari-forum.com/viewtopic.php?f= ... 50#p329758

Thank you very much. I see it is only the original demo. It is nonetheless absolutly impressive! I would have gone crazy for this when I was a teenager!
As an adult this kind of work actually motivates me to keep going for the best on my IC design job. Why? Because these things remind me of why I chose a career in electronics and what it means to do a good job.

Thank you Anima for inspiring us. And keep it up!

[cawley1]

Hi,
Is there any way we can get the version that is working on the Falcon?
Would love to give this a try...

Thanks,
Paul

[Anima]

Hi,
Is there any way we can get the version that is working on the Falcon?
Would love to give this a try...

I'll prepare another test version in the next days which is targeted for the Falcon. Please note that it will contain the same graphic assets as the Atari STE version.

Unfortunately the current renderer isn't compatible with the MC68030. The Blitter code is optimized for the Atari STE and so it needs some additional NOPs to satisfy the faster CPU timing on the Falcon. Also the fine scrolling requires a small fix. I hope that it'll work on MC68030 accelerated Atari STE machines as well. However, the upcoming test version for the Falcon will require 14 MB of RAM.

[Sturm]

Cool ! I'm impatient !

[Estrayk]

I'll prepare another test version in the next days which is targeted for the Falcon. Please note that it will contain the same graphic assets as the Atari STE version.

Unfortunately the current renderer isn't compatible with the MC68030. The Blitter code is optimized for the Atari STE and so it needs some additional NOPs to satisfy the faster CPU timing on the Falcon. Also the fine scrolling requires a small fix. I hope that it'll work on MC68030 accelerated Atari STE machines as well. However, the upcoming test version for the Falcon will require 14 MB of RAM.

We are all expectant!

[dhedberg]

I'll prepare another test version in the next days which is targeted for the Falcon. Please note that it will contain the same graphic assets as the Atari STE version.

Can't wait! Hurry hurry!

[cawley1]

Lol, checking back daily!

[MacFalcon]

Great to see you keep up the good work!

[Anima]

I'll prepare another test version in the next days which is targeted for the Falcon.

I suppose that the term "next days" was too ambitious. Some serious real life things got in between. Also it's quite hard to test the Falcon version on Hatari due to some incompatibilities; so I need more time testing on the real machine which is a bit inconvenient. So far there are some glitches in the background and the bus error handler needs some fixes as well.

However, the Atari STE has still the top priority and I am currently working on the dynamic sprite/tile loading routines.

Stay tuned...