Jump to content

Recommended Posts

Posted (edited)

To get us started, here's my first idea for an effect. It could still be refined a lot, e.g. the boxes could grow and shrink, but you'll get the idea. (Please don't post any videos of this effect.)

2PSCROLL

Edited by Asmusr
  • Like 8
3 hours ago, PeteE said:

This is the effect I've been working on. The placeholder image and text would need to be replaced, but the code seems to work well enough.  (No videos please)

Very nice. It has that "How did he do that?" quality.

  • Like 1
  • Thanks 1
On 5/26/2024 at 11:44 PM, RamonB5^dSr said:

I've always believed that the MSX1 and TI systems are quite compatible in terms of technical capabilities. Following @PeteE lead, I explored some MSX demos for inspiration.

We already borrowed a lot of stuff from those demos for the first demo, but keep in mind that the Z80 is generally faster than the TMS9900. I like the last plasma effect of 'Invasion Of The Big Pixels' and I'm not sure know how they did it. 

 

I'm currently thinking about an isometric scroller that could be fed either from generated or stored data. It wouldn't be smooth, but maybe quite fast?

 

image.png.5744d1ff8319aecbe003f75249555caa.png

 

 

 

 

  • Like 3

Another idea I have thought about for years: Implement the ultimate algorithm for converting an image to the 9918A taking into account that you can have 4 additional sprites per scanline. We will run out of sprites after 128 scanlines, but then we can switch to another sprite attribute table. The picture is from an old demo of mine where you see 48 sprites on the screen at the same time.

image.png.e950619eaced95c345c0a6a08f98e5a3.png

  • Like 5
Posted (edited)

Dear Asmusr,

 

Your idea reminds me of following rhetorical question I asked the member Tursi a few years ago:

 

Tursi's answer was following:

 

If your idea is meant this way, then please go ahead.

 

 

Kind regards

Edited by MueThor
13 hours ago, Asmusr said:

Another idea I have thought about for years: Implement the ultimate algorithm for converting an image to the 9918A taking into account that you can have 4 additional sprites per scanline. We will run out of sprites after 128 scanlines, but then we can switch to another sprite attribute table. The picture is from an old demo of mine where you see 48 sprites on the screen at the same time.

image.png.e950619eaced95c345c0a6a08f98e5a3.png

I think there are a couple of image converters that attempt this.

 

The main issue I see is once you commit to a sprite, you've committed for at least 8 scanlines - whether there's pattern on those lines or now, it will contribute to the four sprites on a line limit. So basically, for each character row, you can get 4 8x8 pixel single color overlays. Will it help? Yeah! But automating it to be the best possible is a challenge.

 

My brute force concepts are defeated by this, since you've got 4*8*8*15*256 possible layouts. (4 sprites, 8 rows of 8 pixels, one of 15 colors, and one of 256 positions.). Made even more complex when you consider that you are applying that to the existing pixel layouts - and how you lay out those pixels can be impacted by the sprite selection.

 

You can get more coverage with double-size 16x16 sprites, of course, but it also commits you for 16 rows.

 

The commit might be controllable by changing the sprite table, but since most register changes take a couple of scanlines, I wonder if it's possible to do that to any value.

 

Anyway, that's why I don't want to tackle it. My code is pure brute force and we aren't brute enough yet for that. A genetic algorithm running on a GPU could potentially do a good job though! (I've seen this done on the C64, actually, it's stunning to watch ;) )

 

  • Like 1
10 hours ago, Asmusr said:

Could you post a link?

 

Unfortunately, no. One of my Imagineer friends did it after seeing my image converter, so I've only seen it run live. I don't think I was ever able to finagle the code from him ;)

 

I did have the non-GPU version of it.. I remember it took a good 20-30 hours to converge. The GPU was done in seconds and you could watch it live. 

 

Would some of the dancing outline effects used in the "State of the Art" Amiga demo by Spaceballs be too pedestrian for a TI demo?  What about the motion video stuff done for "9 Fingers" from the same group?

1 hour ago, OLD CS1 said:

Would some of the dancing outline effects used in the "State of the Art" Amiga demo by Spaceballs be too pedestrian for a TI demo?  What about the motion video stuff done for "9 Fingers" from the same group?

Ahh, dancing outlines was what I was planning to do. If you've already started, go ahead.

 

  • Like 1
12 minutes ago, Tursi said:

Ahh, dancing outlines was what I was planning to do. If you've already started, go ahead.

Not me.  I do not have the time to do any demo work, and I still have other projects to do before I start playing.  I was asking more for a measure of opinions toward effects.

Posted (edited)
1 hour ago, TheMole said:

Wait, what? Does anyone have any idea how this is done?

It could be F18A. That YouTube channel has a few other F18A videos. But it's hard to see because the video is so grainy.

Edited by Asmusr
52 minutes ago, ti99iuc said:

I would like to dream a state of the art dancing woman on the ti99 but maybe it has a too much low resolutiion for that kind of graphics.

Considering what has been done for "Bad Apple" on the 4A, I suspect otherwise.  In fact, regarding "State of the Art," I imagine the dancing outlines could be stored as vectors.

  • Like 1

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...