Jump to content
rbairos

Movie Cart

Recommended Posts

On 8/18/2021 at 8:04 PM, rbairos said:

As part of this project, I also created an automatic box + label graphics maker, I can demo at some point if anyone's interested.
I had a lot of fun creating possible 1970 movie candidates for the final 5 movies I made:

The resolution is too low in this attachment, but all the movie + copyright year info is accurate, and easily editable.

I used https://www.stickeryou.com/  for the vinyl cart stickers (they were great)
and staples copy center for the boxes + instruction manuals (cheap + good quality).
I used a variety of online effect generators like https://www2.lunapic.com/ and some others to get the painterly effects from the photos.
I put everything together in a custom editor I wrote for TouchDesigner.

 

over1.jpg

show the DL demo......i love DL :)

 

Share this post


Link to post
Share on other sites
2 hours ago, gambler172 said:

show the DL demo......i love DL :)

 

What's DL ??

Share this post


Link to post
Share on other sites

So I've started using the newly freed up cycles to see how things might improve if I was able to change the background color each scan line.

In the tests below, the left side is the original, while the right side selects a allows a color change once per scanline.
This causes a big improvement on scenes with solid colored backgrounds.

(I've also made a change to the dithering algorithm to clamp values to 0..1 after distributing error. This eliminates some ghosting and noise after high contrast areas).


I think the changes are markedly better.

Dragon's Lair:
https://youtu.be/5taAXQ1z6Zo

Star Wars Desert Scene:


What's Up Doc? Credits:
 


Dragon's Lair:
 

 

Edited by rbairos
  • Like 9

Share this post


Link to post
Share on other sites
1 hour ago, rbairos said:

I think the changes are markedly better.

 

Vastly improved. These are just proof of concept changes at the moment are they?

Share this post


Link to post
Share on other sites
19 minutes ago, JetSetIlly said:

Vastly improved. 

This! :thumbsup:

Share this post


Link to post
Share on other sites
1 hour ago, rbairos said:

In the tests below, the left side is the original, while the right side selects a allows a color change once per scanline.
This causes a big improvement on scenes with solid colored backgrounds.

(I've also made a change to the dithering algorithm to clamp values to 0..1 after distributing error. This eliminates some ghosting and noise after high contrast areas).


I think the changes are markedly better.

Star Wars Desert Scene:


What's Up Doc? Credits:
 


Dragon's Lair:
 

 

:-o

Share this post


Link to post
Share on other sites

Thanks everyone.

It's still of proof of concept.
Both the original (left), and proposed (right) are missing the final flicker, so there's that.

A little more work should get it emulator next.

I should be able to change background color every 2 scanlines with the current kernel.
But if I switch the flicker back to alternating columns (instead of alternating checkerboards), I should be able to switch every scanline.

To be honest, in hardware, the alternating checkerboards often looked like alternating columns once it got to the screen, so probably not anything lost switching back to alternating columns to get even more cycles back.

Another challenge is picking the background color.
Right now I just brute-forced it, and test all 128 backgrounds per line to pick the best.  Resulting in an 8 hour render time of the 5 minute dragon's lair!  Normally it runs about 20% speed.

Share this post


Link to post
Share on other sites
3 hours ago, rbairos said:

Another challenge is picking the background color.
Right now I just brute-forced it, and test all 128 backgrounds per line to pick the best.  Resulting in an 8 hour render time of the 5 minute dragon's lair!  Normally it runs about 20% speed.

Maybe you can observe a pattern of the final color chosen. So that you can guess the correct color and then only check the nearby colors.

  • Like 1

Share this post


Link to post
Share on other sites

Okay, so I realized, even though I'm now using background as a separate color, I was accidentally still using it to average each 8 pixel foreground block.  This is wrong.

Fixing that resulted in sharper transitions between foreground and background now:

(Left is original, middle is using background, right is ignoring background for foreground colors)



 

res1.jpg

res2.jpg

res3.jpg

res4.jpg

res0.jpg

Edited by rbairos
  • Like 7

Share this post


Link to post
Share on other sites
20 minutes ago, rbairos said:

 

res0.jpg

Huge improvement.  I really want to see some more video material!

Also, wanted to ask if you carry pixel errors across frames to give some time-based correction there?

 

 

Share this post


Link to post
Share on other sites
7 minutes ago, Andrew Davie said:

Huge improvement.  I really want to see some more video material!

Also, wanted to ask if you carry pixel errors across frames to give some time-based correction there?

 

 

I've restarted the 5 minute Dragon's Lair . Hopefully tomorrow morning I can post.

I tried time-based correction at one point.
Didn't get the results I was hoping for.
Was expecting the star-wars star field to cause more twinkling spots, for stars that got filtered out normally,
but ended up with solid-coloured areas.
I may have coded it wrong.
Might investigate again.

PS. I need to hurry up and push these changes to the emulator, to make sure its not some sweet illusion.


 

Edited by rbairos

Share this post


Link to post
Share on other sites

Here's the next test with the better foreground separation.
Note in this case, it includes the simulated 60-hz checkerboard flicker, so its best scene at 720p60
 

 

  • Like 9

Share this post


Link to post
Share on other sites
37 minutes ago, rbairos said:

Here's the next test with the better foreground separation.
Note in this case, it includes the simulated 60-hz checkerboard flicker, so its best scene at 720p60

Well firstly, this is magnificent. Amazing job; I'm in awe.

 

I have reviewed and I think/hope there are some minor improvements you can do. I have observed that the yellow is too bright and almost white, so you lose a lot of detail in yellow/white areas. And the purple/blue choice seems to be incorrect. I've grabbed a couple of screenshots where this is (to me) clearly apparent. If possible, I think a review of the RGB to '2600 colour choices you are making might improve the visuals even more. Of course I may be misunderstanding the capabilities, but I'm hoping...

 

 

378264977_ScreenShot2021-08-31at9_58_44pm.thumb.png.0256c223b338383a9098f3ac8590abfa.png

 

In the above, the purple platform is a poor match.

 

1166154461_ScreenShot2021-08-31at9_57_51pm.thumb.png.b2b41c764db60b6ab0b3abdca1ebf596.png

 

Another instance with platform colour choice, and also here the red pit is very yellow instead of red/pink.

 

1052889699_ScreenShot2021-08-31at9_55_47pm.thumb.png.f00fbfbcdae831f4d3f78915719c4bfd.png

 

Here, the yellow is not contrasting enough - a shade darker would be nice.

 

1797747872_ScreenShot2021-08-31at9_54_33pm.thumb.png.ead551da2ef47c511b97f5028d2840e9.png

 

Again, yellow a shade darker might work...?

 

 

So, well done again. Hoping you can find even more improvements. What a great bit of work :)

 

 

Share this post


Link to post
Share on other sites

rbarios, I see you're probably using a DVD-Video source. 
I don't know if there could be improvements in your (awesome) job if you'd use another "master"... maybe the process could run smoothly if done with a clip which has different resolution or color gamma.

If you'd like I have some different captures of NTSC LaserDiscs, some other from PAL LaserDiscs and also other edited clips taken from HD sources, matched to the NTSC LD frame number/length. 

I personally produced some of these clips years ago for the Daphne Team to have them working on the emulator, and they were shared through torrents (now "dead").

I don't deny that HD clips have their "charm" but in my opinion the best are those taken from the LDs and "reworked" just enough to remove traces of "laser-rot" and small imperfections. 

Share this post


Link to post
Share on other sites
3 hours ago, Andrew Davie said:

 

In the above, the purple platform is a poor match.

 

 

Another instance with platform colour choice, and also here the red pit is very yellow instead of red/pink.

 

 

Here, the yellow is not contrasting enough - a shade darker would be nice.

 

 

Again, yellow a shade darker might work...?

 

 

So, well done again. Hoping you can find even more improvements. What a great bit of work :)

 

 

Thanks!
In the above, the same colors its picking for encoding are the same colors its displaying.
So what I think is going on, is I'm preprocessing the colors too much.
I made a patch that plays with gamma and contrast, and tries stretching out each red,blue,green range independently to its max.
Maybe I should relax those now that the color selection is improved.
And use your screenshots as a guide.



 

Edited by rbairos

Share this post


Link to post
Share on other sites
51 minutes ago, macdlsa said:

rbarios, I see you're probably using a DVD-Video source. 
I don't know if there could be improvements in your (awesome) job if you'd use another "master"... maybe the process could run smoothly if done with a clip which has different resolution or color gamma.

If you'd like I have some different captures of NTSC LaserDiscs, some other from PAL LaserDiscs and also other edited clips taken from HD sources, matched to the NTSC LD frame number/length. 

I personally produced some of these clips years ago for the Daphne Team to have them working on the emulator, and they were shared through torrents (now "dead").

I don't deny that HD clips have their "charm" but in my opinion the best are those taken from the LDs and "reworked" just enough to remove traces of "laser-rot" and small imperfections. 

Likely a DVD source. Just something I quickly grabbed off youtube.
Not sure if different sources would improve the output much, as so much is already lost in the encoding with all the atari hardware limitations.
I'll play around with my pre-processing stage though.
Cheers.
 

Share this post


Link to post
Share on other sites
40 minutes ago, rbairos said:

Thanks!
In the above, the same colors its picking for encoding are the same colors its displaying.
So what I think is going on, is I'm preprocessing the colors too much.
I made a patch that plays with gamma and contrast, and tries stretching out each red,blue,green range independently to its max.
Maybe I should relax those now that the color selection is improved.
And use your screenshots as a guide.



 

Also, as an aside, my observations are that intensities are not the same for all colours.  That is, brightness of 14 is not the same for colour x as it is for colour Y.  White is, for example, quite bright even at intensity 10.  Blue (say, colour 8 ) is not nearly so bright. So perhaps your conversion may need to take this into account, too?  I guess I'd do it by getting the Stella palette RGB values (at least it's a starting point) and using those for doing the matching.  In my experiments in displaying images, long ago, I found that indeed stretching the range and increasing contrast made a big difference.

Share this post


Link to post
Share on other sites
3 hours ago, rbairos said:

Likely a DVD source. Just something I quickly grabbed off youtube.
Not sure if different sources would improve the output much, as so much is already lost in the encoding with all the atari hardware limitations.
I'll play around with my pre-processing stage though.
Cheers.
 

My thought was not for a simple qualitative increase... I mean that a clip which originally has fewer compression artifacts, more defined colors and margins can facilitate the work of many video data decoding and compression programs and so improve the results. It was a fact at least for the work I had done during the "rips" of the LaserDiscs. Obviously I don't know the methodologies used in this case... in "my" case I mainly used ATI acquisition interfaces, post-processing programs like VirtualDub and lot of disk space due to "non-compression" de/encoding systems like HuffYUV.

Daphne_Differences+info.jpg

  • Like 1

Share this post


Link to post
Share on other sites
3 hours ago, Andrew Davie said:

Also, as an aside, my observations are that intensities are not the same for all colours.  That is, brightness of 14 is not the same for colour x as it is for colour Y.  White is, for example, quite bright even at intensity 10.  Blue (say, colour 8 ) is not nearly so bright. So perhaps your conversion may need to take this into account, too?  I guess I'd do it by getting the Stella palette RGB values (at least it's a starting point) and using those for doing the matching.  In my experiments in displaying images, long ago, I found that indeed stretching the range and increasing contrast made a big difference.

I use the Random Terrain palette as I recall it looked closer on real hardware, though I made the stella palette an option as well.
In my search, I use a simple algorithm where I compare Euclidean distance for best match.
I experimented with HSV searches as well, but similar results.
When picking the color to best represent each block of 8 pixels, I use an average, weighted by its distance from the background.
In both cases, I weigh green 60%, red 30%, blue 10%.
I think I am distorting the color too much before this stage though.



 

Share this post


Link to post
Share on other sites
56 minutes ago, macdlsa said:

My thought was not for a simple qualitative increase... I mean that a clip which originally has fewer compression artifacts, more defined colors and margins can facilitate the work of many video data decoding and compression programs and so improve the results. It was a fact at least for the work I had done during the "rips" of the LaserDiscs. Obviously I don't know the methodologies used in this case... in "my" case I mainly used ATI acquisition interfaces, post-processing programs like VirtualDub and lot of disk space due to "non-compression" de/encoding systems like HuffYUV.

Daphne_Differences+info.jpg

Those look interesting.
I think I can receive similar varieties with the preprocessing I do.
Color enhancement is done with TouchDesigner, which is a full-fledged real time graphics editor.
Right now, Im just blindly cutting 15% off the sides, to zoom in the middle a bit.
In terms of compression artifacts, my encoder butchers things so severely, I don't think the original artifacts stand a chance unfortunately.
 

Edited by rbairos

Share this post


Link to post
Share on other sites
5 hours ago, rbairos said:

In both cases, I weigh green 60%, red 30%, blue 10%.

I know the eye sensitivity for colours is the thinking behind this, but I am not understanding why the above is necessary here.

This is just going to skew the colours, no?  Have you tried just a simple average?

 

Share this post


Link to post
Share on other sites
6 hours ago, rbairos said:

Those look interesting.
I think I can receive similar varieties with the preprocessing I do.
Color enhancement is done with TouchDesigner, which is a full-fledged real time graphics editor.
Right now, Im just blindly cutting 15% off the sides, to zoom in the middle a bit.
In terms of compression artifacts, my encoder butchers things so severely, I don't think the original artifacts stand a chance unfortunately.
 

Ahhhh ... I missed the pre-processing part :D .

Can I see, of course if allowed, a pre-processed videoclip just before being fed to the encoder ? It's just a curiosity...

Share this post


Link to post
Share on other sites

The video quality is fantastic! Would it be possible to create an AtariVCR Cart with the encoder and port for a usb cam to film movies right on the Atari or view video in real time?

 

That would be very interesting for games and utilities. It could enable video conferencing capabilities for PlusCart for example, where text conferencing has already been implemented. 

 

Share this post


Link to post
Share on other sites

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...