Jump to content
IGNORED

Quantizator


ilmenit

Recommended Posts

Beta4:

- Details mask added /details=inputfile /details_val=value

- Improved dithering algorithms

- New dithering algorithm (/dither=knoll)

- Dithering strength added (/dither_val=value)

- /distance param replaced /euclid

- new distance color function (CIEDE2000) that solves problems with "too gray" output

- /noborder param removed

 

EDIT: Source code attached, No more time today to push it to Github...

RastaConverterBeta4.zip

Beta4Src.zip

Edited by ilmenit
  • Like 2
Link to comment
Share on other sites

With regards to Beta3, any chance of a version with no allegro as an option?

 

If so, I'll do another BeagleBone version.

 

I merged ilmenit's latest changes into my fork. I added you as a collaborator if you'd like to upload another Beaglebone build. I also just received my own Beaglebone in the mail from Amazon a few hours ago and I just got your build running on it. I'm getting around 220 evaluations per second with only USB power. Amazing little device.

Link to comment
Share on other sites

I merged ilmenit's latest changes into my fork. I added you as a collaborator if you'd like to upload another Beaglebone build. I also just received my own Beaglebone in the mail from Amazon a few hours ago and I just got your build running on it. I'm getting around 220 evaluations per second with only USB power. Amazing little device.

 

Ahh, excellent, thanks for the merge. I'll do a new build soon, though I'm not sure when because I have a bit of overtime to do at work this weekend.

 

Yes, it's a great bit of kit, though it would be nice if they supplied a power supply with it. I'd ordered mine from Farnell. You can get power supplies from them also if you need one. I do find that there isn't much information on the internet about the BeagleBone.

 

Thanks for adding me as a collaborator by the way.

Link to comment
Share on other sites

Beta4 is now uploaded. It appears to be quicker as I'm getting around 300 evaluations per second after 50,000 iterations, but it really depends on the picture.

 

We may be able to bounce some ideas off each other with regards to the BeagleBone. It is a little quirky. Installed 'man' pages would be nice.

Edited by snicklin
Link to comment
Share on other sites

Beta4 is now uploaded. It appears to be quicker as I'm getting around 300 evaluations per second after 50,000 iterations, but it really depends on the picture.

 

We may be able to bounce some ideas off each other with regards to the BeagleBone. It is a little quirky. Installed 'man' pages would be nice.

 

I guess running rastaconv on a Cortex-A8 could benefit a lot from converting the usage of floats and doubles to fixed point math.

Link to comment
Share on other sites

Beta4 is now uploaded.

 

I realized my fork doesn't have all the Beta4 changes since ilmenit still hasn't pushed all of them to github yet.

 

I guess running rastaconv on a Cortex-A8 could benefit a lot from converting the usage of floats and doubles to fixed point math.

 

EDIT: I had posted that the summing and comparing of distances were still floating point, but like phaeron said, it's integer math. You could speed up the initial distance computations and dithering but those don't take too much time.

Edited by Xuel
Link to comment
Share on other sites

On around version 196 of this project in the year 2025, maybe we could have it connected into a central database, with a centrally stored datastore of pictures which people have worked on. Then it stores different people's pictures after 10K, 100K, 1M, 10M, 100M etc evaluations. The user who has the best quality has their image stored, then other people can work off that. It would be a form of distributed computing.

 

I can only dream...!

Link to comment
Share on other sites

I have some new feature requests:

 

1) Pause. I'd just like some way that I can pause it and continue later, without stopping it and then continuing it from the command line. Maybe hitting the space bar would do this? Would there be much of an overhead in scanning the keyboard?

2) A gallery function. Could it create and store a picture after 100,000 iterations, 1M, 10M, 100M etc? This might seem a bit useless, but it would nice to be able to show comparisons on here of different pictures using different techniques/settings and how well they are converting.

3) A way of supplying a filename for the generator, rather than having to have all my files begin with "output".

Link to comment
Share on other sites

Just for fun, here's another video that abuses the fact that block loads are instantaneous in Altirra and AtariWin800. (Haven't tested Atari800)

 

- Video: Fullscreen 160x240, 4++ bpp

- Audio: 4-bit 15.6kHz

- Length: 1m23s

- Uncompressed size: 46MB

- Link: im.zip

 

Next step is getting this working on real hardware using my Beaglebone. :grin: Need to buy some level shifters...

  • Like 5
Link to comment
Share on other sites

Just for fun, here's another video that abuses the fact that block loads are instantaneous in Altirra and AtariWin800. (Haven't tested Atari800)

 

- Video: Fullscreen 160x240, 4++ bpp

- Audio: 4-bit 15.6kHz

- Length: 1m23s

- Uncompressed size: 46MB

- Link: im.zip

 

Next step is getting this working on real hardware using my Beaglebone. :grin: Need to buy some level shifters...

 

 

LOL

Link to comment
Share on other sites

:(

 

My BeagleBone won't work. As soon as I kill the Putty session, rastaconverter is killed. I can put it in the background and I can use the Linux "screen" command, but it doesn't matter what I do, due to a bug in Angstrom, it'll kill my processes.

 

I need to work a way around this.... , I want to leave it processing all day and night.

Link to comment
Share on other sites

Just for fun, here's another video that abuses the fact that block loads are instantaneous in Altirra and AtariWin800. (Haven't tested Atari800)

...

Next step is getting this working on real hardware using my Beaglebone. :grin: Need to buy some level shifters...

 

Oh my god... I can't belive my eyes and ears! :-D

 

Btw, next time try to play with higher values of /dither_val (=2 or 3) if your images get too gray. It works especially better with 'chess' dithering.

Also /distance=ciede may solve the grayness problem, but the initial preprocess can take a lot of time then.

Edited by ilmenit
Link to comment
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...