Jump to content
IGNORED

Images generated by RastaConverter


Philsan

Recommended Posts

Some more fuel for the fire!

 

I desaturated the shadows on the julia bond pic to remove her chicken wings...

 

post-23476-0-39213700-1345158958_thumb.pngRevEng_JBond.xex

 

 

Some classic pin-up gals...

 

post-23476-0-02685800-1345158973_thumb.pngRevEng_PinupGal01.xex

post-23476-0-37192600-1345158975_thumb.pngRevEng_PinupGal02.xex

post-23476-0-81854600-1345158955_thumb.pngRevEng_HawleyCoke.xex

 

 

Some historical photos...

 

post-23476-0-40592700-1345158953_thumb.pngRevEng_amelia.xex

post-23476-0-06646400-1345158951_thumb.pngRevEng_Ali2.xex

 

 

And an experiment mixing a strong pre-dither in the background (via gimp) and floyd dither 0.5 strength for Mona's face (via Rasta)...

 

post-23476-0-40347300-1345158961_thumb.pngRevEng_MonaLisa.xex

  • Like 7
Link to comment
Share on other sites

I'm back from my holidays so I will start to work on RC soon :)

In the picture from the Amberstar the detail mask on the person should have been used. Wonderland by Piesiu has too many colors to be handled by RC, but the result is acceptable.

post-22831-0-45771100-1345450045_thumb.pngilmenit_carrion-fire.xex

post-22831-0-10310700-1345450041_thumb.pngilmenit_carrion-face.xex

post-22831-0-10041900-1345450047_thumb.pngilmenit_cows.xex

post-22831-0-40251000-1345450048_thumb.pngilmenit_giant.xex

post-22831-0-68153500-1345450050_thumb.pngilmenit_wonderland_by_piesiu.xex

post-22831-0-27434900-1345450132_thumb.pngilmenit_amberstar.xex

  • Like 6
Link to comment
Share on other sites

Use the attached files and the following command line to see the effect of details mask:

RastaConverter.exe amberstar_title.png /h=200 /filter=box /details=amberstar_details.png /details_val=4

For the best effect you should use details only in one area in each line. Remember that forcing the algorithm to put more details on some area of the picture will cause the rest to look worse.

You should also increase the /s number for a better output.

 

With details mask the error (colour distance) of a pixel is multiplied by mask brightness rescaled to range [1,2), where 1 is black and 2 is white, multiplied by details_val.

 

From the FAQ:

"Best areas to cover with more details are not the same for a human eye and for a computer. Computer works on the color distance while humans prefer to see more details for example on faces, small but important objects, center of a picture, objects in the first plan etc. Use the /details mask to choose where do you want to have details."

post-22831-0-16903400-1345455360_thumb.png

post-22831-0-54839500-1345455361_thumb.png

Edited by ilmenit
Link to comment
Share on other sites

Detail mask example

 

Each picture was created with /s=100 and /max_evals=4000000 (4 million iterations).

 

No details mask:

post-22831-0-84657500-1345457422_thumb.png

 

Mask from the previous post and /details_val=4:

post-22831-0-28010800-1345457429_thumb.png

 

Mask from the previous post and /details_val=15:

post-22831-0-31093600-1345457430_thumb.png

 

Without a details mask the guy lacks of details. On the second picture the amount of details on the robe is much higher, but the head of the bird got a bit red. On the last picture (with the mask too strong) the robe is well detailed but the bird wings got colors of the robe and the roof on the castle lost colors for the sky-face beard details.

Edited by ilmenit
Link to comment
Share on other sites

So, this is the new piccy thread. I haven't played with it for awhile, so I thought I'd try to see what the latest version could make of an abstract landscape. This photo has areas of relatively flat color, so I knew that it would come out pretty well, but I'm pretty impressed with the early outcome, thus far. I'm going to let it cook some more, then append the file here.

 

Some information on the photo:

 

Source and author copyright:

 

http://www.flickr.co...in/photostream/

 

Background info:

 

https://en.wikipedia...Roche,_Cornwall

 

Go ahead and click for a doube-sized image. :)

 

Sorry about all of the errors in the image. They were present in the original destination image, but they didn't bother me too much, so i decided to continue instead of restarting. :)

post-3306-0-95397000-1345564009_thumb.png

AtariNerd_rocherock.xex

  • Like 3
Link to comment
Share on other sites

I think 8bit artists will always like to have a pixel-perfect control over their work. But... the newest version of G2F will support loading RC output for further edition :-)

 

My artist uses his ST-tools to make the pic. However, he is not an 8-Bit guy...yet :)

 

BTW, would you recommend to use "double pixel" before conversion, or will the result of RC be better when he uses full 320x200 res?

We are a bit under time pressure, so there is not much time to experiment on our side. :(

Link to comment
Share on other sites

The input picture should be as close as possible to Atari capabilities if it is not photo-like. The "double pixel" will help, proper Atari palette (f.e. laoo) will also help. Both of you should try to start experimenting with the converter as soon as possible - it does not guarantee good effects always. You can also drop me a private message with the picture and I will tell you how the conversion should be done.

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...
×
×
  • Create New...