Generally the process I use is iterative.
For complex images that need to be converted from hi-color graphics I do a first pass where I resize the image to 240x200 and bump the contrast and brightness to use the full range of Oric colors. Then I use the "libpipi" (Sam Hocevar) converter to get a base image that displays on the Oric, but I output to PNG format instead of HIR so I can easily see how bad the result is.
Often fixing the image is as simple as separating the elements on the screen to give a chance to the converter to put attributes where possible.
image_2025-09-01_083553838.png
On the Wolfenstein image you can see that the two characters as well as the wall were converted automatically, but the logo was completely cleaned out to be "pixel clean" and then a gradient added to make it nicer.
You can also see that there's a 6 pixels black column between the hero and the wall to ensure the wall stays blue, and then there's black on the right of the wall to limit the glitches on the soldier, and the helmet was forced painted to white.
image_2025-09-01_083840460.png
On the Encounter image I painted the background characters in black and dark shaded the right side of the front character to avoid glitches.
So basically on the Oric you need to work with "where" are things on screen so you can avoid things to impact each others.
Me personally, everything is done in PaintShopPro 9, with layers, using a 6 pixels wide grid so I have everything aligned properly so the converter can exploit alignment to put the proper attributes, or things like "6 white pixels" will be drawn just by using invert video if they are aligned and there was some black around.
You do not have the required permissions to view the files attached to this post.
