Blastfrog Posted October 30, 2012 I'd rather not derail the Mocha Doom thread, and this is more about what's "correct" in general than relative to any particular port, anyway, so I've created this thread. Maes and I have been discussing color tinting effects like pain/berserk and others relative to colormaps in a truecolor software renderer. Mocha Doom generates the colormaps starting from the color tinted base palette first, and then applies the darkness levels. I suggest that it'd be better to use palette 0 as a base for all of the color tinted colormaps, and apply color tint after the darkness levels have already been calculated, but he insists that his method is better. Maes says that id using the same colormap across the different palettes was a way to save memory and that they probably didn't put much thought into the feature, which is true, but I still think that it looks more visually coherent to keep the same general appearance as the original blending method. I put together two sets of colormaps to demonstrate what it is that I'm talking about. On the left is the results of what my proposed method of colormap generation would look like, and on the right is representative of what the results of Mocha Doom's method looks like. What are your thoughts on this? 0 Share this post Link to post
Gez Posted October 30, 2012 I think I prefer the version where the palette flash does not override darkness entirely. Pure red is not a red-tinted black, it's pure red. 0 Share this post Link to post
qoncept Posted October 30, 2012 I think your colormap is more true to the original, but I prefer Maes'. I'd like to see both in game but Mae's looks like it should give a lot more contrast and end up more playable. I always try to avoid these berzerks and invulnerability because I hate the palette changes. 0 Share this post Link to post
Dragonsbrethren Posted October 30, 2012 Yeah, the left one is definitely truer to the original, and I prefer that. (The radsuit palette in Maes's looks too subtle, too.) 0 Share this post Link to post
RestlessRodent Posted October 30, 2012 Why do you guys not solve for the color shift values? i.e. solve for the color difference in each channel to determine how much should change,. 0 Share this post Link to post