Talk:Composite artifact colors

Color Low‑importance

	This article is supported by WikiProject Color, a project that provides a central approach to color-related subjects on Wikipedia. Help us improve articles to good and 1.0 standards; visit the wikiproject page for more details.ColorWikipedia:WikiProject ColorTemplate:WikiProject Colorcolor articles
Low	This article has been rated as Low-importance on the project's importance scale.

Computing Low‑importance

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
Low	This article has been rated as Low-importance on the project's importance scale.

Images were ruined

Someone (or something) rescaled the "King's Quest (CGA)" image and ruined it. In the RGB image, it's not possible to see the vertical lines that are part of the artifact anymore. The explanations of the article just don't make any sense without them. Schmidlin (talk) 13:18, 15 September 2017 (UTC)[reply]

I just wanted to note that I started a discussion of this here: Wikipedia_talk:Non-free_content#When_is_downsampling_the_wrong_strategy_for_dealing_with_non-free_content? to see if we can get an exception when article focuses on the micro vs macro details of the images. --Linux dr (talk) 06:58, 16 January 2018 (UTC)[reply]

Since no one replied at NFC - here was my last comment - As for the Composite artifact colors article - NF images should be as small as possible - why is it necessary to show the whole of each screen to show the artifacts - if one cropped a quarter of each screen (say top left), then four images combined would be 320x200. Ping me if you want me to arrange that as a trial Ronhjones ^(Talk) 14:42, 20 January 2018 (UTC)[reply]

Dithering and the composite blurring myth

I originally removed the following "The Mega Drive takes advantage of horizontal blurring of checkerboard dither patterns to simulate transparency effects on many games." It was restored with the reason: "Megadrive didn't output RGB exclusively so let's keep it for now; Indeed arcade machines used it without any blurring, so please add them to the article if you wish. Overall this is better discussed on the talk page." So let's talk it out.

1) My point in mentioning it being used on RGB-only systems wasn't about the MD/Genesis' RGB output. If "dithered transparency" was used with the developer expecting composite video's properties to smooth it out, why was it used so widely on systems that had no composite video output?

2) I have never read or seen any developer mention that they used dithering with the intention for it to look a certain way with composite video (barring actual artifact colours), just magister dixit statements online. In fact, the section I deleted is not cited (though this fable is so entrenched it's probably easy to find someone of some regard in retro gaming repeating it). Most dev setups back then would have had RGB output, I know Sega's Digitizer System did.

3) dithering to simulate transparency is not producing a new colour, which is the subject of this article.

I'll admit this myth being repeated over and over is a pet peeve of mine. It's a post hoc explanation that I believe emerged once people noticed how glaring dithering looks in emulators, perhaps influenced by knowing true composite artifact colours were used by some early home computers. The thing is, back in the day I played Genesis on RF, which if anything would have made the dithering blurrier. I could still tell it was dithering, not transparent. I thought it looked bad, but what else were they supposed to do? Yes it was used intentionally, but not specifically because it was going to be viewed over composite, because it was the only way to get a transparent-like effect on the MD/Genesis! And any other system before true transparency effects became a thing.

The MD/Genesis part should be deleted. CrinklyCrunk (talk) 19:53, 14 July 2022 (UTC)[reply]

Thanks for your feedback! You are right in mentioning that better references are needed. I added a few references to Megadrive/Genesis and added some text about the Amiga. I'm not 100% happy with the quality but they do exist and provide clues for better ones.

1) There's overlap between this article (composite colors) and the article on dither, and I think your question is best answered there. But I agree that we should mention dither on the intro paragraph here, perhaps as the technique used to produce (intentional) composite artifacts.

2) I added some citations, but I agree that we need better quality ones, specially from back in the day. There are a few programs for the ZX Spectrum the generate "128 colors" through blurring, and those came out on magazines from the 1980/90s. https://ia600604.us.archive.org/view_archive.php?archive=/1/items/World_of_Spectrum_June_2017_Mirror/World%20of%20Spectrum%20June%202017%20Mirror.zip&file=World%20of%20Spectrum%20June%202017%20Mirror/sinclair/books/123/49ExplosiveGamesForTheZXSpectrum.pdf see page 270. That would count as a period reference for a PAL based system. Yes, I know that many game graphics were done on PC or Amiga, and those had RGB monitors. The question is - did they had composite blurring in mind ? Agree that we need a specific reference for that.

3) Partially agree... Composite artifacts are NTSC specific, where the generated colors have no relation to the "original" ones. But simple mixing, also creates new colors (light yellow= red + white) - from a color palette POV, mixing will create new colors... The article hints at this on the PAL section, but agreed that it should be clearer.

Yes, on RF you could tell it was dithering, because blurring affected more the Chroma signal. The underlying different luminance pixels remained somewhat visible. It was the same on the Amiga or the SNES, from my experience with PAL machines.

In short, I think the article should be improved and reorganized. Most of what we wrote is a good start. Mentioning emulation and composite filters make sense. About the MD/Genesis, we might end up with something like: "The MegaDrive is usually mentioned as using dither [ref][ref], but those assumptions arise from modern emulation using filters[ref][ref] to simulate the look the system on older CRT home displays."

But again, we must find those references. Feel free to add them, and change sentences as needed. I'll try to improve the intro to address dithering in the next few days. 4throck (talk) 12:22, 15 July 2022 (UTC)[reply]

Replying to myself ;-) I've added text mentioning artifact vs blur colors. Both are diferent concepts but at least for now let's keep both in the article. Cleaned up the PAL section a bit. Trying to add references for every system and for specific titles.4throck (talk) 18:42, 20 July 2022 (UTC)[reply]

Just to add the Megadrive used mostly vertical dithering (vertical stripes), and that makes sense if you consider horizontal blur. Other systems normally used a checkerboard pattern. 4throck (talk) 22:40, 17 July 2022 (UTC)[reply]