Talk:Remote Imaging Protocol
 | This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||||||||||||||||||
| |||||||||||||||||||||||||||
RIP Bezier Curves
editThe cubic Bezier curve formula was widely published at the time, and RIPterm/RIPpaint used it. The issue is that it explicitly used double floating point math, so was at the mercy of the floating point library used. This wasn't a failing to publish, it was a mistake in using floating point for something where precision matters more than accuracy.
I don't know that I agree, because various implementations that can be found in the likes of the SWAG Pascal archive don't agree on how RIP implemented Bezier curves. The RIPscrip specification also does not provide any specifics on implementation. Do you have a source? Peter raines (talk) 06:22, 27 May 2020 (UTC)
A source for the cubic Bezier curve formula being published, or for RIPscrip using it? I think there's a wiki page already for Bezier curves, and I'm pretty sure RIPscrip specifically states it uses the cubic curve, and there's zero evidence that it doesn't. The only problems anyone had was exact pixel match, and it's pretty trivial to show that the errors were all from rounding, but that would be original research. — Preceding unsigned comment added by 68.61.118.158 (talk) 00:40, 15 March 2021 (UTC)
I changed the description of the problem to more general terms to highlight that this problem is not an opinion but a matter of fact. I have tried a lot to replicate the bezier curves pixel exact. It isn't easy. If someone knows about an open implementation that does it, please link. GrimWang (talk) 16:37, 14 February 2026 (UTC)
Supernumbers
editI was at a presentation at One BBSCON in 1994, I believe, by Telegrafix, where they mentioned that the RIP standard used "super numbers" that were base-36, which is all the letters of the alphabet and the digits 0-9. 71.234.186.80 (talk) 21:26, 14 March 2023 (UTC)