This could be an XY problem, that is, you’re trying to solve problem X, rather than the underlying problem Y. Y here being: Why do you need things to be in decimal in the first place?
I wouldn’t say it’s needed, but this is more of a fun thing for me. The only thing I’m using this is for Tupper’s Self-Referential formula, and my current approach of converting base 1>>24 to base 1e7 works instantly for 106x17 binary digits. When I load a image to that filter that’s greater than somewhere over 256x256, delays are noticeable because the underlying algorithm isn’t that great, but it could have to do with the fact that G’MIC is interpretative, and despite the JIT support in it, this is not the kind of problem it’s meant to solve (Domain-Specific). On the bright side of thing, this algorithm will work with any data type as long as one data type is one level higher than the other, and in this case, I’m using the lowest level (single and double), and the bigger data type, much faster it can be.
The problem is printing big binary to decimal. That’s not a easy problem because 10 is not a power 2. If we live in a base-hex world, this would be very easy to solve in O(n).
Also, I can’t access that as G’MIC is a language that can’t really communicate with other language as it’s not meant to share memory.