Old Dip Tracker

> I will rework this tomorrow. Perhaps I can just backport the unit test from here into there, and unblock #3313 that way? Then you can focus on the RNG...

I benched this on both an x86_64 Intel I7 and an arm64 Apple M1. Results: ### On i7-970 #### Current FastRandomContext (at 8d4e0bd085) ```csv #Benchmark,count,min,max,average,min_cycles,max_cycles,average_cycles FastRandom_1bit,640,0.001588508486748,0.001606859266758,0.001596468687057,5144478,5203874,5170219 FastRandom_32bit,640,0.001587942242622,0.001599781215191,0.001593051850796,5142448,5180947,5159149 ``` #### New...

I picked the following on top for a ~10% performance increase on the 32-bit FastRandomContext call: - 2bc2b8b49ad87376d98591b14403dc3b46e3166b (to get easier pick) - 2dfe2751713c814aea53b5a7563eb74ad1baea00 (to get another bench baseline to...

Thanks for this work, I think we could merge this out-of-phase to please the eye and avoid confusion - that's fine. As for the license files, these will need to...

> Would this be a rebase what it is now? Yes because all the CI changes are squashed into a UI commit. No worries, we'll get that done. For now...

Answering the easy one first: > Is there a plan to switch 1.21-dev to a different "Continuous Integration" process? I see 1.14.7-dev is doing it a bit differently but haven't...

> The MacOS task is now compiling successfully, but is failing due to failed PSBT tests. Hmm no, it is failing due to berkdb 5.3 runtime issues - I suspect...

I just saw that Bitcoin Core fixed this by no longer supporting "legacy" bdb based wallets in the macOS arm build on 24.1 / 25.0 - it looks like we're...

Hmm it's only for macOS native builds - cross-compilation with bdb still works, here too. So it's not depreciation of the feature, it is specific to the platform. I've personally...

Note: I think that we can port #3308 to 1.21 to unblock the bdb issues. Will work on that as soon as we've seen some successful testing on the 1.14...