Well those that have difficulties to understand *how* TESVAL modifies the exe, will be completely puzzled by *what* it does. The actual optimisations can't be written in C, and anyone wanting the source to cooperate on one of the project (or start his own), needs to understand the *what*.
Isn't the confusion perfect now?
So what are the adverse effects of either version? Why use one, why the other?
"stuck scripts" sound very severe to me, and should not happen with a tool that pretends just to replace obsolete far calls with inline code.
Anyway, whether SD, SKSE, or a one-hit unwrap & optimize tool, development and release should be discussed in different threads.
JOG said this much better than I could have. I am not going to force Alexander to start his own thread though.
Just finished a test playthrough with Skyboost. Did one mages guild side quest, one thieves guild main quest and 2 side quests, and one daedric quest (sanguine). No script problems or odd NPC behavior observed.
I really don't care which version goes forward, SB or TV. What matters to me is performance, stability, and support.
Exactly, and my personal preference would be that both go forward because it does offer more choices.
I agree with this - I found the original asm code much easier to read than the C conversion. If we had to write large portions of SKSE in assembly I would probably be using an external assembler to generate static libraries for those sections of code; the current patches are annoying enough to deal with.
TESVAL uses SKSE to load and provides versioning information so if problems are found with one version it's possible to disable older versions and give the user update information. SkyBoost uses an ASI loader to load and patches slightly more. AFAIK those are the only differences.
edit: was wrong about how SkyBoost loads, fixed
That was due to a typo in the original code. None of the currently distributed versions have that problem.
Thanks ianpatt. I find the explanation very helpful.
