openmw.org

A split uniform logarithmic ratio of 1 would probably be a much better default for people using distant terrain.

Discussion about that keeps swinging back and forth. Quite a while ago, I was under the impression that 0.5 was best with the default viewing distance and 1.0 was better with an extended distance, but when shadows were merged, people were telling me that actually, 0.95-ish was best for the default viewing distance. When I started mentioning that, people then started telling me that actually, 0.5 was the only value that didn't absolutely ruin distant shadows.

Well, sure. But at the very least, distant shadows being messed up is a lot more normal and "easy to expect" than nearby shadows being messed up. If someone looks at a plant and the shadows go from crisp to a smudged blobby mess, they're going to notice and get pulled out of the experience. That won't happen if they look at a hill in the distance and a shadow on it is flickering in a weird way; they won't even consider blaming the engine in that case.

Now that Zini is back in some capacity and able call the shot on getting the release process started, would now be a good time to do so?

As far as I can tell there should definitely be enough warranting it, with a good number of commits since last release, and shadows being merged into master as a major feature for the release.

Can we wait, like seriously, a week? So I can get through finals?

Too early for 0.46 IMO:
1. There is a lot of pending pull requests
2. Nightlies still do not work with Bullet, compiled with a double precision support
3. Shadows still need some polishing and break rendering on some machines even when they are disabled.

break rendering on some machines even when they are disabled

The only case I know of where that was a thing has been resolved a couple of days ago.

AnyOldName3 wrote: ↑06 May 2019, 13:48 The only case I know of where that was a thing has been resolved a couple of days ago.

Well, this bug is still opened.
Also there is a bug with GLES renderer when sky is turning red, even if shadows are disabled. It appeared once shadows were merged.

Okay. As they're platform-specific, I reckon it's more likely they're due to driver or GL4ES bugs, even if we do eventually end up providing a workaround.

Any news on .46? I see 180 completed issues on GITLab but it is quite quiet around here of late...

It'll be done when it's done. Right now, you're probably best off using nightlies.