backwardsEric

So it sounds like this is the same problem mentioned here, https://stackoverflow.com/questions/50301271/sdl2-texture-cleared-if-i-go-to-windows-lock-screen-and-come-back-or-resize-wind . I haven't found much in the way of a solution or workaround on Angband's end. There's https://discourse.libsdl.org/t/renderer-freeze-when-lock-screen-and-login-again/21654...

Given that using OpenGL should go directly to the OpenGL implementation on Windows rather than being routed through X11, the performance could be acceptable. Though, It looks like there'd have...

It does look like o_obj_known_damage() does include the additive term (see lines 1046, 1077, 1082, and 1113 of obj-info.c in the current master branch). The additive term is in OAngband:...

Though from what I see, the scaling of the added term in the object information calculations is not correct thus https://github.com/angband/angband/pull/5703 .

I believe it is. What I see is that the actual calculations and those for the object information agree, and that they match what's in OAngband. tomm may want to...

As an alternative, https://github.com/angband/angband/pull/5536 addresses 1 and 2 . 3 may be less of an issue when and if https://github.com/angband/angband/issues/5499 is resolved. I can't speak for the rationale for main-ibm.c...

For emulating code page 437 in the SDL 2 front end, if you have access to a fixed-width TTF font with Unicode indexing that has support for the additional symbols...

As you found, a font with code page 437 indexing would work with font-ibm.prf. For a font with Unicode indexing, I was recommending replacing the values in font-ibm.prf with the...

The hurdle I see for using a runic alphabet is knowing whether the font the front-end uses has the necessary glyphs (there could be hook functions for the front ends...

Leaving aside the concerns about the font details, you might describe how your proposal would look like (with a fully unidentified, partially unidentified, fully identified item what would be seen...