vladislavbelov

It happens on start? The assertion was added few weeks ago. It might be that the bug was silent and now it's detected by the assertion.

Could post your graphics options (advanced tab)? Also do you have steps to reproduce?

Could you update the game to the last SVN?

Thanks! Is it possible for you to run and play the game with GDB (it allows to resolve the stack with related source lines)?

Yep, thanks! Could you tell which map it was and which steps did you have (like have you just started the game and the map or it was at the end of the match)?

Do you have logs and crashdump from that run? The error might happen on out of memory.

In my opinion the new cursor looks a bit strange, more cartoonish than we had before. (first mentioned here https://code.wildfiregames.com/rP26448)

It has nothing about architecture of the engine. The answer is: We don't build 64bit game for Windows yet, like we do for other platforms.

Fixed in https://code.wildfiregames.com/rP26302.

Thank you for the report, fixed in https://code.wildfiregames.com/rP26302 (it was a side effect of an implicit relation). Could you reproduce the slowing at the end? If yes could you enable profiler (usually press F11) when it's noticeably slow and make a screenshot and attach here?

It's a known thing, thanks. My next patch about framebuffers should fix that.

I don't have an implemented solution yet.

Because somethings else is broken I might have a good solution, but not in priority at the moment.

It's incorrect to replace 2.0 by 1.5 because it should convert [0, 1] range to [-1, 1] range, not [-0.75, 0.75].

By the way if the exit button is not so important, is it VIM? I suppose it's not so stretched. It's aligned to the left (right if RTL) with a fixed padding. Which a bit different type of layout in UX. Blender: Office: Browser: The problem with your approach is that you need to skip by eyes a lot of empty spaces to find the each next visual anchor (button text). macOS users I agree that the grouping is a good thing. Though I believe that it needs to be grouped in a more natural way, without a lot of empty space.

Seems like designed for mobiles, not PC displays. Eyes have to walk for a long distance to find important buttons. Which leads to Brownian motion of eyes.

Fixed in rP26229.

What's the map?

Could you attach logs and system_info?

No promises, but I'm working on that

We have some old but relevant (because of our audience and we don't have LODs) rules about polycount: https://trac.wildfiregames.com/wiki/ArtPolyCountGuidelines, but sadly some new models broke them in a bad manner

Have you ran update-workspaces.bat?

Are there some numbers? (Just wondering, I've met various approaches) I suppose it may not tell about Git, but products like Plastic/Perforce. Because they are commercial and companies might pay for their integration.

It means you have never tried to make a signal decoder or a simple calculator with red stone