Has there been any news about a newer version of trenchbroom? 

I've been creating some info_null entities and changing their classname to "func_illusionary", in order to make corpses, but TB seems to make the entities disappear after I deselect them. They still compile and appear in-game, but there is no way to edit them or find them again after creating them.

Any idea what might be causing this? 

I think. 

Today I decided to man up and start to use Trenchbroom. Loving it so far. Feeling proud of myself.

Sorry, continue. 

as soon as I deselect the func_illusionary entities they seem to disappear...

Check the View menu upper left side of 3d window. Make sure you somehow didn't accidentally uncheck func_liiusiionary.

Aside form that you could edit the FDG to include a func_illusionary point entity... but I have no idea how to do that and it could mess up brushes.... gah! I will add a request for this on the TB2 Github because a lot of hack need this type of functionality. Or @ericw might have a suggestion on a workaround? 

I just filed a bug: https://github.com/kduske/TrenchBroom/issues/1805

I'm not sure if TB will deal with a classname appearing twice in the fgd (point + brush), it's worth a try though. 

I've run into the same issue myself while doing the same thing (using func_illusionaries to display models) and was actually just about to post it- what's happening isn't that the entities are disappearing, it's that Trenchbroom is hiding brush entities with no brushes included. You can find the missing entities by going to the Issues tab below the map view- look for entries that say 'Entity 'func_illusionary' does not contain any brushes.' If you have a bunch of illusionaries- or a bunch of other, unrelated issues, like brushes with non-integer positions (i.e. you were doing a bunch of clipping at grid level 1), it might be difficult to find the entity you want.

Click the issue and the purple box for the func_illusionary will show up as the entity is selected and you can edit it in the Entity pane, but it'll disappear again once you click on something else in the map view. So far as I remember this bug/feature wasn't present in older versions of TB, I only saw it happen once I moved to the latest 2.0 release candidate (RC4). Word of advice though, so far as I can tell using that method to show models/corpses/etc. means they'll always be aligned to face 'forward' (0,0 by viewpos) so if you want your dead bodies to be positioned more imaginatively, it's better to use info_notnulls with the proper modelindex and model fields filled out instead. 

But modelindex isn't too reliable. If you, for instance, delete an earlier placed monster and recompile, the indices might shift on you. 

True, it's half-dozen one, six of another when it comes to placing models. Worst comes to worst, get all your enemies and entities set and working how you want before you start detailing things and use the 'edict <number>' console command to find the one you're looking for (as edictlist won't show the whole list if you have enough stuff in your map)- also helpful to prevent precache errors caused by an info_notnull trying to use a model that hasn't been loaded yet due to showing up later in the entity list. 

Nice, cheers guys.
Spuds idea of the "Issues" tab reveals the entities nicely so that should be a fine work-around 

After this latest jam it has come to mind that almost every brush in say: Skacky's or sock's maps are func_detail. greatly decreasing vis times obviously. I wonder if it would be possible to have a toggle button similar to texture lock where as you create brushes they are added as func_detail automatically. I think ideally this would be added to the map "under the hood" and not displayed in the editor by default. I also think it should be independent of groups/layers.

Or perhaps a "Seal Brush" command would be smarter? Every brush is automatically func_detail (again behind the scenes) and you select brushes and remove this attribute as needed to seal your map.

Posting here for discussion first before entering this on Github. 

https://github.com/kduske/TrenchBroom/issues/1616

I have a submission for a "Default Entity Hotkey"

One hotkey for turning a brush into the desired brush entity and another for placing the desired point entity.

:) 

...or copy behavior from JACK, where you can set what default brush is. You basically type func name and every new brush will be automatically set as specified func.

It's more universal this way, so you can use it in Q2 HL etc. 

I guess the best way to do this currently is to use layers correctly and select "all in layer" and make them func_detail all at once. I'm just not in the habit of using layers. I only use them for large areas and troublesome brushes. 

To upgrade TB2 from build 2f3c498 to RC4, can I overwrite or is it better to do a fresh install? 

NP 

could you add support for Kingpin loc 

When trying to open any map, I am get this error.

Failed to copy the file (AppData/Roaming/TrenchBroom/TrenchBroom.log to "My Documents/trenchbroom-crash.log (error 3: the system cannot find the path specified.)

I can get to this path that TB cannot find so I am confused here. Is there a way to reset where TB thinks the log file is or can tell TB to generate a new one?

Thanks for any assistance. 

It looks like TB actually crashes every time you are opening a map, and then fails to copy the log. Does it really happen with any map file? Can you post the log file? 

What does "I am get this error" mean, exactly? Is it a dialog? Can you please post a screenshot? 

Yes it is crashing every time I open a map, even a brand new one. I sent you a message on Twitter so we can discuss at a better pace.

I am get this error was a typo and I meant: "I am getting this error." Which is an error dialog window that prompts when I try to load any map, old or new. 

I don't use Twitter that much, and I need to track issues in a single place. 

It looks like you moved your user data directory to a network location and that's not supported by the underlying library that TB uses. 

Yes that would be what occurred although not my intention of course. I submitted an issue on GitHub.

Thanks!