Dragofer

My own take on the beta process is that devs on one side commit to adding and improving features, but also to ensuring that the new releases are, as far as reasonably achievable, free of bugs and that the new features aren't broken on release. To balance that we have a general policy to freeze the code when the beta starts, where code changes should primarily be fixes to issues detected in the beta. Exceptions are made on a semi-regular basis for changes that go further if they can be justified. That's mainly the case when there's still enough time left in the beta to get the changes properly tested and reviewed, or the risk of breakage is low i.e. because you're adding, but not changing things. For example, I worked extensively on assets this release cycle and didn't manage to get all the work done before the beta, so I was still busy until about the middle of the beta phase. However, I prioritised my work so that asset changes were ready ahead of the beta, while asset additions (which won't break anything) were made during the beta. I also treaded particularly carefully when adding the new assets to compensate for their shorter beta testing time. I might not have been around as much lately, but I only heard of Daft Mugi's slew of patches when we were discussing whether the next beta build should be the release candidate, which gets released as-is if no significant issues are reported. At that point any changes will either prolong the beta testing phase, or carry the risk of not working as intended. The plan was to reach this point at the end of january (if this was not prominently communicated somewhere it was a mistake). Personally I definitely see the value in trying to accommodate as many of these changes as possible because they make TDM more attractive to Thief players, Daft Mugi is even standing ready to introduce them to the TTLG community, and it looks like they're fairly simple changes. However, prolonging the beta comes with its own drawbacks. One is that players and mappers are delayed in getting access to all the other new features. Another is that the beta phase is a relatively draining part of the release cycle for the devs because new projects are mostly on hold while the focus is on getting to a stable build. One of the ideas behind dev builds was to shorten the beta phase by spreading out the testing. Ultimately, we've designated Stgatilov as the project lead so it's his task to call the shots on how to balance these demands.

Its been quite a busy time, and another author has hopped in and is doing some marvellous stuff with the unique segment of this FM. Waiting for that to complete before betatesting continues.

It would be interesting to see how the FPS is when you use func_static versions of the grass instead. This way we would know how much of this drop is caused by (A) lots of transparent textures and (B) lots of animated models. If the problem is more the latter, then it could be remedied by using larger merged models, and maybe there are some low-hanging fruit for optimising the animation code. The former would require simplification of the mesh, to the detriment of the visual quality.

That doesn't work for worldspawn geometry, since it all counts as a single entity so they share spawnargs.

No, I dont have time to fix them at the moment, and I think a lot of FMs that use custom briefing GUIs are affected by numerous new warnings, too.

DR has a relatively new map merge feature. It should allow multiple people to work on the same map and periodically merge their map versions together. It probably works best if mappers work on separate regions of the map.

This does indeed sound like a case for layers.

Stgatilov recently improved GUI console warnings, so a lot of mistakes in various GUIs have been exposed. I think it's mainly unused remnants from GUIs that these GUIs were copied/inherited from, as I suppose people would've noticed if they tried to use those broken components of the GUIs.

I believe I altered map_of.tga or a GUI reference to it as it was throwing console warnings. I'm surprised any FM actually tries to use map_of.tga as it is in core since it looks to me like a highly specific placeholder.

One of them is the subtitles/readables overflow?

Do you mean because of difficulties aligning the grass with the floor? That wasnt easy for me either, at least with the large grass models.

I could whip up a custom script that does this by comparing the player origin with the bounds of grass models and then applies a lightgem modifier. No idea about a "proper" in-engine solution - maybe it'd be something similar.

I've tried using SEED to automatically clone and distribute a model across a large terrain patch, but the performance was single-digits, probably a lot worse than if the same number of models had been hand placed. It also didnt seem to respect the spawnarg to floor the models along the patch and created too many models, but that might be inexperience in handling it. Regarding the close-ups of the grass, I think Arcturus' map just had a much higher density of alpha cards. Mappers might be able to achieve that look by closely overlapping each grass model.

The old videos don't really apply because kingsal had to create new animations and rigs due to modelling app incompatibilities and a lack of source files. Not to mention he also added more variety in shape and appearance. (My previous comment was also a nudge-nudge-wink-wink kind of thing to FM authors with suitable WIPs).

Yep, would sure be cool to see authors share what it looks like in action in their WIP.

This is why you have to fix console warnings.

Kingsal has graciously converted the highly map-specific animated grass from Arcturus' gorgeous "moonlit castle by a lake with grass plains blowing in the wind" demo map into something that can easily be used by mappers, now also featuring different varieties of grass, different sizes, 2 different wind speeds and LOD stages: I'd like to run a small beta of this asset here for mappers to try it out in their FMs, so feedback on it (i.e. animation, performance impact, looks) before it gets added to core assets would be very appreciated. I won't go too much into the details because the entity & spawnarg tooltips are part of the testing. You can grab the asset here and just drop the .pk4 in the folder of your FM, no need to unpack it. Look for the animated entities in the entity menu. Once the assets are in core you can just delete the .pk4.

A legacy is something of significance that a person leaves behind when they die. These are missions grayman never finished and left behind for others to finish.

Yeah I think every mapper had a huge burst early this year but is still reeling from the exertion. I do know one of Grayman's legacy missions "Seeking Lady Leicester" is about to go up for beta around now.

My antivirus puts the .exe in quarantine mode every time the exe changes. It seems to be normal for antivirus programs to be extra cautious with any .exe - I haven't seen that kind of warning about locking a Windows file, though.

I've made a change so that m_ignorePlayer doesn't stop the current alert from counting towards the alert score. Now the stealth statistics become +1 sightings, +5 score as soon as the player is fully detected, and seem to stay there regardless of what happens afterwards. One extra stealth score bug I noticed while testing this is that if an AI is fighting you and you duck out of sight for some seconds, then duck back into sight before the AI leaves combat mode, the time where you weren't visible is added to seenTime. Probably not as high priority to fix asap (i.e. during this beta period) as full detections counting as 0 score.

If Im not mistaken, sys.wait() works by checking every frame whether time elapsed > wait time. 0.01667 seems like a risky wait duration because if your frames are i.e. 0.0160s long the engine would presumably wait 0.0320s because that's the first check where time is greater than 0.01667. So your 60 fps script becomes something like 31 fps.

For some reason when I run "testmap stealth", with stealth.map in the .pk4, the console just tells me "loading maps/stealth" rather than "loading fms/bcd/bcd_stealth.pk4/maps/stealth". Maybe just unpack and delete that .pk4 after renaming the extension to .zip, so that you get darkmod/fms/bcd/maps/stealth.map. Maybe also split up the console commands into "dmap stealth", then "map stealth". You don't need to load bcd beforehand.

@MirceaKitsuneA lot of the unexpected detections sound like they may be a consequence of this: https://bugs.thedarkmod.com/view.php?id=6091

It looks like this is a false alert (no pun intended). AIs that are about to enter combat will have the m_ignorePlayer flag enabled for the duration of their combat preparations (reason given by grayman: AIs ramped up to combat too quickly), which temporarily causes the alert to not be counted until the AI has fully entered combat. I extracted that area into its own test map, put the player start on that ledge and placed the guard just before where he turns around to see the player. I took a series of screenshots showing the resulting behaviour: 1) Level 1 alert (doesn't count for stealth score). 2) Increases to score 2. 3) Increases to score 3, with agitated searching and starts to draw sword 4) Full detection of the player, AI is entering combat. All score stats drop to 0. 5) The AI is done entering combat, score stats are restored and increased to a score of 5. 6) The AI puts away his sword and starts throwing rocks. Score is as before. 7) I duck around a corner where the AI no longer sees me. Now that the AI no longer sees me the seenTime stat is updated. I suppose you quick loaded around 4? I could also imagine that quickly ducking out of view while the AI is entering combat could let you get away with 0 score for that episode, but apart from that possibility it seems it's working as intended. The stealth score was evidently never intended to be exposed like this during the mission so in some situations it only updates after some delay. You can try the test map yourself if you're interested, just put this next to the bcd.pk4 in darkmod/fms/bcd, then enter this console command in TDM: "testmap stealth". bcd_stealth.pk4