STiFU

Maybe some moderator should move these posts about videoscreens into a separate discussion, as people might find the information useful in the future and will find it better this way.

Just FYI, Ubisoft now also gives away games over the course of the month. This list is not yet complete. More are to come: https://free.ubisoft.com/

Custom tdm_player_thief.def or tdm_sfx_movement.sndshd in you mission?

Both are already in place. Did you actually play with sound enabled? You can find more details here: https://bugs.thedarkmod.com/view.php?id=4992

I have addressed both issues. The changes should be in the next beta release. @kingsal, @lowenz, I actually had a quick look into mobilizing the player during the shouldering animation, but apparently it is not easily possible because the player actually clips through the floor and allowing regular movement during that leads to the player falling through the floor. Sorry guys, this cannot be changed without creating a completely new type of viewport animation, which is way besond the scope of this beta-phase. However, I made sure that you can actually disable the animation via pm_shoulderAnim_msecs = 0.

I just realized that immobilizing the player is actually not OK while the player is in mid-air. Sure, this is a very rare corner case ("grabbing an AI while in the air"), but playing out the animation and then dropping on the floor is really weird. It also still feels kind of weird, that the player view cannot be adjusted while the animation plays out. I will have to revisit the code.

Oh wow, yes, that's a huge gain in loading time. My loading time for Saint Lucia in 2.08 Beta 1 decreased from 0:28 min to 0:17 min when run from my regular HDD. Wow! Just out of curiousity, I also copied the game to my lightning fast m2-SSD and surprisingly, there was no difference in loading time: It was 0:17 min again! Tried it numerous times in both cases, to verify. I conclude, that hdd access are not the bottleneck in regards to loading time.

Yeah, I know. The way I understand it, you'd need Mipmaps and stream the higher resolution mipmaps later. I just figured you might be interested in this.

Player's movement is locked for example while mantling and just like in that case, it makes sense to immobilize the player here: Lifting a 140 kg guard is not an easy task, so you cannot move while doing it. And as I said above, this is also meant to balance shouldering vs dragging because shouldering is so overpowered that nobody ever would use dragging (which has been discussed here and in the linked thread). With this brief immobilization, there is a small disadvantage when using shouldering, so the player might consider rather dragging a body than shouldering it in some situations. Anyway, if you don't like the feature, you can just disable it via the pm_shoulderAnim cvars, as I already said above.

There was an internal discussion (linked for reference for other team members) about this. It was decided to immobilize the player during the viewport animation in order get a better balancing between shouldering and dragging of bodies, issue #1932. I will therefore close issue #5202. If you don't like it, you should be able to disable the shouldering animation by settings the times to 0 via pm_shoulderAnim* cvars.

A mostly technical review about Doom Eternal and Id Tech 7 and why that engine is so amazing. @stgatilov @duzenko there is also a lot of detail about their texture and LOD streaming and why it is so crazy fast.

Mappers can disable ambient locally, if they want. However, I am of the same oppinion as New Horizon. That being said, you can disable ambient via r_skipAmbient 2. Please note that it also disables SSAO, as that effect only applies to ambient.

How can you tell? Did you profile map loading? If so, I'd be very interested to see the results!! Yes, this is definitely a 2.09 task.

Do you have a converter for RGTC @duzenko? I'd like to conduct an experiment so see how loading times are affected. I assume their is no code support needed to load RGTC normals at this point?

Not as much as you'd expect. A while back, Aluminium Haste benchmarked loading times using hdd, ssd and RAM-Disk. Loading time went down from 1:45 min to 1:28 min at most. I also did a quick benchmark just now in order to check what the effect of the zip/pk4 compression is. I used 2.08 Beta 1 and compared it to SVN (where everything is unzipped), while using the same executable. I loaded Saint Lucia for this test. Loadtime zipped: 0:28 min Loadtime unzipped: 0:23 min Again, not much to gain here. I wonder how loadtimes would be affected if all normals were converted to RGTC?