Beta Testing 2.06

[duzenko]

Guys, how does the reloadimages command work for you? Yesterday it made the soft shadows disappear, now I got the whole screen go black. x64 on Intel CPU, nvidia GPU, seta r_fboResolution "1",

seta r_fboSharedDepth "0", seta r_fboSharedColor "0", seta r_useGLSL "1" seta r_useFBO "1".

 

Also, while it's not super important, but r_showShadows 1 and 2 don't work either, making soft shadows disappear as well.

The FBO textures might be getting reset with reloadimages. Looks like a bug.

r_showShadows always make shadows disappear, but with soft shadows on they also don't produce the shadow volume geometry because soft shadows are rendered off screen. Looks like a bug too.

[peter_spy]

I just reported this in bugtracker.

[Destined]

Yes

Great! I have added a small comment on that to the Wiki-page for sleeping AI.

[grayman]

Great! I have added a small comment on that to the Wiki-page for sleeping AI.

 

Thanks.

[grayman]

I would just upload the .pk4 to the darkmod ftp (which runs on the same server as the wiki and other darkmod resources) and link it on the wiki page.

 

Getting a "cannot connect" error atm.

 

I'll try again later.

[grayman]

Flashbombs in 2.06

 

Per this issue, can folks report on their experience with flashbombs?

 

Any problems?

 

Thanks.

[Bikerdude]

Getting a "cannot connect" error atm.

I beleive I posted the ftp detail for that ftp sever in a members only thread, but dont have access to that anymore. So I can haver a look for it when I get home, otherwise drop Taaaki an email.

[grayman]

I beleive I posted the ftp detail for that ftp sever in a members only thread, but dont have access to that anymore. So I can haver a look for it when I get home, otherwise drop Taaaki an email.

 

Yes, I have the details.

 

I'll deal with this after 2.06 is gone. I don't have time to tinker with this now. It's good enough as is. It's not like Dropbox is suddenly going to fail.

[Bikerdude]

• I'll deal with this after 2.06 is gone.
• I don't have time to tinker with this now. It's good enough as is. It's not like Dropbox is suddenly going to fail.

• fair enough
• Is it on your personal DB account, I only ask as there a number of dead links on the forum/forum where people have used the aformentioned instead of the darkmod/official mirrors.

We should ask Taaaki is there is any reason why team members dont or cant have the ability to upload files other than images to the wiki and for those files to be bigger than 500k.

[stgatilov]

How is dmap working in the beta?

 

Have you met any maps on which dmap crashes?

As of now, Bikerdude's WIP seems to be the only map with serious dmap issues (as noted in 4661).

 

[peter_spy]

Didn't have any problems with dmap on x64, but I only have a small WIP with lots of custom meshes and one test AI walking around.

[grayman]

How is dmap working in the beta?

 

Have you met any maps on which dmap crashes?

As of now, Bikerdude's WIP seems to be the only map with serious dmap issues (as noted in 4661).

 

 

I just dmapped Behind Closed Doors using 64-bit dmap. BCD's map file is 46MB, and it took 5:14 on an SSD. No crash.

 

BCD is one of the largest maps.

[grayman]

Dmapped Politics (Nhat3).

 

35MB.

 

5:12, no crash.

[grayman]

Dmapped Lords & Legacy.

 

35MB

 

2:30, no crash.

[Bikerdude]

I will give my city map a run under 2.06 this evening.

[AluminumHaste]

Physics completely broken still at high FPS.

When running a very simple test map with uncapped FPS, I get over 666 fps.

 

When the mission starts, I'll jump and get stuck in the air. I can look around and randomly can drop back down.

Also, while running the player doesn't make any footstep sounds.

 

Limiting FPS via rivatuner fixes the issue, as well as enabling VSync

[teh_saccade]

http://forums.thedarkmod.com/topic/18708-bugs-with-update-to-205-strange-physics/?p=415315

Explanation as to why - good luck fixing the physics over 60fps.

 

//assume refresh rate of monitor is 60/59hz therefore fps w/vsync will be same.

Edited December 14, 2017 by teh_saccade

[peter_spy]

Yup, it's not like the physics is broken, it was never designed to work with such high fps. Other engines have this too. Everything is designed for 60 Hz max.

[Springheel]

Physics completely broken still at high FPS.

When running a very simple test map with uncapped FPS, I get over 666 fps.

 

I thought we were going to cap it by default and let users uncap it if they think it provides them some benefit, taking on any risk that goes along with it.

[grayman]

If there are specific bugs that beta testers have reported, please file bugtracker issues (or have someone do it for you).

 

I don't want everything on the known-bugs list dealt with, only to discover there's another 10 things that people have neglected to file issues about.

[grayman]

I thought we were going to cap it by default and let users uncap it if they think it provides them some benefit, taking on any risk that goes along with it.

 

So the suggestion is--and I agree--that FPS is capped at 60 by default (where we know physics will at least work the way it always has), and if the player wants to uncap it, a warning gets put in the Release Notes that physics can become unreliable the higher the fps goes.

 

At this point, I don't think anyone should be spending time hunting down all the wonky scenarios caused by high fps. If someone wants to take that on in 2.07, that would be great, after which we can get rid of the warning.

 

Agreeeeeeed???

[cabalistic]

Afaik, uncapped fps sets com_fixedTic, correct? From my understanding, com_fixedTic does indeed result in one game tic run per each frame, which is undesirable. When I originally wrote the multi-threading patch, I did the fps uncapping in a different way, ensuring that game tics run at a constant 60 Hz and only rendering is uncapped. This is now also the behaviour you get when you enable both com_smp and com_fixedTic, and with both enabled I have not had any trouble at higher fps. Without com_smp, though, com_fixedTic is, imho, broken.

Edited December 14, 2017 by cabalistic

[teh_saccade]

splitting the game physics and drawing/rendering is a good idea:

http://kirbysayshi.com/2013/09/24/interpolated-physics-rendering.html

although probably more hassle than simply applying a choke.

It's also worth noting that while the human eye is capable of seeing up to 10ms flash of light / 100ms avg. - humans see fastest in peripheral vision for motion.
Anything lighted/motion over 60hz (~30fps) in peripheral vision such as a monitor in a computer game can cause disorientation and create state of unreality/motion sickness - another reason we often would cap console games at 30fps draw time for consoles and reach engine cap of 60fps for hardware that could manage it [ie, not on interlaced tv and with the juice to process]. Also why VR headsets are shunned by Disneyworld and Augmented Reality is becoming a more viable alternative in games/interactive entertainment development.

It has been noted that some people might not be able to experience "FPS" at "gaming" speeds even in real life at the speed of a monitor: https://en.wikipedia.org/wiki/Wagon-wheel_effect

Over 120hz (60fps) most humans are unable to tell the difference - it takes training to respond this quickly and notice - some of the best fighter pilots have been measured as seeing a recognisable object with a light flash of 1/225 (225fps).
It is simply a waste of processing and drawing time when the average human can generally see at only 45fps.
Roughly 25% of people can see faster than this.

However:

The fps of reality is 1/tp - inverse of Planck time (10^44fps).

Human brain processing has been shown to operate at around 60 fps.

Considering humans created the physics that go into these engines, and our wetware/neural processors are "capped" at around 60fps - I am unsure if there will be any time soon some physics model required that will render faster than our own brains, or if it would be necessary - at least until we all evolve and get over that something such as increasing the fluidity of motion in a game that fewer and fewer people will notice is less important than coming up with more creative level design within the limits of what we already have.

https://xcorr.net/2011/11/20/whats-the-maximal-frame-rate-humans-can-perceive/

NB diagram for "score vs frame rate and how the curve falls - much extra effort for such little gain.


(I'll put picture in case cba to read science article)

There is a reason.

But - hey - what do I know..?

Edited December 15, 2017 by teh_saccade

[duzenko]

About 4665 (framerate and mouse sensitivity)

Does the unlocked fps option make any difference to you? I am asking whoever can notice this problem because I can't.

[grayman]

The next package build is available for testing.

 

7342 – Remove debugging statement (grayman)

7343 - Better fix for #4609. (Beleaguered Fence inmate ragdolls alerting guards at mission start.) (grayman)

7345 – 4595: Install VC redistributable pack during cleanup phase, if it is not installed yet. (stgatilov)

 

15117 - Fix some issues with the new church modules (Springheel)

15118 – 4529: Update missions to match repacking (nbohr1more)

15119 - Add MSVC redistributable packs for user-side install (included into manifest too). (stgatilov)

15120 – Include redistributables in TDM package (grayman)