[[ 2.10 ] New Frob Shader]

It’s a little crazy

We talk about broken missions as a critical bug but still no mission name and getviepos coordinates

[[ 2.10 ] New Frob Shader]

1 hour ago, geegee said:

put a "secret" switch under a desk.

In DR? Didn't you say the existing missions break with frob outline? I'm interested in that

[[ 2.10 ] New Frob Shader]

16 hours ago, geegee said:

It isn't a bug, it's a feature.  With the frob highlight on as currently set default in the 2.10 dev builds, you can see it thru' desks, walls, etc.  So if this is implemented in the final build I'll have to go through my wannabe FM taking this new feature into account, relocating switches, loot, readables etc., or covering the entities with target_set_frobable, or whatever else that'll work to make it play right.  The game is a THIEF homage, and there's stuff that's supposed to be hidden from thieves, not telegraphed through walls etc.  That's the whole point of the game.

I never made TDM - I'm late to the party - so I'll go with whatever and do so thankfully because I appreciate the work, the result, and it's great fun working with DR then checking it out in a great game engine.  Going thru' my wannabe FM to redo things to account for this feature won't be much problem, the time spent a drop in the bucket compared to the time already spent learning tricks.  

But what about all the FM authors who're no longer here, who's going to fix up their FMs to put back the mystery.

What are the steps for me to reproduce?

[[ 2.10 ] New Frob Shader]

Is there a simple way to reproduce the whatever bug some people think the frob highlight has?

[Dev builds (2.10)]

18 hours ago, HMart said:

I've seen something similar, in my case all sliders in the menu lack the visual movable slide and the value label, also some missions lack the "start mission" string, I assume all this is because of the changes being done to the menu system by stgatilov and will be solved sooner than later. 

Image_preload 0?

[[ 2.10 ] New Frob Shader]

Pity, I grew fond of the new frob visuals

[Convert normal maps to RGTC?]

23 minutes ago, stgatilov said:

The new dev build is available, so now everyone can test the new approach.

To be honest, I forgot to add a cvar to enable/disable the new RGTC code, so on the new dev build it is always on.
I'm afraid you'll have to install old/new versions side-by-side if you want to make a comparison.

image_useNormalCompression does that?

You mean there's no more driver compression, only ours or none

[Why so long?]

Quote

I believe you can look in tdm_installer logs or game console (at least on the latest dev build) to see which URL is used.

I can't see url's in the game console

[ARM64 Support?]

Use a .cfg cvar for running in a window

Then it's debugger time for you

[ARM64 Support?]

@wootyOh dear, I just realized. This is probably the same bug I hit earlier this year

It does not like two 'darkmod' in the file path. Try renaming the parent folder to something like 'tdm'

[ARM64 Support?]

22 minutes ago, wooty said:

@stgatilov I followed the readme.md conan instructions and was able to build all dependencies and compile "thedarkmod.custom". I navigated back to my base directory and into the newly populated darkmod folder. Running TDM gave me an error about "default.cfg" not being found, so I dropped the one from the source directory into the root of the TDM folder. Running again gives me a "Can't open journal files" and "Sys_Error: _default material not found" as a result.

I feel I am just missing some files that need to be manually placed, or something of that nature.

 

Do you have the game itself installed? It needs all the assets naturally

Here's a partial screenshot of the file content

[ARM64 Support?]

1 hour ago, wooty said:

@duzenko Heres some on the fly bogompis like tests, need to run phoronix test suite later.

sysbench --test=cpu --cpu-max-prime=20000 --threads=1 run

CPU speed:     events per second:   780.44
 General statistics:     total time:   10.0002s     total number of events:   7807
 Latency (ms):     min:  1.28     avg:  1.28     max:  1.36     95th percentile:  1.30     sum:  9997.18
 Threads fairness:     events (avg/stddev):   7807.0000/0.00     execution time (avg/stddev):   9.9972/0.00

sysbench --test=cpu --cpu-max-prime=20000 --threads=16 run

CPU speed:     events per second:   12331.51
 General statistics:     total time:   10.0013s     total number of events:   123371
 Latency (ms):     min:   1.27     avg:   1.30     max:   2.60     95th percentile:   1.44     sum:   159944.14
 Threads fairness:     events (avg/stddev):   7710.6875/241.13     execution time (avg/stddev):   9.9965/0.00

glmark2 Score: 4500+-

Respectfully, that's a lot of numbers but no flops

Quote

I've been working on this for months and I am starting the benchmark phase.

I'd benchmark that on day 1 and that might as well have saved those months, LOL

Maybe just run any web benchmark?

[ARM64 Support?]

42 minutes ago, wooty said:

HoneyComb LX2K

What's your estimate on single-thread FLOPS there? I'm kinda skeptical on its speed, considering it's 16 cores and passively cooled

EDIT I can see on some pictures it has a tiny cooler, but still looking like a 20-30W max. Something to drive a file server.

[ARM64 Support?]

I think we're missing the static libs for ARM @stgatilov

Your contribution to fix this will be very much welcome however I'm curious what's the make and model of your computer? Is this some kind of a server?

[Convert normal maps to RGTC?]

17 hours ago, HMart said:

oh yes a good internal ssd or even a nvem/M.2 ssd can be fast even with preload on, but I have the game in a external USB 3.0 2.5 HDD, so preload may impact me more I assume. 

Preload is definitely preferred for HDDs

[Convert normal maps to RGTC?]

On 8/20/2021 at 2:13 AM, HMart said:

I bet you guys already know this but if you don't, just for kicks I set image_preload 0 and image_useCompression 0 and the missions I tried load way quicker! Is obvious why but I underestimated how impactful this options were!

And I didn't felt any performance loss from it. I hope this is true for all! 

It seems I will play with those off going forward.  

Btw I also found that using preload 0 and compression ON, causes the game to start with no textures for a few micro seconds, similar to some UE3 games! TDM is now among the big league engines! ;D kd

Is not a big problem and the same happens with compression off but for a lower time, unfortunately, the game has visible pauses, when opening doors and the game as to load and compress the textures on the fly. 

 

edit: I didn't played the latest version with stgatilov new RGTC compressor in SIMD, so things could be different on it. 

With uncompressed assets on ssd it's quite fast but then again full preload is not too slow either

[Dark mod 2.07 shadows issue]

2 minutes ago, nbohr1more said:

Try r_useGLSL 0

Something about uniform transforms maybe too?

[Dark mod 2.07 shadows issue]

12 minutes ago, OrbWeaver said:

Intel HD 3000 officially supports DirectX 10.1 and OpenGL 3.1, according to Intel. This should be enough, surely? I thought we required OpenGL 3.0.

3.1 for 2.08
3.3 for 2.09
Afair

But in reality i did a quick test with 2.08 and intel 2000/3000. Regret the drivers for that chip are utter garbage.

[Dark mod 2.07 shadows issue]

Sorry, no support for Intel bugs here

Your options

• go further back to 2.06
• search the beta thread for 2.07 for new rendering cvars
• make sure you have the latest drivers for HD 3000 on Intel website (not the laptop producer's)
• consider trading up on your hardware

Sincere condolences on your GPU

[Convert normal maps to RGTC?]

1 hour ago, MirceaKitsune said:

Will keep it in mind. IIRC though that means the game will stutter whenever an object with a previously unloaded texture comes into view, so I'd likely get slowdowns during the game instead.

The current svn performs quite well actually

[Convert normal maps to RGTC?]

2 hours ago, MirceaKitsune said:

This is really great! Any noticeable difference in loading time too? That's one of the main aspects I was interested in, TDM still has some really long load times compared to many other engines.

Try image_preload 0

[How does the "ignore user" function work?]

Shit, Kurshok, I'm almost sorry for you now

You clearly can't control yourself, it's now matter of time when you commit that suicide-by-moderator

It's like watching a propane can in fire

I wish I could say something to trigger you and end the pain

[Convert normal maps to RGTC?]

39 minutes ago, stgatilov said:

Yes, it makes sense.
But make sure to disable it woth in HandleCompression and in SelectInternalFormat.

Huh, it seems I was wrong about that

Completed: At revision: 9563 

Quote

I'd say we can probably refactor SelectInternalFormat into two parts: the one which is O(1) and return format without any analysis, and the one which can additionally check the image data for final internal format. In this case I'd call the first function in HandleCompression instead of writing the same conditions again.

Also a good idea

[Hidden Hands: Initiation (27/12/2018)]

This mission is regretfully under-optimised

E.g. this window lacks a visportal

[Convert normal maps to RGTC?]

7 minutes ago, OrbWeaver said:

If I understand correctly, the advantage of doing RGTC compression in software is performance, because it can be done in parallel rather than image-by-image on the GPU?

If so, I guess any testing should focus on performance (particularly the time to load a new mission from scratch), rather than GPU memory usage which should be the same whether RGTC compression is done in software or by OpenGL.

With dynamic image loading driver-side compression is just too slow and results in stuttering

Even if not used widely ATM it's already technologically superior