PolymerNG - Xbox One and Windows 10
#841 Posted 25 June 2016 - 03:02 PM
#842 Posted 25 June 2016 - 03:49 PM
Steveeeie, on 25 June 2016 - 02:58 PM, said:
When we going to see more from the 3d sprites and when will we start to see materials?
If you still plan to do PBR can we go with the metalness workflow as opposed to the specular workflow?
So the next goal I have is to get the performance under control, and clear up the bugs we talked about a few posts up. I want to get this part working 100% then I can move on to materials. Once materials attributes working I can move on to PBR. I want to get each milestone as bug free as possible before moving on, so PBR is still in the pipe, don't worry I'm just working up to it. The reason why I got mapster working at this stage was to help me with bug fixes and performance fixes.
This post has been edited by icecoldduke: 25 June 2016 - 04:01 PM
#843 Posted 25 June 2016 - 05:43 PM
Steveeeie, on 25 June 2016 - 03:02 PM, said:
You mean def code?
MusicallyInspired, on 25 June 2016 - 01:55 PM, said:
There is a test option in mapster which instantly boots the map into the game with the starting your position at the current position which is pretty useful. You can test stuff out, then simply close eduke and continue working in the editor. Still, there's a tiny amount of time between switching which starts to add up if you do it frequently (and is probably a lot longer when using high res assets and models). Plus it might be nice to get a point in the map, and want to go straight to that point in the editor when you close, so I can see the advantages.
#844 Posted 26 June 2016 - 01:48 AM
#845 Posted 26 June 2016 - 10:11 AM
#846 Posted 26 June 2016 - 10:19 AM
Quote
from the wiki
so how do it work
like this? r_pr_overridespecular <TILENUM> <VALUE> ??
#847 Posted 26 June 2016 - 10:32 AM
Plagman, on 26 June 2016 - 10:11 AM, said:
Drek, on 26 June 2016 - 10:19 AM, said:
so how do it work
like this? r_pr_overridespecular <TILENUM> <VALUE> ??
I agree if your having to restart the game to edit materials then that is a awful workflow, but I don't think the cvar approach is great either(since it doesn't directly save to the material), is there currently a reloaddefs/reloadmaterials command in the console?
This post has been edited by icecoldduke: 26 June 2016 - 10:34 AM
#848 Posted 26 June 2016 - 11:01 AM
icecoldduke, on 26 June 2016 - 10:32 AM, said:
Not documented on the wiki, so probably no.
Reloading defs ingame would be a huge time saver.
#849 Posted 26 June 2016 - 11:59 AM
Drek, on 26 June 2016 - 10:19 AM, said:
so how do it work
like this? r_pr_overridespecular <TILENUM> <VALUE> ??
You just set the first cvar to 1 to enable override mode, then you fine-tune power and factor through the other two cvars it says to use. You do have to alt-tab and write it out to the DEF file when you find the good values, but at least you're not rerunning the game just to test a different value.
#850 Posted 26 June 2016 - 12:29 PM
This post has been edited by icecoldduke: 26 June 2016 - 12:36 PM
#851 Posted 26 June 2016 - 12:43 PM
#852 Posted 26 June 2016 - 12:48 PM
#853 Posted 26 June 2016 - 12:51 PM
TerminX, on 26 June 2016 - 12:43 PM, said:
That's on my todo list
Tea Monster, on 26 June 2016 - 12:48 PM, said:
That's possible, I can confirm I re-exported the above normal maps and everything looks fine now.
This post has been edited by icecoldduke: 26 June 2016 - 12:57 PM
#854 Posted 26 June 2016 - 01:34 PM
#855 Posted 26 June 2016 - 01:34 PM
#856 Posted 26 June 2016 - 01:55 PM
icecoldduke, on 26 June 2016 - 12:51 PM, said:
The NVIDIA normal map plugin will have removed the alpha channel as normal maps should have no transparency.
Like others have said as the alpha channel is not needed by the normal map, content creators are utilising the spare channel to store the displacement / parallax maps so it doesn't have to require its own file and saves some space.
#857 Posted 26 June 2016 - 01:59 PM
Plagman, on 26 June 2016 - 01:34 PM, said:
You are correct I should have screen grabbed the fail case, in order for that to happen I have re-run the texture payload tool with the fail case in the data and I'm not able to do that right now. All I did was generate a normal map with the NVIDIA Photoshop normal map plugin and exported it out. I am not accidentally scaling the light contribution by anything, but that was a good idea. This is definitely a data issue, I'm just not sure if its because I'm handling something improperly, or because bad data slipped through.
It's possible the alpha channel has something to do with it(might be an error in the texture payload generation tool), I'll double check if any of the other normal maps that work have a alpha channel.
This post has been edited by icecoldduke: 26 June 2016 - 02:04 PM
#858 Posted 26 June 2016 - 02:12 PM
icecoldduke, on 26 June 2016 - 01:59 PM, said:
It's possible the alpha channel has something to do with it(might be an error in the texture payload generation tool), I'll double check if any of the other textures that work have a alpha channel.
You should only be taking RGB channels into consideration for normal maps, other engines will usually strip the alpha channel out themselves on compile.
Related Note: If you are looking to reduce file size by compiling the textures, a lot of other engines will strip out the blue channel from normal maps on compile. This is because the blue channel is always an even spread of 100% grey, you could fake this channel being there through code which other engines do.
This post has been edited by Steveeeie: 26 June 2016 - 02:13 PM
#859 Posted 26 June 2016 - 02:16 PM
Steveeeie, on 26 June 2016 - 02:12 PM, said:
Related Note: If you are looking to reduce file size by compiling the textures, a lot of other engines will strip out the blue channel from normal maps on compile. This is because the blue channel is always an even spread of 100% grey, you could fake this channel being there through code which other engines do.
When sampling from the normal map I only use the RGB channels, but when the data gets saved out to the payloads it does get DXT5 compressed(which isn't ideal for normal maps), its possible thats fucking it up.. It's a stretch but its something I can't rule out without testing.
This post has been edited by icecoldduke: 26 June 2016 - 02:19 PM
#860 Posted 26 June 2016 - 02:36 PM

#861 Posted 26 June 2016 - 02:56 PM
Some of the best performing HRP tiles have normal map textures with partial transparency.
This post has been edited by Mark.: 26 June 2016 - 02:57 PM
#862 Posted 26 June 2016 - 03:10 PM
Mark., on 26 June 2016 - 02:56 PM, said:
Some of the best performing HRP tiles have normal map textures with partial transparency.
You are failing to see how they work.
Normal maps should never be transparent, its just not how they work. The only channels that really matter are the red and green channels.
The HRP normals only appear to be transparent if viewed in photoshop or explorer because content creators are storing parallax maps in the alpha slot.
#863 Posted 26 June 2016 - 03:14 PM
Plagman, on 26 June 2016 - 02:36 PM, said:
Oh I know I just haven't added multi compression type support to the payload format yet. I guess now is the best time to do that
#864 Posted 26 June 2016 - 03:21 PM
This post has been edited by Drek: 26 June 2016 - 04:03 PM
#865 Posted 26 June 2016 - 03:52 PM
Tea Monster, on 26 June 2016 - 12:48 PM, said:
Mark., on 26 June 2016 - 01:34 PM, said:
#866 Posted 27 June 2016 - 04:40 AM
This post has been edited by icecoldduke: 27 June 2016 - 04:51 AM
#870 Posted 28 June 2016 - 05:11 AM
Drek, on 28 June 2016 - 05:07 AM, said:
You are an incredibly innovative thinker and a man that possesses wonderful intellectual ability.

Help
Duke4.net
DNF #1
Duke 3D #1









