3889 replies to this topic

[toxie]

Alright while I experiment with the HDR environments I figured I would share some samples with people, maybe it'll help others...

 

from this link, I decided to try a few on my Sopranos build to see what happens. For this, all my GI lighting is set to white/full white (not how I had it set in the previous preview but it looks better with white lighting using these environments

http://www.smartibl....bl/archive.html

 

Am I doing this right?

 

This looks like you're doing it right, yes (as long as you are NOT using the .hdr files in these packages that have their filename end with '_env', as these are the blurred variants for the diffuse approximations of some other rendering engines, VP does compute this itself).

 

So please use the original/high res environments, and then please downscale them using a tool like this: http://qtpfsgui.sour...net/?page_id=10

(open the file, then 'edit->resize' and enter something like 512x256 or 1024x512 at maximum, then save the file, as otherwise you will get a lot of aliasing on specular/glossy surfaces!

do NOT use the tonemapping stuff in there, which is the main usecase of this program usually. so just use resize! also always save it as either .hdr or .exr again!)

 

Then the old scheme should still apply:

1) test each of the two lighting components separately, so set light emission scale to 0, then test how it looks like (this will show -only- the environment emission)

2) then restore the old value for the light, and set the environment emission scale to 0, and test again how it looks (this will show -only- the lights emission)

 

Make sure that both look reasonable, as the lights will give you a nice basic lighting on the table and also some nice specular highlights, while the environment will make the

reflections look good and add some nice variety to the lighting in general.

heres a couple disney HDR's I took last year...

 

these look spectacular, but unfortunately not useful for VP..  :/

 

what you need is something like this:

https://www.google.d...nments&tbm=isch

 

and then make sure that the pictures are in a 'real' HDR format like .hdr or .exr, then please downscale them using the tool i mentioned above, to keep them as a high dynamic range image.

[toxie]

here i also found a nice beginner tutorial what all this HDR, environment, IBL madness is supposed to mean:  http://www.trinisica...ode=1&issue=002

 

if you're all scared off by this stuff, then simply ignore it for now and wait for some of the hardcore artists in here to demonstrate it for you (i hope )..

[hauntfreaks]

so they need to be not only hdr but globe... because these are multi-exposure HDR photos... but have been flattened... I test one in VPX... I see how its being used on shinny objects like chrome part... pretty cool..

[Dozer316]

A very early CV with Milky Way HDR map and relatively high light emission scale and Bloom settings.

 

[fuzzel]

W.O.W.

[bolt]

Looks fantastic Dozer.

[nicco84]

Oh ohhhhhh

[toxie]

so they need to be not only hdr but globe... because these are multi-exposure HDR photos... but have been flattened... I test one in VPX... I see how its being used on shinny objects like chrome part... pretty cool..

 

of course you can just use arbitrary HDR pictures and see how it looks, but if at some point VP would support dynamic camera(s) or if you have animated objects on your table, then it will look strange in motion.

 

you could place such "flat" HDRs on stuff like flashers though, or the backglass (NOT on other stuff though, and i also included a check for this that should warn you in case you try).

the difference there is not that striking though, but this might become more and more important as soon as the first consumer-affordable HDR and WCG capable TVs and monitors will be available and we (hopefully) can add support for these into VP (might be a bit painful due to DX9, but could be possible i think).

then you can actually really -see- the much richer contrast and brightness levels that HDR capable rendering allows for. currently you can also see the difference, but due to the necessary tonemapping step (=bringing the brightness back to the range what your monitor can display) its not -that- obvious.

[Outhere]

A very early CV with Milky Way HDR map and relatively high light emission scale and Bloom settings.

 

"""*******NICE*******""""""

[jimmyfingers]

Toxie, can you please clarify how we go about just making things look like the way the did before (how do we get the old environment image back or do we set it to blank).  These new options seem like they could be cool but to be honest I was happy with the way it looked and how I had personally tweaked the lighting before these changes so would like to be able to continue in that realm while exploring these others.

 

Also, you mention playfield reflection on the ball is now reenabled / working but it is not showing anything on the ball from the playfield texture bound to the table no matter what view (FS / desktop) or settings I try. I've used very obvious patterns for the PF as a test and nothing is displayed on the ball. Are you only reflecting these new static environment pictures in the ball?  I thought the idea was as worded for the playfield to be reflected in the ball but it's just not happening even on the default / new table (take a look with some different PF textures and views and the ball appearance doesn't do anything even with reflection set to 255.  Also, I see in the latest build that you have even removed the options further and now the ball option for reflection is simply an enabled flag (-1, 0, 1) and all settings are controlled still by a single strength for reflection in a different tab now.  Do you not think that the reflection of the playfield on a steel ball should be able to be set differently then the reflection of the ball and table objects on a clear coat PF surface (but even right now it doesn't appear that PF reflection on ball is actually showing anything)?  

 

I've posted some other items in our initial testing PM so would really appreciate if you could respond to some of these things.  Also, performance has dropped by about 5-10% with all other things being held equal on my testing rig since these builds even when reflection is disabled.

A very early CV with Milky Way HDR map and relatively high light emission scale and Bloom settings.

 

Glad to see Dozer that you're working on a VPX CV.  The new features will give us more to tinker with and it's very early in checking things out, however, one thing I don't get about this view and the amount of general thumbs up about it is the random shiny / coloured spots on the wood at the top right and left side especially?  That part doesn't seem realistic to me for these blotches to show on wood?  Also the lower right sling shot seems overly saturated with a purple hue for the lights off view, did you set your light emission colour to a more purple one intentionally or is that all from the environment image you used?  The steel ball guides / sides at the top look quite nice though.

Edited by jimmyfingers, 12 October 2015 - 02:32 PM.

[Dozer316]

Hello,

 

This was an example of an over saturated HDR map to give an idea of how things could look if you were going for that type of effect.

 

I like this particular map because on higher wire forms the light illuminates the ramp from below like it would in a dark room with GI bouncing off the underside of the ramps.

The random over saturated spots are are indeed unrealistic but the table is about 5% complete too so those wood decals won't appear in the final build.

 

The reason the right slingshot is purple is because none of the parts that make it up have a material assigned to them.  This has nothing to do with the environment map.

 

Toxie, can you please clarify how we go about just making things look like the way the did before (how do we get the old environment image back or do we set it to blank).  These new options seem like they could be cool but to be honest I was happy with the way it looked and how I had personally tweaked the lighting before these changes so would like to be able to continue in that realm while exploring these others.

 

Also, you mention playfield reflection on the ball is now reenabled / working but it is not showing anything on the ball from the playfield texture bound to the table no matter what view (FS / desktop) or settings I try. I've used very obvious patterns for the PF as a test and nothing is displayed on the ball. Are you only reflecting these new static environment pictures in the ball?  I thought the idea was as worded for the playfield to be reflected in the ball but it's just not happening even on the default / new table (take a look with some different PF textures and views and the ball appearance doesn't do anything even with reflection set to 255.  Also, I see in the latest build that you have even removed the options further and now the ball option for reflection is simply an enabled flag (-1, 0, 1) and all settings are controlled still by a single strength for reflection in a different tab now.  Do you not think that the reflection of the playfield on a steel ball should be able to be set differently then the reflection of the ball and table objects on a clear coat PF surface (but even right now it doesn't appear that PF reflection on ball is actually showing anything)?  

 

I've posted some other items in our initial testing PM so would really appreciate if you could respond to some of these things.  Also, performance has dropped by about 5-10% with all other things being held equal on my testing rig since these builds even when reflection is disabled.

A very early CV with Milky Way HDR map and relatively high light emission scale and Bloom settings.

 

Glad to see Dozer that you're working on a VPX CV.  The new features will give us more to tinker with and it's very early in checking things out, however, one thing I don't get about this view and the amount of general thumbs up about it is the random shiny / coloured spots on the wood at the top right and left side especially?  That part doesn't seem realistic to me for these blotches to show on wood?  Also the lower right sling shot seems overly saturated with a purple hue for the lights off view, did you set your light emission colour to a more purple one intentionally or is that all from the environment image you used?  The steel ball guides / sides at the top look quite nice though.

 

[jimmyfingers]

Hello,

 

This was an example of an over saturated HDR map to give an idea of how things could look if you were going for that type of effect.

 

I like this particular map because on higher wire forms the light illuminates the ramp from below like it would in a dark room with GI bouncing off the underside of the ramps.

The random over saturated spots are are indeed unrealistic but the table is about 5% complete too so those wood decals won't appear in the final build.

 

The reason the right slingshot is purple is because none of the parts that make it up have a material assigned to them.  This has nothing to do with the environment map.

 

Right, after posting I thought more about the purple and figured that it was likely the materials missing with the default purple colour and the overall way it looked (I've changed my default material colour to a medium grey for playing and switch to a green for editing so haven't seen the purple in a while).  Thanks for your thoughts on the other aspects.  So those ramp / metal ramp highlights now are done with the map vs. you manually doing them with flashers, which I think you did on some other tables for that effect?  Only thing about using the maps for lighting highlights like this is that it will be static so when the GI goes off, the highlights will remain (from what I understand, Toxie said that they cannot be switched during game play).  

[Dozer316]

 

Hello,

 

This was an example of an over saturated HDR map to give an idea of how things could look if you were going for that type of effect.

 

I like this particular map because on higher wire forms the light illuminates the ramp from below like it would in a dark room with GI bouncing off the underside of the ramps.

The random over saturated spots are are indeed unrealistic but the table is about 5% complete too so those wood decals won't appear in the final build.

 

The reason the right slingshot is purple is because none of the parts that make it up have a material assigned to them.  This has nothing to do with the environment map.

 

Right, after posting I thought more about the purple and figured that it was likely the materials missing with the default purple colour and the overall way it looked (I've changed my default material colour to a medium grey for playing and switch to a green for editing so haven't seen the purple in a while).  Thanks for your thoughts on the other aspects.  So those ramp / metal ramp highlights now are done with the map vs. you manually doing them with flashers, which I think you did on some other tables for that effect?  Only thing about using the maps for lighting highlights like this is that it will be static so when the GI goes off, the highlights will remain (from what I understand, Toxie said that they cannot be switched during game play).  

 

Yes, the resultant spots are because of the map but remember it is set very high at the moment so the effect is greatly exaggerated.   On the final version I will pull it back a bit as I am sure you are correct about the map staying static when the GI brightness is variable.  I am hoping that on this game because the GI never really goes out - (sits on step 1 at the lowest I think) it won't be too jarring to the eye to still see subtle specular highlights on the wire forms even though the GI brightness changes. 

I guess the thinking though is to actually compliment the static environment map with variable highlights as I used to do on some of the other tables I modded in the past.  This way I guess the table could be thought of as sitting in a room with a light source outside it which is constant and variable light sources coming from within.  Using this approach a map which wasn't so directional to the underside of wireforms would probably be a better choice.

[dark]

I made my display pic using HDRI lighting and reflections which produces very realistic effects.

 

This page explains the differences between HDRI (environment) images and how to convert them.

http://gl.ict.usc.ed.../tutorial04.php

 

If you just google 'HDR maps" or "HDRI maps", lots will start to come up, you won't know for sure how the maps will effect your lighting and reflections until you try them out, so have fun experimenting.

Edited by dark, 12 October 2015 - 03:43 PM.

[toxie]

Toxie, can you please clarify how we go about just making things look like the way the did before (how do we get the old environment image back or do we set it to blank).  These new options seem like they could be cool but to be honest I was happy with the way it looked and how I had personally tweaked the lighting before these changes so would like to be able to continue in that realm while exploring these others.

 

Also, you mention playfield reflection on the ball is now reenabled / working but it is not showing anything on the ball from the playfield texture bound to the table no matter what view (FS / desktop) or settings I try. I've used very obvious patterns for the PF as a test and nothing is displayed on the ball. Are you only reflecting these new static environment pictures in the ball?  I thought the idea was as worded for the playfield to be reflected in the ball but it's just not happening even on the default / new table (take a look with some different PF textures and views and the ball appearance doesn't do anything even with reflection set to 255.  Also, I see in the latest build that you have even removed the options further and now the ball option for reflection is simply an enabled flag (-1, 0, 1) and all settings are controlled still by a single strength for reflection in a different tab now.  Do you not think that the reflection of the playfield on a steel ball should be able to be set differently then the reflection of the ball and table objects on a clear coat PF surface (but even right now it doesn't appear that PF reflection on ball is actually showing anything)?  

 

I've posted some other items in our initial testing PM so would really appreciate if you could respond to some of these things.  Also, performance has dropped by about 5-10% with all other things being held equal on my testing rig since these builds even when reflection is disabled.

 

Glad to see Dozer that you're working on a VPX CV.  The new features will give us more to tinker with and it's very early in checking things out, however, one thing I don't get about this view and the amount of general thumbs up about it is the random shiny / coloured spots on the wood at the top right and left side especially?  That part doesn't seem realistic to me for these blotches to show on wood?  Also the lower right sling shot seems overly saturated with a purple hue for the lights off view, did you set your light emission colour to a more purple one intentionally or is that all from the environment image you used?  The steel ball guides / sides at the top look quite nice though.

 

You can simply set the environment to <none>, or leave it as-is (e.g. not set it) on your table, then its the same as before. But you should really try out some other ones just for the fun of it, especially as the old default one is not neutral gray as one would expect (sorry, but it all started as a quick test, and then it took me forever to follow-up on that specific issue ).

 

As for the reflection of the playfield on the ball: There is still a bug in the build available here, that i fixed this moring, but one table where you can see it clearly for example with the current build would be JPs Nascar MOD (slightly plunge the ball up to the letters in the lane). With the fix you should see it on all tables that do use the "original" builtin playfield image like it is intended (e.g. no additional walls or whatever on top of the playfield).

 

As for the removal of the reflection flags: This was actually Fuzzel, i just moved stuff around.  And it basically just removed the specific ball reflection setting, as now this is generalized into the playfield reflection setting due to how it is all implemented (e.g. separation not possible).

 

As for the PM thread: I know, i know. Please bare with me. I have to catch up on some things like this one, but first wanted to get some of these old bugs and ill features ironed out.

[atarian]

Alright while I experiment with the HDR environments I figured I would share some samples with people, maybe it'll help others...
 
from this link, I decided to try a few on my Sopranos build to see what happens. For this, all my GI lighting is set to white/full white (not how I had it set in the previous preview but it looks better with white lighting using these environments
http://www.smartibl....bl/archive.html
 
here's a link to a gallery of some samples with the hdr file names
 
Am I doing this right?
Please share any other free HDR environment resources you come across... a lot of these come out a little yellow/orange. Here's my lighting settings:

This is really nice, they look so different. There have been times where I thought it would be nice to change the look of a table in that way and had limited success with SweetFX under VP9. The lobby centre one for example brings a really nice warmth to the table - a look particular suitable to EMs played in a dark room.

It's a really awesome feature.

Edited by atarian, 12 October 2015 - 05:03 PM.

[freneticamnesic]

I used the environment maps from those links. As toxie noted that is wrong I'll try again with the full detail hdrs

Sent from my SCH-I545 using Tapatalk

[toxie]

yes, all this naming for this is completely confusing anyhow, there is no real standard unfortunately in the rendering community so far.  :/

(i personally would vote for HDR environment lighting or emission)

 

 

btw: could you guys please test FS vs DT lighting again? i changed the behavior (at least) for the FS setup, and i -think- it should now roughly match up between FS and DT, but then again i only tested it sparsely.

(context: unfortunately i cannot find the posts anymore, but there were complaints that the lighting behaves differently in DT or FS mode, like being "not rotated along" with the table)

Edited by toxie, 12 October 2015 - 05:34 PM.

[fuzzel]

rev2223 is up:

 

- bug fixes for the playfield<->ball reflection

- some cleanups

[dyopp21]

yes, all this naming for this is completely confusing anyhow, there is no real standard unfortunately in the rendering community so far.  :/

(i personally would vote for HDR environment lighting or emission)

 

 

btw: could you guys please test FS vs DT lighting again? i changed the behavior (at least) for the FS setup, and i -think- it should now roughly match up between FS and DT, but then again i only tested it sparsely.

(context: unfortunately i cannot find the posts anymore, but there were complaints that the lighting behaves differently in DT or FS mode, like being "not rotated along" with the table)

 

EDIT: just for clarification I'm playing in FS mode.

 

 2044.PNG   1.16MB   19 downloads  2217.PNG   963.12KB   19 downloads

 

Not sure if this is part of the HDR environment that has been introduced, but 32assassin and I have been working on a Blackout beta table and I have been playing it in a few different versions.  In an older build (2044, I know, I haven't updated my laptop build in a while!) the insert lights have a nice amount of bloom on them and are plenty bright.  However on the new build (2217) they seem more muted and the circular area where it is lit, vs. not lit is far more distinguishable.   I tried setting the HDR environment to "none" but the lights still seem muted.

 

Also, on the older version the ball movement seems natural but on the new build it does some strange things in the inlanes (goes zipping back up the table) and off the top rubbers ("spins" off them in a lateral and down motion), as well as making some strange lateral movements off the the upper sling rubbers from, I assume, ball spin?

 

Here is a link to the table currently if anyone would like to see the differences in lighting and ball movement in the earlier builds vs. the new build:

https://www.dropbox....a V0.4.vpx?dl=0

 

Admins: hope I'm not putting this in the wrong place.  I'm thinking this is as much an application anomaly as it is a table build issue.  

Edited by dyopp21, 12 October 2015 - 07:33 PM.