Build A Better Skydome - Terragen 2 ?
#151
Posted 15 January 2023 - 12:58 AM
Agree.
About clouds flattened dome: for overcast ans rainy days, the effect is really looking better: now with changes, made by James, feeling of low clouds is being improved a lot; thanks to James once again.
But in scattered or almost clear conditions of sky,,, something important is absent, something not plausible is present. Sorry to say so.
@Dave.
I have noticed long ago, how ORTS makes colors less contrast in many places.
Though, I have never played with ambient brightness slider so far.
I've already posted multiple times here and there about phenomenon: some routes (later I've discovered - even part of large routes, made in different years) have "cold", but less of them - "warm" gamma. Second looks more pleasure for me.
About gray on Your downtown... I think, real city smog brings this effect. Crystal-clear air and "warm" sunlight make things looking warmer, but how many hours with such condition have people, living in Chicago?
I suspect, that may be tied with current "pale" color of Sun in ORTS (or am I wrong?) take a look on spectral stripes of heated hydrogenium and helium, which produce sunlight, filling all the space on Earth. Foggy athmosphere filters some shorter waves, when is cloudy.
#152
Posted 15 January 2023 - 06:43 AM
James Ross, on 14 January 2023 - 06:15 PM, said:
Currently it looks like we define precipitation density from 0 to 1 by simply scaling from zero to the maximum number of particles allowed, which is completely arbitrary but probably not a good idea to simply increase - we can probably do a better job of making it appear heavier than we currently do :)
We should be defining it using real-world measurements, such as mm/h and perhaps using the METAR list of precipitation types
In any case, the wind needs fixing first so that at least the clouds and precipitation match the simulation (and weather change events); after that, maybe other bits of the weather can be improved :)
Hi James,
Intimately familiar with METAR's and TAF's.
;)
Over the years - I have played with both the Rain and Snow textures in an attempt to make them look better and denser. One thing one did try with Rain was putting more than one rain drop on the single Rain Texture used by ORTS. While it did appear denser - you immediately see the repeating pattern which ruined the effect. Perhaps it's time to put some of those other (10) textures to use in ORTS? Also - the precip never seems to hit the ground which has been a common complaint for years.
Thanks for the hard work you are putting into this.
Regards,
Scott
#153
Posted 16 January 2023 - 01:30 PM
Genma Saotome, on 14 January 2023 - 07:28 PM, said:
Genma Saotome, on 14 January 2023 - 07:28 PM, said:
...
Much more interesting.
Agreed, and the contrast is more commonly what you can adjust in other games, I believe
As the Open Rails setting for ambient brightness dates back to 2014, and I can't dig up any discussions in the forum about its creation, it's not clear to me why it was added (vs just fixing a problem with the brightness)
So I'd put this setting on notice as part of the validate brightness, colours, and visible textures stage of the plan
Genma Saotome, on 14 January 2023 - 07:28 PM, said:
Genma Saotome, on 14 January 2023 - 07:44 PM, said:
Thanks!
Jonatan, on 14 January 2023 - 09:06 PM, said:
It is a striking visual phenomenon of the sky that usually can't be ignored and would be a nice addition to the sim.
The idea of lightning has been a recurring theme; perhaps even the only topic to come up every single year for over a decade: 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022 :sweatingbullets:
Once the planned work is done we will be able to investigate new weather effects light lightning, thunder, etc.
Weter, on 15 January 2023 - 12:58 AM, said:
But in scattered or almost clear conditions of sky,,, something important is absent, something not plausible is present. Sorry to say so.
Glad to hear the low clouds look good
Hopefully when we add extra layers of clouds and adjust the colouring, the scattered or almost clear skies will look better :)
scottb613, on 15 January 2023 - 06:43 AM, said:
Curious about the precipitation not hitting the ground, as it is coded specifically to stop right at ground level; something to investigate :)
I am going to have a search for recommendations for in-game weather and density; there are a few options we can probably combine (multiple drops per rendered primitive, instancing, alternative textures, etc.)
#154
Posted 16 January 2023 - 01:49 PM
James Ross, on 02 January 2023 - 09:02 AM, said:
- Fix distortion of sky textures - in progress right now
- Validate sun and moon positions calculated from latitude/longitude/season (including horizon crossing time)
- Validate brightness, colours, and visible textures (e.g. stars) for the day/night and three stages of twilight (dawn/dusk)
- Merge the disconnected wind simulations (currently there are THREE separate wind speeds and directions inside Open Rails)
- More general investigation of ways to improve the appearance of weather, etc.
I have completed the first 2 of these (in PRs and available to test in Unstable Version)
But instead of validating the brightness, etc., next, I am going to fix the wind and other weather parameters first - this will help avoid any confusion by having everything running off the same values, and they'll all be visible in the WEATHER debug HUD page
New order:
Fix distortion of sky textures- doneValidate sun and moon positions calculated from latitude/longitude/season (including horizon crossing time)- done- Merge the disconnected wind simulations (currently there are FOUR separate wind speeds and directions inside Open Rails) - in progress
- Validate brightness, colours, and visible textures (e.g. stars) for the day/night and three stages of twilight (dawn/dusk)
- More general investigation of ways to improve the appearance of weather, etc.
#155
Posted 16 January 2023 - 02:10 PM
Need to bring this to your attention. In the last couple unstable versions, users are experiencing crashes that I believe is related to these recent changes. See this thread:
http://www.elvastowe...table-versions/
I get similar crashes on my (TrainSimulations') Mullan Pass route, but not on some other routes. I can send one of my Log files tonight if the one if the above thread is not sufficient to track down the issue.
Has something changed in the code where it is now trying to reference something in the ENV folder in a strange way?
For what it's worth, in the routes that I can get to load, the new code for the sun path across the sky, and the sunrise/sunset times in different seasons is a BIG improvement. I also noticed the weird effect of dim/muted light that used to be present in early morning and late evening light is now gone, so that's an improvement as well.
#156
Posted 16 January 2023 - 05:23 PM
I feel compelled to ask, a few questions.
1- In relation to ambient brightness and what correlates it, is it possible to modify or implement atmospheric haze? currently everything depends on the Fog, while the latter should have a dependent function. It would be enough to set a sort of purplish haze as shown in this photo.
the second question is related to the implementation of atmospheric phenomena, interesting to know that we are working on a whole series of real phenomena! It would be interesting to understand if by correlating it with the weather in real time according to the coordinates using the web server, it is possible to implement a mix of various situations, for example: Only lightning, lightning and thunder, wind in correlation with them or not, rain, storm and rain..etc..etc..
Before closing.. little Off Topic..
I read that the improvement of the water shader is foreseen in the roadmap, in correlation to atmospheric events, it would also be interesting to understand and hope if it will be possible to obtain a shader rendering of this type .. this especially in coastal lines, could be related to the function of the wind and therefore having rough sea conditions, on the coast lines it would be a breathtaking sight ..!
#157
Posted 17 January 2023 - 01:22 PM
PerryPlatypus, on 16 January 2023 - 02:10 PM, said:
http://www.elvastowe...table-versions/
I get similar crashes on my (TrainSimulations') Mullan Pass route, but not on some other routes. I can send one of my Log files tonight if the one if the above thread is not sufficient to track down the issue.
Has something changed in the code where it is now trying to reference something in the ENV folder in a strange way?
The correct sunrise/sunset times come from the ENV file so changes were made to that code
Thanks for alerting me to the issue and other thread, I am investigating it now :sign_thanks:
FS.E652 091, on 16 January 2023 - 05:23 PM, said:
Yes, haze should be possible; it would probably be implemented together with any fog improvements
FS.E652 091, on 16 January 2023 - 05:23 PM, said:
We should extend the weather file / weather change events to be able to control all weather features in all appropriate ways
Hooking in real-time weather should also be possible, but would come after making sure the above is done, and might not be straightforward
FS.E652 091, on 16 January 2023 - 05:23 PM, said:
Water shaders like this are certainly possible, but one of the issues we'd have is that there is little to distinguish open sea, coast, rivers, and lakes in MSTS data
We may end up with some extra tile options or some other means of indicating the type of water body
#158
Posted 17 January 2023 - 01:55 PM
PerryPlatypus, on 16 January 2023 - 02:10 PM, said:
http://www.elvastowe...table-versions/
I get similar crashes on my (TrainSimulations') Mullan Pass route, but not on some other routes. I can send one of my Log files tonight if the one if the above thread is not sufficient to track down the issue.
Has something changed in the code where it is now trying to reference something in the ENV folder in a strange way?
Should be fixed in U2023.01.17-2151
#159
Posted 17 January 2023 - 02:40 PM
#160
Posted 17 January 2023 - 04:33 PM
Genma Saotome, on 17 January 2023 - 02:40 PM, said:
Yes, unless you're doing ray tracing, reflections are probably the worst possible thing to render - as you'd expect, you typically need to render the reflected objects twice (once normally and once reflected)
There can be cheats sometimes, but usually only when the camera has limited motion, e.g. early racing games could reflect scenery onto the track as a blended reflection by mirroring what was already on the screen, because the reflection was always far-away scenery drawn first, and the camera was low that the angles didn't make it too obviously wrong, but not everything reflects correctly, e.g. in Grand Prix 3:
https://james-ross.co.uk/temp/f1gp3_1.jpg
But you can often get a decent effect with simple bump mapping - which makes the surface normal fluctuate in various directions, making patterns when it points more towards the sun (brighter) and more away from the sun (darker) than if it was flat - combined with the right colour textures (e.g. blues) and translucency