Tenkoku - Dynamic Sky

I don't see why anyone wouldn't migrate to Unity 5. Do it!

 

I don't see how or why anyone would not jump to unity 5. I mean even if they paid for 4 pro... they are basically getting all the pro features if they move to 5 without paying anyway... no point in being on 4 anymore. None at all. So by all means.. i say as an asset store developer... stop trying to make things work for 4 as well if u dont have to.

 

I'm only using Unity 5 now. No plans to use 4 again.

 

That is not entirely true. If you make over 100k, you will have to buy pro and if you already have Unity 4 Pro, it may be hard justifying spending $1500 ( or more if you target mobile or have multiple employees) to get Unity Pro 5.

 

Is it really that hard to justify spending $1500 is you already made $100k ? Thats a little greedy dont u think? lol.. I mean you can write unity off your taxes.. and if u dont.. the government is going to take that money from you instead. So with that being said.. i really dont understand your logic. Not to mention if u had unity 4 pro.. its like half the price to get unity 5 pro.. not $1500..

 

Yes, I did forget the half price off. But still, even then if have also have iOS and Android Pro, it will cost you $2250 to upgrade. Plus, if you have several employees in your studio it adds up to even more. For even a small studio of just five employees (including yourself), the cost rises to $11,250 (5 * 2250 = $11,250). $11,250 is nothing to laugh about even when you make $100K and tax write-offs. And your argument that spending $1500 when you make $100 is nothing is invalid. If it was true, people would not have stayed on Unity 3 when Unity 4 came out. And the $100k doesn't eveIt is totally reasonable why such a person would stay on Unity 4. That said, most Unity 4 Pro (and Free) users will likely jump ship to Unity 5 and Tenkoku should focus on that.

 

If all you made off unity was 100k and you have 5 employees... lemme tell you.. your already out of buisness... and it has nothing to do with unity. You shouldnt have that many employees and that many licenses. So your argument is again some far reaching thing ... where no one is in that position. Thats like saying what if nazi germany rose again and destroyed europe and decided to invade unity studios... some how i doubt thats happening to anyone right now.

 

Well I can certainly understand why some people may not jump to Unity 5, especially not right away. If you're in the middle of a project and have already invested resources then the move may be more of a burden than a help (depending on the project). I can also see how having to upgrade multiple licenses for a whole team can be something not to scoff at. Making a "$100k" sounds like a lot, but a Unity upgrade can quickly become only one of many things that that money will need to address. When it comes to your business these things need to be considered carefully.

I don't know if we need to bring up Nazi Germany here... this thread doesn't need to win an oscar. I was just trying to get an overall idea of whether 4 is worth supporting still. Thank you everyone for the answers so far!

 

I can understand some users not wanting to hear this, but my 2c is as you have not released it yet, its not such a big deal to support only 5+. Had you released it, different story and it would even be a bit unethical as many users would have built a lot around it and need a tech change. Optionally you could ship it WITH a second 4.x compatible version (or sell that separately) - which will never get new features (apart from any necessary bug fixes), so those totally committed to 4.x have that as an option.

Re being server-ready, yes the ability to access variables is needed, but the holy grail when it comes to server comms is minimal traffic. So nice if a simple timestamp can be passed to synch TOD as well as carry a simple value which will be used to orchestrate a weather pattern. I guess its something the developers can do, I was just wondering if you had anything like that built in. Also, whether you have any randomness built-in, because while its great for a single-player game, for multi-player syncing its a problem. If you do, then I suggest you allow the "random value" you use to be overwritten as a variable too.

 

We're switching to 5, and would not hesitate to buy 5+ only assets. ;-)

 

I really want this awesome sky for my next project (Unity5)! When will it come to the assetstore?!
Will it work with the OculusRift?

 

This looks amazing; already using Suimono.

Question: does this support 'Fantasy/Sci-fi skies?'

By that I mean, I want to control the color/texture of the clouds and atmosphere; something that wouldn't be present on earth. Multiple moons/custom celestial bodies would also be great.

So for example, would I be able to create an atmosphere similar to Venus, or other planets like that?

Looks awesome though! Excited for release.

 

I tested a beta in my Rift. It has an issue with the particle clouds right now but the rest of the system works great.

 

Great! Do you have still beta access? How to become a beta user?

 

Thanks guys.

The beta access was chosen on a per-project basis and wasn't exactly open to everyone, and I'm not currently adding any more beta slots unfortunately. The beta process gave me some real things to think about and illuminated some issues that needed fixing, so once again I've completely blown my projected release date. However I don't expect it to be in beta for much longer. as @CaptainMurphy notes there were some real issues with the cloud rendering and in the past couple weeks I've gone back and completely rewritten how the clouds are rendered. Below is a preview of the updated cloud system (will be released to you beta users soon )...



@markefus The system can be custom colored however you like, either by using a general tint color property, or by selectively editing a color texture ramp in photoshop, so sure you could create an alien sky with a little tweaking.

 

Preview shots look very nice & real!

Sounds like it could take a few more weeks / month until this great asset will hit the assetstore?!

 

That looks better than the version I had. Looks like you changed the orientation of the clouds to be more horizontal than spherical overhead.

 

Yes, and you can now control the direction that those clouds move. The clouds work better with Suimono now and should be much better with Oculus Rift.

 

Cool! Can't wait to get a chance to try it.

 

@chingwa Are you able to post a video of Tenkoku and Suimono working together over a full (obviously accelerated!) day-night cycle.

Thanks!

Andy

 

Wow! This looks amazing

 

@ivyroxy Thanks!

@EiKON-Andy Here ya go.....

 

That is looking good. Did you remove the particle effect clouds for now? Are they planned to be added back in?

 

Only problem I seen with it is at the very beginning you see the sunrise reflecting on the water but the sun is rising behind the mountains. Other than that, it is looking good...

 

@CaptainMurphy I removed them, yes. While they were really looking nice in a lot of cases, the incredible overdraw from the particles was quite an issue and try as I might, I couldn't really solve it without lowering the particle count to unacceptable levels. Probably won't be adding them back unless they can somehow be made more performance friendly.

@Shawn67 Thanks! Yeah the sun sparkle is a small issue. I might be able to fix this on the Suimono side in Unity 5.

 

Beautiful! Thanks

 

That is an issue that a lot of engines run into. Anytime there is a global type light like that they will not calculate anything that might be in the way when doing the specular reflection. Just about every lighting calculation doesn't handle that. It is annoying for having objects in the water like a ship as well because the reflection will go right through the ship when it should have a shadow there. Not much you can do about it until the engine handles it.

 

Is there any date when this is beeing released?

its a long time now under development. And would be so awesome to get it

 

Hi UltraTM,

I don't have a date to give out. Everytime I do I end up missing it so I don't want to keep frustrating people. Currently I have an issue with the accuracy of the star/sun/moon positioning. It works well for some dates, and then other dates seem off by a large percentage... I concede that for most people this probably isn't an issue, but it will be for those who are looking for accurate positions.

This is the last real bug that I'm looking at fixing before putting out the 1.0 release, and I hope to have it fixed soon!

 

Yes, most users wouldn't even notice that (or know its wrong, or care!) ... so its definitely a good candidate for a post release patch.

Tenkoko + Suimono look good together ... that video makes me want to go fishing ... how about a special "2 for 1" limited time release package deal
Some users have other water solutions, and some other atmospheric solutions ... that will help them to make the switch to both.

 

Glad to report that Justin has nailed the bug with regards to accuracy of Sun/Moon and Star positioning .
Current tests I've run have been 100% accurate using a in-game version of a long standing Mariner measuring device.
The Navigational sextant.



Some of my Results complete - with real world data to verify.

Click to see full size.


Click to see full size.

(Drawn in the red lines on the Bid Dipper to make it easier to see)

Cant recommend this package enough for any of your environmental needs.

 

@John G. Thanks for your detailed testing in this regard. That sextant you're building is damn cool

So yes, this was the main bug that was preventing me from submitting version 1.0 to the store. I'm going over today working on some additional UI polish and should be submitting soon

 

Can't wait. Really looking forward to the release.

 

Looks great.

 

I'm in. I want to check it out, particularly performance on consoles

 

The wait is over...

 

YEY!!!!!

 

Sweet!

 

I am really looking forward to buying this although my current budget might have me waiting till the end of the month but oh well. Anyway I was wondering, how well does the system work in multiplayer environments? Is it possible to sync the weather changes between clients or is that not possible in its current state?
Oh and also I wondered if it is already possible to access elements like wind, temperature and/or humidity so that it can influence other game elements? I know this had already been talked about but I can understand this to be a low priority feature.

 

Was lucky enough to scoop a preview version to look at. Pretty amazing that the author's essentially managed to make his own version of trueSKY but with features like sound ambience and more. Thumbs up, really.

 

Currently you can access various cloud settings, rain, snow, fog, and wind settings. However I do plan on building and adding a more advanced weather automation system which would include variables such as temperature, barometric pressure and have tolerances be "tuneable" to the season and latitude. This will be for the future though.

There isn't currently any specific multiplayer code... however all variables are accessible and you could write some code that sends this data from the server to each client. There were previous discussions about packet and network bandwidth related to this, and i've been brainstorming with the best way to approach this.

I think the ideal would be to have the server do a Tenkoku "send" command which would encapsulate all data in one variable that can be sent over the network, and then each client could do a "decode" on that single data packet to then display the correct Tenkoku state... so everything would be done in one batch send. I'd like to get a bit more feedback on this first though...see how people use the current system and what would be most needed in the future.

I should have some additional docs showing how to access Tenkoku variables and change them via code soon.

 

I'll go ahead and put a link here to the main asset store thread. I'll keep this thread open for a little bit until the new one gains some traction though...

TENKOKU - Dynamic Sky
http://forum.unity3d.com/threads/tenkoku-dynamic-sky.318166/


Also... this...

 

I will stay in u4.x as long as I can. GI Lighting sucks as baking takes forever. Anyone wirh a mid/ large project will get stuck with lighting. Plus anyone in the middle of a project would consider this as a suicide move.

 

I've heard a lot of reports about crazy bake times in Unity 5. It must be a bug as I can't believe this is working as Unity intended (but, you never know). But otherwise I've only seen and heard praise about improvements in lighting. Care to elaborate/share your lighting pain?

 

I dont whant to drive off topic but in regards to your question, by lighting I meant to the GI lighting. I'll correct this. Without baking GI we've lost a really important part of the graphics power up tools. During the beta people was warning... and it has happened. Luckilly Unity has an awesome community and GI alternatives seem to be coming in the realtime GI side (I'm not talking about Livendas vaporware). But this remembers me to the umbra adition, it was announces with all the honors and when reached first to Unity you could only bake small scenes where it was less needed... hope they manage to fix overtime.

Your sky looks superduper, but the clouds looked better in your initial vids, now they look like stiched to a plane that moves in one direction.

 

OK Yeah, it seems like features get caught up in hype, and then the marketing takes over the reality... it's happened before... remmeber Shuriken? Remeber Mecanim? (I still prefer to use the original animation system) I personally haven't had issues with their "realtime" GI in 5.0, but I've only used it in small scenes.

 

As I said, small scenes ok, bigger stuff exponential time increase [unusable], at least Beast keeps handling large stuff so meanwhile I'll stick to this. If things dont change and really need to take advantage of something exclusive for U5 I'll bake with Vray plus Flatiron but this wont be as convenient as doing within the engine + cost increase.

Pesonally I think Mecanim works pretty good, in fact its one of the things that had been closest to what I expected.

I think unity should focus in providing a super flexible full featured core so that talented people like you can plug their assets without limitations. If might worth from their part investing in devs that cand bring superb assets instead of these marketing sh*tpiles.

 

I think perhaps this comment was directed to users of Unity Indie. 5 is a vast improvement for indie users because it has opened up a ton of features that used to be Pro-only. So the incentive to upgrade for Pro users is not as great as it is for indie users. I can't see why an indie user wouldn't want to upgrade.

 

To be more on topic, I have a question about Tenkoku.

How easy is it to access the time, date, and weather status via script? To use for AI, Quests, lights, etc.? I'm currently using Unistorm, which was developed with RPGs in mind, so all that is easy to do, there was even a R.A.I.N.-based demo for controlling NPC behaviours based on time of day, season, and weather. Is Tenkoku able to do the same sort of thing?

 

Accessing variables via script is very easy. I actually don't have this put into the manual yet, but I'll be writing out some example pages soon. All you have to do is cache a reference to the main module and then you can directly edit specific variables from anywhere in your project...

First Reference main module...

Code (CSharp):

1. private var tenkokuModule : TenkokuModule;
2. function Start () {
3.     tenkokuModule = GameObject.Find("Tenkoku DynamicSky").GetComponent(TenkokuModule) as TenkokuModule;
4. }

Then you can access/change any of the time variables...

Code (CSharp):

1. tenkokuModule.currentYear = 2015;
2. tenkokuModule.currentMonth = 3;
3. tenkokuModule.currentDay = 21;
4. tenkokuModule.currentHour = 7;
5. tenkokuModule.currentMinute = 30;
6. tenkokuModule.currentSecond =0;

And you can set position variables...

Code (CSharp):

1. tenkokuModule.setLatitude = 40.2;
2. tenkokuModule.setLongitude = 0.0;

And you can set weather variables...

Code (CSharp):

1. tenkokuModule.weather_cloudAltoStratusAmt = 0.1;
2. tenkokuModule.weather_cloudCirrusAmt = 0.2;
3. tenkokuModule.weather_cloudCumulusAmt = 0.5;
4. tenkokuModule.weather_OvercastAmt = 0.0;
5. tenkokuModule.weather_cloudScale = 1.0;
6. tenkokuModule.weather_cloudSpeed = 0.2;
7. tenkokuModule.weather_RainAmt = 0.0;
8. tenkokuModule.weather_SnowAmt = 0.0;
9. tenkokuModule.weather_FogAmt = 0.0;
10. tenkokuModule.weather_WindAmt = 0.3;
11. tenkokuModule.weather_WindDir = 180.0;

Or you can setup and trigger the automatic(random) weather generator...
the below will turn on the auto weather, set the weather pattern to play for 5 minutes, and set the between-pattern transition to about 10 seconds. However it's best to also set a boolean variable (below called randomizePattern) so that you can force a new pattern update when you want to.

Code (CSharp):

1. tenkokuModule.weatherTypeIndex = 1;
2. tenkokuModule.weather_autoForecastTime = 5.0;
3. tenkokuModule.weather_TransitionTime = 0.1;
4. if (randomizePattern){
5.      randomizePattern = false;
6.      tenkokuModule.weather_forceUpdate = true;
7. }

By editing the above variables you can essentially program Tenkoku to do anything you need, whenever you need it. However, other than the random weather pattern generator, there currently aren't any event-driven systems built into Tenkoku. I do plan on building a more advanced weather generator based on more realistic input, such as seasonal shift, temperature, latitude etc. I'd love to hear ideas for extending this functionality though.