LISTEN TO PODCAST SESSION 30
New features of Terragen 3.2 – Interview with Oshyan Greene from Planetside Software
I am very pleased to introduce Oshyan Greene, the business manager of Planetside Software, which produces the highly regarded Terragen 3, a powerful solution for rendering and animating realistic natural environments. In this session we're going to discuss the features of a new update for Terragen 3.
In this podcast session we'll learn:-
- Highlights for Planetside and Terragen 3 during 2014
- All about new glass shader.
- Stereo camera handling and support of the Oculus Rift.
- A fisheye lens can now be applied in Terragen 3.2, we discuss uses for the fisheye in Terragen.
- More control over cloud shapes can be applied in 3.2 with the "Taper top and base" parameter.
- Notable media productions created by Terragen 3
- Plans for Terragen and Planetside in 2015
Be sure to sign up for our free monthly magazine delivered right to your inbox http://3dartdirect.com/ (http://3dartdirect.com/)
http://3dartdirect.com/3dad-030-new-features-of-terragen-3-2-interview-with-oshyan-greene-from-planetside-software/ (http://3dartdirect.com/3dad-030-new-features-of-terragen-3-2-interview-with-oshyan-greene-from-planetside-software/)
(http://3dartdirect.com/wp-content/uploads/2015/01/FrankBasinski.jpg)
Image by Frank Basinski
procedural erosion??? Holly crap! OMG, OMG! I want that stuff :D :D
I hope you'll announce when we get to play with the SDK plug-ins when you announce them this month. ;D
Quote from: Upon Infinity on January 10, 2015, 06:13:07 PM
I hope you'll announce when we get to play with the SDK plug-ins when you announce them this month. ;D
Quote from: archonforest on January 10, 2015, 10:25:11 AM
procedural erosion??? Holly crap! OMG, OMG! I want that stuff :D :D
??? This is in the podcast, you already heard it?! Or you guys are really stoned right now?
Quote from: TheBadger on January 10, 2015, 07:16:26 PM
Quote from: Upon Infinity on January 10, 2015, 06:13:07 PM
I hope you'll announce when we get to play with the SDK plug-ins when you announce them this month. ;D
Quote from: archonforest on January 10, 2015, 10:25:11 AM
procedural erosion??? Holly crap! OMG, OMG! I want that stuff :D :D
??? This is in the podcast, you already heard it?! Or you guys are really stoned right now?
Uh, yes it's in there...
:o
Now I am high too!
:D
- Oshyan
Sounds really cool, procedural erosion. Great podcast, by the way, Oshyan.
lets get high without drugs!! Procedural erosion baby :D
I liked a lot how Oshyan sounded more excited about the projects people who hangout in this forum do, relative to how the big studios and such use TG. That was really cool.
And yeah, this erosion sounds amazing. Cant wait to see the first images from it. Also VERY interested what other plugins are coming... I have a long wish list ;D
But none of it matters. Because when the VR stuff is in full take over mode, I am already planning on dropping out of reality to become a full time VR addict. I wanted to create environments. But I think I will be too busy playing with the places other people create.
Oshyan (or anyone who is working with VR) Have you any idea what kind of market places there are or will be for VR content? Its really not too costly to get some of the basic things you need to build VR enviro (non games). no way other than through the big companies to get content to the coming flood of users yet? YOUtube for rift!?... or something
By the way, I just saw the samsung/Oculus Rift, it only works with one of their phones. It will be interesting to see this happen. I hope I don't need a different rift for every content producer!
The Rift development kit is cheap. Come on guys! Lets Make TG dominate the indie envrio content (when we find where the market is)! "Mood worlds to trip out to"... Or some other marketing title ;D.. "EXO PLANETS ON PINK FLOYD"?
When you guys are ready lets start a open thread on RIFT and TG!!! Where we can talk about all aspects!
I am gona get an occulus for sure! Gonna play FEAR for a week :D :D
And imagine some of us creates a TG image and then an Occulus version of it just in case someone wants to walk between the trees!!!!!!! Fawesome!!!!
Quote from: archonforest on January 11, 2015, 02:37:01 PM
I am gona get an occulus for sure! Gonna play FEAR for a week :D :D
And imagine some of us creates a TG image and then an Occulus version of it just in case someone wants to walk between the trees!!!!!!! Fawesome!!!!
All real human contact is about to end forever. GREAT! No more flu then ;D See you in the matrix Bud ;)
Quote from: TheBadger on January 11, 2015, 02:39:08 PM
Quote from: archonforest on January 11, 2015, 02:37:01 PM
I am gona get an occulus for sure! Gonna play FEAR for a week :D :D
And imagine some of us creates a TG image and then an Occulus version of it just in case someone wants to walk between the trees!!!!!!! Fawesome!!!!
All real human contact is about to end forever. GREAT! No more flu then ;D See you in the matrix Bud ;)
LOL!
Thanks for the feedback everyone. We're very excited about the work Daniil is doing and can't wait to show it to you. This will be a first for publicly-available technology (Dmytry's MojoWorld erosion work was never actually available for anyone to try, unfortunately), so it's a huge deal.
As for VR/Oculus Rift/etc., well it has a lot of potential, certainly. And I think more than ever before in the history of VR, mainstream content developers are *getting onboard*. Mostly it's games because the head tracking has the most obvious and natural application to a realtime and interactive environment. I think 360 degree "cinema" or video of any kind is more of a novelty *and* harder to get right. Ensuring your viewer is looking in the right place at the right time, or adapting to them not doing so, will be a huge challenge.
Just for one example, imagine you're making a great Sci Fi/Action/Horror movie and the tech issues of filming your scenes in 360 degrees easily have already been solved (obviously traditional sets are an impossibility at this point, so lots of CGI). Now it's in the final 3rd of the film and you have the dramatic reveal of your Alien villain, who will jump out from behind a door that appears to be empty when it's first opened. Your viewer, with VR helmet, is looking at the door open, then as soon as he sees it's empty, he looks behind him expecting to be caught by surprise. Then he hears a dramatic music cue and sound effect, but sees nothing, only to turn back around and see the alien in the doorway, but now a second or two after the initial jump scare, totally changing the impact of the moment. This is basically a hugely magnified version of the issue with Depth of Field and viewer focus in 3D, where some people don't like watching 3D because their eyes expect to be able to focus on any part of the environment (since it's 3D), but Depth of Field is still used (necessarily) by the filmmaker to draw viewer attention to the areas of importance. When you have a whole 360 degrees of possible view, it's going to be very challenging to make any kind of traditional cinema.
SO, with that in mind, yes I think Terragen-type stuff has a lot more potential to really take off because it generally doesn't require that same kind of specific focus. I can certainly imagine a wonderful flight around Mt. St. Helens or a Mars environment or any number of other things that are made even more enjoyable by the ability to look all around as the camera moves. I do think this has a lot of potential, which almost puts TG users ahead of the game in terms of content production for non-game uses in these VR systems, because as I said traditional film storytelling techniques are really going to struggle to do so.
I am sure that services like YouTube will quickly catch up and implement necessary playback tech if the hardware ultimately takes off. It's going to take a critical mass of hardware support though, just as it did for 4k and 60p. It's easier for YouTube and other content aggregators/players to support this kind of tech than on the *production* side of it, of course, but VR in particular is a more demanding area than just bumping up to 4k resolution (a simple increase in resolution vs. a total change in the image projection and scope), or going to 60fps. I'm not aware of a real standard existing for video in 360 degrees yet, though I know Quicktime and other formats do support it. So one or more "standards" will have to gain support first, and in the meantime it's going to be challenging to get your content viewed on all devices and platforms.
It's an exciting time. :)
- Oshyan
^^ YES!
QuoteI think 360 degree "cinema" or video of any kind is more of a novelty *and* harder to get right. Ensuring your viewer is looking in the right place at the right time, or adapting to them not doing so, will be a huge challenge.
Death to the hollywood model! Its been here since silent films. Enough already, lets move on!... Because of everything you wrote in your next paragraph, so how about this:
Instead of the hollywood model (and repetitive narrative) We make "movies" of one single event. So one fight in a war, one or two hours during a calamity on mars, ect ect. But instead of a single narrative on one linear timeline, every thing that happens, happens simultaneously. All of the 3D characters, and events go on in real time over that two hours regardles if anyone is there to see or hear it or not. Every movie that is made would instantly become re-watchable almost forever. Every time you saw the "movie" it would be unique to you and whoever was with you.
With millions of people watching (unseen by one another or not) there could be whole communities set up to discuss everything you should see the next time you see the show!
Or, like in a game, certain parts of a film only happen once the viewer has triggered it. So a mystery film that has some interactivity, maybe. The possibilities are endless!
Quoteyes I think Terragen-type stuff has a lot more potential to really take off because it generally doesn't require that same kind of specific focus. I can certainly imagine a wonderful flight around Mt. St. Helens or a Mars environment or any number of other things that are made even more enjoyable by the ability to look all around as the camera moves. I do think this has a lot of potential, which almost puts TG users ahead of the game in terms of content production for non-game uses in these VR systems,
Oh man! Add music and sound effects and you got your self the new "crack". We need wind, particles, and better object animation in TG, STAT!.. plus maybe a few other things to put the icing on the cake... Even just google maps would be fun ;D
QuoteI said traditional film storytelling techniques are really going to struggle
Its also likely going to cost them 2-3 times more a film if they try to keep their standard models. But in any case there is going to be a lot more jobs for modelers and asset makers!
QuoteI am sure that services like YouTube will quickly catch up and implement necessary playback tech if the hardware ultimately takes off. It's going to take a critical mass of hardware support though, just as it did for 4k and 60p. It's easier for YouTube and other content aggregators/players to support this kind of tech than on the *production* side of it, of course, but VR in particular is a more demanding area than just bumping up to 4k resolution (a simple increase in resolution vs. a total change in the image projection and scope), or going to 60fps. I'm not aware of a real standard existing for video in 360 degrees yet, though I know Quicktime and other formats do support it. So one or more "standards" will have to gain support first, and in the meantime it's going to be challenging to get your content viewed on all devices and platforms.
Content will dictate winners. The site with the *most* quality content will win. So... Well likely (if you don't really define "Quality" too strictly, than it will be a porn site) but after that, yeah youtube could really grow even monstrously larger than it is. And since you can get youtube on every device...
I completely agree TG is the right tool for this! It makes sense even to a lay person like me. We just need a few more things. But even now we could do some really fun small scale stuff that people would like if there was a way to get it to them.
Like I said you can get a developers kit now for cheap. But I still don't know much about the part where you take a TG world, and put it in a RIFT, where the viewer can do more than just stand still and look around.
this will work with TG, and rift http://www.kolor.com/video and its affordable.
Rendering the TG world as a 3d environment (whee you can move through it) though is hard to imagine how.
If its as good as people are saying it already is. It sounds almost like a sure thing. No?
The "hollywood model" means different things, what you're talking about isn't really hollywood, it's just traditional film-based storytelling, which is derived from writing, which itself generally has a single "perspective" (at a time), or at least each of its possible perspectives can generally be mapped to a literal perspective/camera on film (the "omniscient observer" being the hardest, generally accomplished through narration, which is reasonably effective though not perfect). The hollywood funding and content selection model is terrible, I'd agree, but the storytelling approach is more about historical precedent and expectation, as well as potential limitations (or at least preferences) in how humans experience narrative and relate to stories. VR isn't going to help address either of these.
The only way to currently create what you're talking about (this huge "meta-narrative") would be total CGI, game environments basically. Even offline rendering (non-realtime) wouldn't really be possible, let alone filming in real life. Either that or accept the limitations of specific camera positions, if not angles (assume 360 degree cameras, but they still have to be positioned *somewhere*, and then you have a finite number of cameras, and perhaps more importantly a finite number of *microphones*. And can you imagine trying to film this kind of thing? It would probably all have to be one take, or at least would be better as one. Even one 1hr take, I mean it's like a play, except on a massive scale. Sounds incredibly hard, if not impossible with current tech. CGI, realtime or otherwise would be more manageable but still a huuuuge undertaking, multiplied by however many stories you want to try to depict.
As for TG's contribution, realistically it is only either as A: a sort of background (like in games presently, skyboxes, etc.), or B: as a less interactive experience (you can only look around, not control your character movement, so to speak), simply due to the limitations of TG's non-realtime rendering approach.
- Oshyan
Well I cant argue that we don't have all the power yet that we need for what I want. But if it takes off (and I am convinced it will in one brand or another) than development will be a whirlwind! Much faster and pervasive than HDTV, it will be more like smart phones and tablets and PCs in that regard. Having not tried it yet myself, I am basing this on the squeals of joy from people who use it, especially non-critic off the street people who have been demoed.
The thing is, it does not have to be the matrix at first, just a personal IMAX, and it will change everything with in a few short years. And its already so much more than that already.
Here is a link to a story about one of the first movies: http://www.foxnews.com/tech/2014/02/19/forget-3d-glasses-new-movie-puts-directly-in-action/
About youtube, this story says NETFLIX is already onboard as a way to distribute, first their library, then arguably new content when its ready: http://www.digitaltrends.com/movies/oculix-mashes-oculus-rift-together-netflix/ It is not so hard to sell content to Netflix as it is to studios, especially if you have a finished product, they buy everything they can.
Now here is a interesting twist on it all: http://www.foxnews.com/tech/2014/10/22/magic-leap-could-be-oculus-rift-killer/
As far as TG goes now, I am learning 360 *video*. While I know and understand what your saying about TG's non-realtime rendering, One thing that we can do now thats still pretty cool is a series of 360 "rooms, where , no you *can't* freely move (only look around), but like you can see in a web browser now, you can "warp" from one place to another. This would still be very entertaining to start with! With audio and some visual effects like wind and such!
QuoteThe "hollywood model" means different things, what you're talking about isn't really hollywood, it's just traditional film-based storytelling,
Well yeah, i am talking about it in terms of Film history classes at film school. Which is how its talked about in criticism and academia. While I did not care much about race and gender in film studies (who goes to see a hollywood movie to get social justice?) I do agree with my teachers and the huge majority of scholars that the "hollywood distribution model" and the "Hollywood naritive", are totally boring and corrupt. I also agree with you that their way of picking films to develop is not the best for sure.
VR will change both because of all the reasons you said, and many many more. I guess we agree, but FILm history is a field of study and thats the only way I know how to talk about it where definitions are mostly consistent.
I don't know, Maybe its not as fascinating to others. But I am in love with this VR stuff. Besides entertainment, I am fascinated by the new ways learning will happen when content is pervasive. One of the problems with "e-learning" is all the studies show that it it completely inferior to in person learning. That learning always works best in person... VR history and geography in person?! Maybe not for a decade more yet. But the more excited we get (and spend) the faster development will come. Games, porn and movies will all but guarantee this.
One last thing.
Did you ever play halo3? At the end there is this slightly interactive monument to the in game war. Where you can move through the battle and see models of everything (like a table top model of the civil war at a museum). This is a good visual of what I meant by an all at once movie. I have no problem at all with it being 3D instead of live actors. Just make it move and there you go http://img2.wikia.nocookie.net/__cb20070915135832/halo/images/7/78/Time_to_Believe.jpg
http://2.bp.blogspot.com/_EjOWGtkKY4g/TDzDqAH3RLI/AAAAAAAAAW4/2Qrg9EgCUZA/s1600/BT+game+001.jpg
Maybe its too intensive for servers to distribute now, but I think that will change too.
Its like you said in the past. We need to develop better compression methods for data.
Ok so maybe I will be wrong. But I think I will be mostly right in the end. Even if it takes slightly longer than I (we?) want.
Oh, and, that is why I am glad TG is getting in the game as much as it can. If RIFT and the sony and others fly at all, they will soar! Otherwise it will just crash and burn. I just cant imagine how it can be in the middle. Its just too cool. It represents the next mile stone to me. And I don't think you have to be an expert in anything to see it, mostly.
Some interesting links there. The Netflix thing is really just a gimick, but I was honestly surprised to see someone is already making a documentary for Oculus Rift. Pretty cool. He describes the challenges well, and suggests that using game design-derived techniques is a good way to start reframing the "movie" development process, which I agree with. I will say, however, that many story-driven games these days *still* use somewhat forced viewer frames, for example in a first-person shooter during a story segment, your ability to control your character is often limited or none, maybe you can turn your head a *bit* but not fully away from the action for example. That kind of limitation becomes a much bigger problem on Oculus though because the user can *physically* turn their head wherever they want, and if you start limiting their frame at all, then their physical movement no longer maps properly to their view and it becomes disorienting and/or frustrating. So, still lots of challenges to deal with, but it's good to see people are working on them.
- Oshyan
Hey Oshyan, You have not said if you are planing to buy? Do you have a development kit (or matt maybe?) Or will you? Or are you going to wait and see?
You definitly seem more cautiously optimistic on a personal side. I dont count PS, because that is business, and its completely understandable why you would control excitement expectations. But personally are you going to buy? Maybe the sony?
I mean I know that you are working with developers, and you can't talk about that part. But that counts as secrete knowledge, and I cant argue or learn from a secrete I am not in on, very saldly (that Im anot in o n it ;D)
Maybe you will spill the beans and tell us whats happening behind the curtian! ;)
I don't have one and I plan to wait and see how the initial production run goes. Matt might buy one though. ;)
- Oshya
MATT! Spill the beans matt. NDA pffft, feed the beast!
I have the DK2, but other priorities have meant I haven't spent much time with it yet.
Matt
Any chance the pro erosion will be out in time to use in the contest? Man what a aspirin that would be for my headaches!
No, I'm afraid not. But there's plenty of good Iceland terrain around. :)
- Oshyan
"Zero Point" is out for download, Oshyan. take a look http://www.theverge.com/2014/10/28/7078925/zero-point-vr-oculus-rift-documentary-released
Cool. A shame I don't have a Rift, but maybe Matt can check it out at some point and let us know, hehe.
- Oshyan
^^ That would be cool.
Is there any update on this announcement?
Matt?
Khmm...procedural erosion??
Fair question, fair question. Daniil is squashing some bugs he wanted to fix before showing results to the public, but I'll try to get something good for you guys to see soon!
- Oshyan
Quote from: Oshyan on March 05, 2015, 05:32:22 PM
Fair question, fair question. Daniil is squashing some bugs he wanted to fix before showing results to the public, but I'll try to get something good for you guys to see soon!
- Oshyan
2 thumbs up
Quote from: Oshyan on March 05, 2015, 05:32:22 PM
Fair question, fair question. Daniil is squashing some bugs he wanted to fix before showing results to the public, but I'll try to get something good for you guys to see soon!
- Oshyan
Awesome! :D
Oshyan- I enjoyed the podcast. (Only) when you are talking there is an annoying noise in the background. It sounds as though your interviewer is thumping his fingers on the table while you are talking (but not when he is talking). Could it have been that this was a remote interview (via Skype or some such thing) and you had some sort of background noise going on at your end of the link?
It was indeed via Skype. There is an occasional noise that is due to a train that passes by my house every 10 minutes or so. But I didn't notice a more regular sound...
- Oshyan
Quote from: Oshyan on March 05, 2015, 05:32:22 PM
Fair question, fair question. Daniil is squashing some bugs he wanted to fix before showing results to the public, but I'll try to get something good for you guys to see soon!
- Oshyan
Any updates on the erosion stuff?
Hehe, I see you're on top of the long-term project status updates. ;) As it happens I just got a new test version from Daniil a couple of days ago. We're hoping to officially show a full animation and some detailed shots soon, but it's not quite ready for prime time...
Since you've been so patient though, here are four images. :D
First, the un-eroded terrain.
[attach=1]
Then, terrain eroded with Daniil's plugin.
[attach=2]
Then, a close-up shot, center-frame.
[attach=3]
Finally, a zoomed-out distant shot of the same area.
[attach=4]
All detail is from procedural erosion on a base Power Fractal terrain with only 8 octaves of noise. There is no additional displacement.
More to come!
Please note that these are preliminary results. The algorithms are in flux and output may change significantly, but you can bet it will only improve from here!
- Oshyan
This gots potential! Will this be in the 3.3 release?
Good Lord! 8)
I know it is too early to inquire but naturally, there will be anticipation. Hopefully we have some nice controls over the channels and sedimentation as well.
Extra fine results; looking forward to this.
One more, from 160km above...
- Oshyan
Holy cow, that's awesome! I advice you to put it in alpha tomorrow ;)
One can only try and imagine how lovely that will look when fully shaded.
I can't wait - WOW!!!!
Can we mask it with painted shaders? Please say yes!
I see a river in my future
You can mask it just like any other shader, so yes the Painted Shader will work. You can also texture it and add additional displacement, again just like any other shader.
Here's a higher-altitude view with some texturing added (courtesy of FrankB's NWDA planet), approximately 420km (no atmosphere of course). And speaking of masking, I added some large-scale fractal noise masking here to give a better distribution of mountains so it looks more like mountain chains and not just "all mountains, all the time". Granted a very dense area of mountains, but not out of the question. Just think of it as British Columbia: https://goo.gl/maps/fl8NH
;)
- Oshyan
Stop doing that!!! ;)
Speaking of rivers, that is the only other ingredient missing here.
Yeah, no rivers in this yet. But who knows what the future holds. Daniil's a clever guy, and Matt's no slouch either. :D
- Oshyan
I could imagine Rivers just being a modification of the Erosion Plugin. The way to possibly think about it in a simple sense is a very less-iterated, more clamped long and wide channeled erosion.
Just to add an appreciation of the realism behind the upcoming Procedural Erosion, I wanted to do a comparison of the other noise types to see how it all look from the same 160 km altitude as the Eroded image. All the other noises are Mt. Everest height.
Hi Chris,
With the Alpine Fractal, try adding 2 or 3 "lead-in octaves". This is good for creating larger structures that look like mountain ranges. But Daniil's erosion plugin is still better :)
I have some R&D in progress which specifically attempts to create river channels (and lots of them). It could be combined with Daniil's erosion, or used alone. Lots of good things are coming soon :)
Matt
Hello,
For the moment a 'general' workflow is often to import WM (with erosion masks) '2D' terrain, and use terragen to give more details into vertical surfaces.
But somehow erosion 'should' be computed as a last stage, so this new feature sounds so exciting !
I'm really curious to see how this 3d erosion works on overhanging shapes. (after a strong twist and shear shader for example).
Please keep posting results 8)
Quote from: Matt on May 19, 2015, 08:06:43 PM
Hi Chris,
With the Alpine Fractal, try adding 2 or 3 "lead-in octaves". This is good for creating larger structures that look like mountain ranges. But Daniil's erosion plugin is still better :)
I have some R&D in progress which specifically attempts to create river channels (and lots of them). It could be combined with Daniil's erosion, or used alone. Lots of good things are coming soon :)
Matt
Indeed...
Darn you, quit tempting us!
MojoWorld was the closest program to make continental-size river systems with a plug-in some fellow had made, forgot his name but that is as far as anyone had come with planetary river systems.
Quote from: paq on May 19, 2015, 08:38:58 PM
Hello,
For the moment a 'general' workflow is often to import WM (with erosion masks) '2D' terrain, and use terragen to give more details into vertical surfaces.
But somehow erosion 'should' be computed as a last stage, so this new feature sounds so exciting !
I'm really curious to see how this 3d erosion works on overhanging shapes. (after a strong twist and shear shader for example).
Please keep posting results 8)
The general workflow for local terrains from World Machine and World Creator is still going to be more powerful because of the many options available. It depends on what you, the individual wants to do. The lovely thing about procedural is near-infinite scales and staying in the same software workflow.
I increased those octaves and not too bad for what it is.
I do have one additional question in regards to the up and coming erosion. Will we be able to erode Terragen's objects as well?
Uh... odd question. Technically yes, but the results may not be as expected. Why do you ask?
- Oshyan
One may want it simply to have a look of being aged.