new game engine for uru?

[andylegate]

Just wanted to give you guys an update on trying to use the CryEngine 3.0 SDK for Uru:

I spent 2 days tearing my hair out, trying to figure out how to import things into the SDK. The SDK's UI is quite alien to me in many ways, but those of you that know me, know that I learn very fast. Search after search was resulting in not understanding (there was one video tutorial that someone did, in which you spend 10 minutes of the tutorial watching him make a sign in Blender 2.5, export it to Max and then export it to CryEngine 3.0....but that last part is less than 20 seconds long ).

Finally this morning I went over to the CryDev forums and posted my plight. I explained that a few of us are trying to import an Age from Uru into CryEngine and I needed some direction. I lucked out as I got replies immediatly (they are very friendly over there. I even got a PM from someone who loves Myst and Uru, but doesn't know anything about Plasma or Age creation. I pointed him to here and the wiki, but he really likes the idea of converting things over and giving CryEngine a try and was glad someone was going to spend some time on it. Of course one of his first questions to me was if we could put Riven in there.....hahahahahahahahaha. I wish! hehehe. I told him that Riven was all prerendered stuff and we don't have the files to try and create it into a 3D format. Instead someone (someone VERY good) would have to sit down and try to physically recreate all of Riven in a 3D format, heh).

Anyways, I learned some very important things: Unit Scale. CryEngine 1 Unit is 1 Meter, so already I have a problem with scale, but not to worry as this is fixed during the importing. To import, I have to take the meshes from Blender....export them in FBX format so that I can then import them into Max. I have to have Max's unit scale set up to 1 unit = 1 meter, and then when importing the FBX file, allow it to scale the meshes up. So that takes care of the scaling problem.

Each mesh MUST have a material applied to it. Again, no problem as FBX allows this to all be exported from Blender and into Max. This is because while you do create different shaders for your material in CryEngine, it wants imported meshes to have materials already applied to them. And if you have different materials applied to one object (think complex objects, like the KI dispenser. It has like 8 materials applied to it), you have to make sure it's set to Multi/Sub material in Max (again, no problem as it get's imported from Blender that way).
You also have to make sure your UV mapping is all set correctly.
THEN you can use the CryEngine Max plugin to export!

One thing you have to make sure of, and this is actually a good thing. You have to make sure that in the CryEngine Max Exporter that surfaces are physicalized based on the material ID, so for every physicalized material (usually this would be the collision proxy) I need to select the Crytek shader for the 3ds Max material and tick the "Physicalize" option. Otherwise the object will be non-solid in Cryengine 3 - and you can't set this parameter in the Sandbox 3 material editor.

The reason that this is good, is that I found in the Unreal SDK, meshes that are terrains when imported: you fall through. Meshes that are imported have to be "closed meshes" to have collision on them. You can have terrain that has collision, but it has to actually be MADE from scratch in the Unreal SDK, as there is no way to import terrain in to the terrain editor in Unreal.

So chalk one up for CryEngine: I can import terrain meshes and walk around on them just fine! Whew!

Okay, I've already started on this, and for now, I'm doing the Ahnonay Linking room again, as it's small and will be easier and faster to do at first, just so we can see how things go and look. If I'm successful, I then move on to the Gahreesen Exterior, as I already have that half way prepped in Blender anyways. I'll come back in a few days and post my results.

btw - just incase people are confused: you can make stuff from scratch in the CryEngine SDK. The reason I'm having to jump through hoops and import is because I want to take what was already made and get it in there so we can see what we can see.

[dendwaler]

Thx for the report Andy, very interesting.!
I follow your findings with pleasure.
You realy stimulate me to take the efforts to learn this.

[Floydman]

Of course one of his first questions to me was if we could put Riven in there.....hahahahahahahahaha. I wish! hehehe. I told him that Riven was all prerendered stuff and we don't have the files to try and create it into a 3D format. Instead someone (someone VERY good) would have to sit down and try to physically recreate all of Riven in a 3D format, heh).

You are familiar with The Starry Expanse Project, right?

[Aloys]

I was about to mention that too, but I don't think that they intend to share their 3D assets; and I don't think either that they'll move their work to CryEngine.
Which only leaves us with one option: convince Cyan to give use access to the original Riven assets! (Rand mentionned in his Mysterium interview he still has all that.)

[andylegate]

@Floydman: Yes I think I've seen it mentioned, but have not really taken the time to look at their project. I do wish them the best of luck however.

@Aloys: After being shaken up by yesterday's earthquake (like Hoikas, I'm here in the SE part of the US in the state of SC, my house gently shook like it does when the washing machine is on it's spin cycle with a heavy load....only the washing machine was not running! hehehe. I was like WTF? and about 15 minutes later, my wife came running in here and said that Virgina had been hit by a earthquake. I used to feel them all the time as a teenager in Naples, Italy.), I got on to skype and was talking to Mark. I asked him about the Riven assets, and he told me this:

[8/23/2011 2:42:05 PM] Mark DeForest: We have all the assets - including the models and textures - However, they are all on a defunct server with no working workstations.
[8/23/2011 2:42:55 PM] Mark DeForest: made with a defunct graphics application

So sound like they have them....in a locked box....but with no key to open the box.....heheheheh.

Anyways, I had sent him links to the CryEngine download and that demo video, he was impressed too, and he sent it to Rand:

[8/23/2011 2:49:25 PM] Mark DeForest: Yup I looked at it - pretty impressive. I did pass it on to Rand - I don't know if he saw it yet or not.... But all these really nice engines coming out and they will just be used in shooters where you don't have time to look at anything, lol.

I asked him both about Cyan possibly using CryEngine for the new game concept that Rand talked about during Mysterium, but also us fans putting Uru into CryEngine:

[8/23/2011 3:05:46 PM] Mark DeForest: buying the commercial licenses would be the small part - staffing a project for 1 to 2 years for a triple A game would be the budget killer... And right, putting Cyan content in different game engine is treading on thin ice, I am obviously not the legal person but you should contact Tony.
[8/23/2011 3:07:37 PM] Mark DeForest: I would think that just taking a small area as a proof of concept (or playing around) is probably ok, but making a more complete working version is where the ice starts to get thin... IMHO

I told him this:

[8/23/2011 3:09:13 PM] Andy Legate: Was afraid you were going to say that. Reason I asked: We (the fans) are looking for a way to put Uru into a new game engine. The results from the Unreal SDK test were....outstanding. Ahnonay actually looked better and was right up there with modern day games in many ways. So now it was CryEngine's turn. But....
[8/23/2011 3:10:12 PM] Andy Legate: the whole idea was to find something that we could use for free, and actually have a way for Fans to make new content for free too (which they can this way), but of course, fans would demand the core content. Well this is a fine pickle. Sigh.

And he replied:

[8/23/2011 3:13:30 PM] Mark DeForest: Right, that is why I was saying before - "what you are doing would not be MOUL but something completely different" - I would think that if y'all are going to create something new it should be entirely new (or very loosely based on MOUL).
[8/23/2011 3:25:58 PM] Mark DeForest: The content is something that a lot of people have glossed over with the open sourcing of the plasma engine - the content (IP) is not open source or public domain and is still in Cyan's rights - and unless Cyan defends it's IP then it could be lost to anyone. The thing I don't understand is if you are going to put so much effort into something, why wouldn't you create something that is new and original - and the fans that want original content can go to the MOUL servers.

I told him basically: "I'm not worthy." heheheheh. But seriously, the idea is not too bad, come up with a MOUL Mark 2. A place where people can go after going through all the original content on MOUL:a. Mark 2 would be based upon a different game engine (as long as the content was freely distributed, etc, etc, the license is free and no legal problems, as long as the content is also original and not taken from MOUL, etc, etc).
I mentioned to Mark, that yah, tis a very cool idea, but that there are groups which seem to be Cyan guys that only want content from Cyan, and then Plasma Guys that do not want to abandon Plasma (not bashing any of you here, your dedication to Plasma is very understandable and all the hard work you put it, hey, it's hard to just walk away from it. I really do understand that.). Nor did I want to be accused of trying to drag people away from Cyan and MOUL. That we do have fans that get rather.....uhm....."defensive" about things...

This is what he told me:

[8/23/2011 3:40:31 PM] Mark DeForest: lol - all the fans are of rosie dispositions -- The fans that won't accept anything unless it comes from Cyan you have already lost. The Plasma Guys are doing it for a different reason which is to incrementally enhance MOUL. And there is your camp which is breaking ties with Plasma which breaks ties with MOUL - which is a tough road but you are striking out in a new direction. IMHO, you need to embrace that or join one of the other two groups. BTW, if you create something new and exploratory, I do think there are plenty in the broader Myst community that would flock to help you.

I don't know about his last sentence....seems a lot of people have taken off to parts unknown over the past year or so. I can think of some people that would be open to the idea, if not out right excited about trying to do something like what Mark mentioned.

I have to admit, the idea does appeal to me: Have a different place that uses the CryEngine with content made by Uru fans in it. Free and open to all, with a way for anyone who wants to can make content too (hrrrmmmm kinda sounds like Alcug Shards already! hehehehehe), but is simply using a more advanced game engine to try and achieve some things that Uru and MOUL are lacking. Nothing official, nothing demanding, just a relaxed place where people can create (just like other shards), but using something new to do it.

I'm still going to do the Ahnonay Linking Room as a proof of concept for now, and nothing is written in stone of course. Even Mark admitted that and told me to send what I make to him and he'll show Rand. Worse that could happen is, well Rand going: "That's nice." and that's it, heh.

Anyways, anyone interested in possibly looking into this, just contact me through pm's or email. I've learned quite a bit about CryEngine in just 2 days and it's just wild.

[Aloys]

Thanks for going info-hunting. Mark definitely has a point about original content. I am interested in porting some Age to another engine just for the fun of it; but as a general rule I totally prefer new content... And I guess most people do.

About the Riven assets:
Back then Cyan used Silicon Graphics workstations; although those are old and obsolete it is nothing that can't be overcome. Taking the hard drives and putting them on a different machine is doable. If worst come to worst, you can find working SGi stations that'll read those HDs on eBay for almost nothing (litterally less that 100$). Also, those assets were created using Softimage 3.5 in 1997. And again that is a bit of a difficulty, but nothing impossible. Softimage also existed as a Windows NT version and it could read the original SGi/Irix files. And it just so happen that I preciously keep my CDs of that version from university days. If Cyan give me access to those hard drives I'll have the files up and running in a modern PC environment in no time. (ok, almost)
So, as always where there's a will there's a way. Obviously a bit of money and time does help too; but my point is that it is technically doable, and not that difficult. The only real problem is the live action videos of the characters; but again there are ways around that. Basically it is possible to port Riven to a modern engine. Of course it'd take work, you don't go from pre-rendered assets to real-time overnight, but the point is: the original files are here and can be ported. It is doable.

The other, more important, point is that it would make quite a bit of sense business-wise.

Riven is a gorgeous game; even by today's standards. Have it run on a modern hardware in realtime (and on consoles too) and it will definitely sell. And more than realMyst (which came too early), if only because it'd look much better. Make it a nice edition with some simple bonuses so that it doesn't look like a cheap port (ideally have an hint system too), and there's definitely a market for that kind of game theses days. Adventure games with high-production values are making a come back lately, supported by indie games (Machinarium/Limbo) and episodic games (Sam & Max/ Back to the Future). Riven is the cream of the crop of adventure games, and it has a proven track record. Also it is more interesting that Myst iOS, because it more than a simple port. It is something new, it's a new real time game and with a level of visual quality than hasn't been seen before.
I haven't thought all this through; but in all objectivity there's definitely a solid business case here. It's a no brainer. It would take ressources, but far less than a brand new AAA game; so it would not even need to sold a lot to be profitable; and it'd be a great way to put Cyan back on the map.

Better: The next generation of consoles is closing in, the new Nintendo WiiU will be here next year, and the next Sony/MS probably in 2013. Those will have tremendous graphic power, and having RealRiven on those at launch time would be a perfect showcase/killer-app... (heck, in a year or two it might even run on iPad3 or 4.). So the timing is just perfect. The technology is ready, the market is here. Start develloping today, ship in a year and a half (and then run to the bank ).

[Chacal]

If Cyan give me access to those hard drives I'll have the files up and running in a modern PC environment in no time. (ok, almost)
So, as always where there's a will there's a way. Obviously a bit of money and time does help too; but my point is that it is technically doable, and not that difficult.

Then give those files to Dustin, I suspect they will be ported in a few weeks' time.

Thanks Andy. I think Mark sums up the current situation quite well.
There is one reason why we like Cyan's Ages so much: we may have good Writers, technically, and a few good artists.
But Cyan had guys like Stéphane Martinière.

I am interested to see the results of your experiments.

[Wamduskasapa]

Thanks Andy. I think Mark sums up the current situation quite well.
There is one reason why we like Cyan's Ages so much: we may have good Writers, technically, and a few good artists.
But Cyan had guys like Stéphane Martinière.

I am interested to see the results of your experiments.

I also am very interested in seeing what you come up with..

[Floydman]

But seriously, the idea is not too bad, come up with a MOUL Mark 2.

This is definitely something I wish would happen. I think the original Uru storyline has been stopped and started and re-hashed so many times that to continue it not only limits us, but it makes for a very bland story. Yeesha holding the bad bahro at bay for years at some unknown location just doesn't interest me.

I've thought before that an "Uru 2" should not simply be a continuation or a branch, but a reboot, sticking to the Uru-ish idea of "What you've seen so far has just been a video game, but this is the real thing." Then there are no feet to tread on. Then we could create a story around the D'ni cavern without having to worry about things like, "Oh, but in Uru this happened, so we can't do that."

[Chacal]

You know, the whole story of Uru ( the DRC's restoration effort) is only a brief and minor part of the history of D'Ni and the cavern.
We could have a new story 50 years before or after.