Comments - DefrostGames

Thanks (yeah a bit late response here)

Custom was necessary, as we want to be able to to VCR style time maniupulation, Unreal Engine and Unity are highly optimized for normal rendering which means none of their systems can really go backward, not to mention that havoc or physx would be absolutely impossible.So animation blending,Particles system, physichs amongst other things are highly optimized in a normal engine to go forwards and therefore can't really go backwards in any easy way just getting our particle system to go backwards properly while being 100% on the gpu was a hassle :)

We started developing the game for XBLIG so the game has always run on the 360 nowdays however our business contacts tell us its not worth the time to make a XBLA verison as sales there are dead. So we are thinking of realeing it in episodes on XBLIG just to support the community that helped us in the beginning. And show off that you could do good stuff with XNA ;)

We were thinkign that perhaps we should have emphazied them even more, For example the split screen mode we run in the last trailer and perhaps less story and more gameplay action :) What do you think ?

yeah we got a bit scared when reset announced but by that time we had revealed our first trailer over a year before so not much to do if they choose to do a similair game. But after seing their actual gameplay footage (not the gameplay trailer) we are not really worried anymore.

We are totaly in agreement with you about first person platforming. Thats one of the major reasons we made Project Temporality a Third person game instead. In the end the platforming is never very complicated, especially since if you miss a jump just rewind time a bit and redo it :)

Great to hear that you are enjoying it:) After near 4 years development cycle its frightening to go out and see what people think of your work.

Nope we didn't learn about prometheus until we released it. But booth this and Prometheus is Derivative from cursor 10. However while prometheus copied cursor 10 straight off(btw I did like it when i found it and played it rhough so I'm not trying to diss it it. It's just a different game). With no rewind and all clones having to restart from the beginning of the game. We instead wanted to fix what we perceived as issues with cursor 10 and a huge issue with it (and promethues) is that you have to start from the beginning every time and the puzzle becomes more about exploring the world than solving puzzles + you got a lot of pointless repeating. By changing so you could rewind time forward and backward and spawn a clone at any time and let it run for any time (promethues had a fixed time per clone) opens an entire new world of puzzles and possibilites so that when and where you spawn your clone matters alot.

So while we now they look Mirilla in a trailer they play very differently. The thing about temporality is as one of our let's players said "You really have to play it to understand why" and that sums it up. The differences in how you think and act are huge even though the visuals might be similair.

We will definitely try to get it on as many platforms as possible. As we are greenlit it will come to steam too :) We are still in the launch period now however and will start contacting other distributors too but for now Desura is the recommended version as we don't know which platforms will let us in.

Yeah Texture map baking is a huge issue for most games thats why a lot of engines try to bake what is called a Basis function (a function that can describe incoming light from all directions) so that they can at least add normal mapping details. But that doesn't change the baked shadows from looking blurred.

Thats why engines try to do as much in real time as possible :)

Last of us is an example of really nice baking they used a really simple basis function and then they made a game that was dependent on soft light therefore there was very little problems with crispiness of the shadows:)

We're still in progress but it will be released on Desura and some other places.

Ambient occlusion (AO) which SSAO is an implementation of works different from shadows. Shadows block direct light if you look at the top image you can clearly see the shadow line. so the parts of the world that are in shadow are dark in the same color (well not exactly the same but that's due to our bounced light solution). AO instead blocks ambient light this is light that has hit something and bounced so you need booth of them. You can bake AO but then you would need a very highres bake. Even engines that supports some baked AO(UDK) also adds on SSAO to compensate for the lack of resolution. As Sparta 3D is all real time all the time and all our shadows are real-time that wasn't an option.

It's in our own Custom Engine Sparta3D ;)

Thanks we are finally getting ready to start making some more noise :)

Thanks. We spent a lot of time working on them so that the game will feel nice and cozy to play. :)

Yeah thanks we were going for that :)

The movie was about beautiful CG of the sun. And got good reviews because it was beautifull not because of the story ;)

We have been doing a lot of these, we even hired in a Director for the trailer to make it more interesting, but it's tough reddit seems like a bad place to go to unless you are a regular because it seems newbie posts just die. But we are actively looking at all options.

At the moment we do belive a playable build is our best bet and are aiming in that direction.

In fact if you have any ideas on how to attract more exposure it would be welcome :) I think we have a good looking game with an interesting concept, and everyone who ha splayed it so far agrees so it's kind of tough.

Well we do have a steam greenlight account and our own webpage, but everything we post is pushed booth to our own webpage and here. But we haven't mad to much noise of us to be honest. Probabely to little but then again its hard to get noticed out here on the net.

For example in the first puzzle in the trailer the only place you can originate clones from to make it succeed is that specific space between the force fields. Just finding this out is a big deal. Also unlocking us from any specific time limitations also changes the game a lot. We go way deeper into the puzze aspect and mix it up much much more than they did. So it's two very different games. But we understand your confusion and we will later launch a gameplay trailer that will make it much easier for you guys to get into the core of Project Temporality. I can write a lot of text about this but I'm not sure if I can explain it better than i have done at our indiedb page or the aforementioned interview. But don't worry just ask and we will try to explain to the best of our ability. And hopefully the next video will clear up a lot of the misconceptions people have. Till then just ask away.

Alright so now I have played Prometheus. And while I can understand from this trailer where people are coming from they are also far off. At first it seemed like Prometheus was just a Curser10 clone (the same game that inspired us for more about that and the origins of this game I refer to Indiegameschannel.com) Sadly it meant it also contained all the parts that made Cursor10 frustrating. It was also very similar to ratchet and clank a crack in time in that you always start a clone form the same position. But thanks to the epimetheus twist halfway the game grew a lot more on me that was a nice idea.

But coming from the view of Project Temporality Prometheus like Ratchet and clank or all other games featuring a type of recording from a fixed position for a fixed time is a very different proposition. While in the trailer it looks like you always spawn a lot of clones at the same time you don’t do that in normal gameplay. You can spawn a clone at any time with any length of recording which stays there permanently. This by itself make the game very different not only do you create clones you have to think very carefully about where and when.

We opted to go for the pretty pictures approach for this trailer because people have had a trouble grasping the gameplay from earlier trailers. So the first 60 seconds establishes the environment and story setting and then we start showing of the gameplay more. we are going to add a second pure gameplay trailer to make sure that people really get it. Because at the moment people seems to grasp only a part of it. But then again it is hard to show it because it happens mostly in your mind.

Thanks. It's hard to loose track when working on a game and then suddenly you don't have any dead of how it really looks.

Thanks for liking it. We are perhaps working to hard on the game and to little on marketing though :/ We want everyone to have heard of it :)

Re-uploaded images with AA on now :)

yeah we have disabled our FXAA somewhere along the way but since that's an easy thing to fix we haven't cared to investigate it. And the auto down sample for the web doesn't exactly help.

Thanks for the kind comment.

Just saw that the post lost all of it's formatting somehow, I fixed it but I'm sorry for you guys who had to read it this way :/

Some good discussions from the implementation in black ops.
Advances.realtimerendering.com(Siggraph%202011%20Advances%20in%20Real-Time%20Rendering%20Course).pptx
More in depth on energy conversation
Rorydriscoll.com
Sebastian Lagarde of DotNod wrote some interestign articles on the subject too
Seblagarde.wordpress.com
and at Seblagarde.wordpress.com
And
Thetenthplanet.de is a good starting point. Between these their links and the links in my papers and of course the two siggraph course
Renderwonk.com and Blog.selfshadow.com there will be enough for anyone who wants to read up on the subject :)

Not really. Shadows from light works the same way as before you can use PCF,PCSS,VSM or as in our case ESM and you get a percentage of occlusion for this light at this pixel, most of these ways can be adapted to area lights too.
For the ambient light it's a matter of using the correct occlusion formula. A Directed Occlussion (SSDO) is optimal and can be implemented using bent normals. In our case we settle for normal occlusions and instead burn cycles on making it really high quality.
But as using a Physically based lighting method in the end only changes how the light for that pixel is calculated and all shadow/collusion methods currently in use either modulate the end result of the lighting equation or the inputs (in the case of bent normals).
Even if we were doing a full Global Illumination solution it would not really effect the shadow casting for the direct light, however shadow casting for indirect light gets hairy that's why people are investigating techniques like Voxel cone Raycasting(UDK4) Light Propagation Volumes (which uses a really coarse collision for bounced light- Crytek3) or Imperfect Shadow maps (If you use Reflective shadow maps (which is what Light Propagation Volumes was base around) they work out pretty well. I have seen some interesting demos).
So to conclude if you are using Phong,Blinn-Phong,Cook-Torrence(the first really well known PBL),Kelemen/Szirmay-Kalos (used a lot for skin shading) In the end you only change what is calculated as the incoming light to a pixel and a shadow either block the incoming light or not.
If things are still unclear I can elaborate but I am trying to stay low tech and not go into all the formulas here, I did enough of that in my paper. I can link to a few more in-depth articles if there is an interest. For me the big advantage of PBL is not that it's physically based but that it simplifies the artists work and makes values that seemed random make sense which accelerates art production a lot.

Then we will try our best not to disappoint you. And you can be sure we will be keeping up with regular updates on the progress.