Hey to anyone who’s stumbled upon this, welcome to my blog of my weekly going ons throughout my university studies, hope you enjoy the (mostly bumpy) ride.
This week I decided it was about time we finally have some form of rough previs for DayDreams. I wanted to make sure it was done as soon as possible so I put it together myself quickly and quite roughly. Rough as it is though I’m sure it’ll go a long way to help with the final animation stages and the overall visualization of the piece.
Over on the Research topic side of things I was able to, with Phils help figure out what was wrong with the way I was doing imf_copy. However after we were able to get that working I found that for some reason my render passes had become tinted red for some reason. When I reran the process at home I found this time they had become tinted blue, I tested the files in every way I could think of in an attempt to fix them. In the end I decided (mostly out of confusion and frustration) to attempted to uninstall and reinstall Nuke, which for some reason worked. So now I have an EXR render with all the passes within working and another with deep data, but of course I’ve found myself a new issue. When trying to view this deep file in Nuke through a deep recolour with or without the color render it attached it crashes Nuke. So stay posted for progress then the next issue there.
This past week I’ve continued work on both the group and research project. I feel like the group still needs a bit of fire in their belly’s over this project and to get motivated which we will hopefully be able to do in the coming week working together in class.
I’ve been continuing with some dynamic simulations trying to figure out exactly how we should be achieving the effects we’re after. I stumbled across a video earlier in the week researching these effects that I think we may be able to take on.
This effect makes use of Maya’s Pond System and some nParticles to create a nice result I figured we could use for our piece, so I got to work on the same kind idea and was able to come up with this.
I’ve been able to come up with a method which I believe will work for our wake effect and that I can confidently reproduce. This is just created using the Wake Emitter on a Maya Pond which works quite well for what we need. Unfortunately in the example vid below I didn’t position the camera so you can see the horizon line so it looks straight out of interstellar.
As far as the research project goes. Tiled EXR stop, why are you not working imf_copy, what is going on. Stay posted for updates there.
The past week we have continued work on the group project. I began some research into some dynamic water effects for us to implement into our piece. The two effects my attention is currently focused on is the wake in the water created by the characters flying over it and the water splashing as one of the characters drops below. So far my tests have been centered around using Maya’s Oceans and Ponds as well as a little bit of Bifrost. Primarily I’ve been trying to use Bifrost to achieve the splash effect and Oceans and Ponds to achieve the wake. Using Oceans and Ponds I was able to get a wake effect working however is wasn’t without its issues. When using the effect on a larger scale as we will need to for our Ocean I found I ran into issues in its results, particularly the need for the for a larger wake emitter to be used to actually generate any effect. The larger emitter meant it would create a strange effect directly where it was placed in the ocean as you could see the water directly displaced by it. I also attempted to use Oceans and Ponds to create the dropping through water effect but found that it didn’t work very well with the water surface not being able to break it looked like some kind of jelly.
Attempting to create the splash effect I used Bifrost and tested dropping a collider object into a container of water. This created a much more dramatic and dynamic effect but found issues in using this method as well. Bifrost takes a considerable amount of time to simulate even for an effect with as little water as my Test scene. For our piece the effect will require much more water so that the effect does not visibly end before it should, this will only increase the time it will take to simulate and render the effect. As I write this I am still searching for methods of achieving these effects and will hopefully be able to produce some effective results soon.
I have been unable to dedicate a lot of time to my research project this week but I was able to make a small amount of progress. Now attempting to render the scene using the EXR frame buffer passes I was able to get all the passes to work. However once I attempted to then render them as deep data I ran into the issue of them being tiled EXR not Scanline which is what Maya and Nuke both require if you wish to view it. So moving forward I will need to find a solution to this problem either using imf_copy to somehow convert the images from tiled to scanline or use a different renderer.
This past week for our group project I’ve really tried to step everyone into gear and get some more progress happening so the project may progress. Ben and I decided it would be greatly beneficial for us to have a discussion over Skype and work on creating a effective Gantt Chart, this was something that probably should have been organised earlier which is the fault of Ben and I. However I do think that now that the plan is in place we will be able to much more effectively work towards our goal. For the past few days I have personally been working on the 3D character models for the dream sequence. It took quite a few hours to get right and I got feedback on the topology quite a few times to ensure it would be suitable for rigging and animation. Steve and I were also able to get into contact with Dan and follow up on some stuff we needed to discuss but hadn’t yet got the chance because he hadn’t been able to make it too class. We were able to assign him work to be completed for this Tuesday and ensure he knew what we were after.
During the week I also worked a bit on my research project, however this work turned out to not of use to me as Maya now has new features I was unaware of. That work was setting up my render layers, separating them into a render layer for each object and pass on that object. This took quite a bit of time and I did run into a few issues with some of the layers. However the next day in class we were given a tutorial by Phil discussing Maya’s new (or at least new to me) render pass method. Now that I understand how to use this method it will be the technique I will use moving forward on the Project.