I've had a great time working on the project using Google Blocks––its allowing someone like me with no 3D modeling experience to get the ball rolling and learn. After you try Oculus First Touch, it's pretty tough to imagine the Future of Farming scenes without that level of interaction with any object in the scene. In order to achieve this, I had an obstacle with my workflow to get individual pieces to be interactable.
Several efficiencies were found after conferring with a more skilled artist than me regarding making discrete objects (i.e. CDs on a desk) from a scene created in Google Blocks. For flow purposes, it's much better to create a scene all in one Google Blocks session. So for example, if I have a TV remote sitting on a table it's a part of the overarching mesh that I export from Blocks. This more experienced artist showed me how to use Maya to go into the living room scene, select the TV remote using either the edge or face tool and then follow this subset of steps which is mostly the same for both.
1. Use Mesh>Extract, while focal object is selected
2. Go into object mode and select the separated geo
3. Go into Mesh>Combine
4. Select the newly created mesh
5. Modify>Center Pivot
6. Rename
With WRLD, I've found the certain long. and lats. don't offer as much in the way of buildings as other coordinates (34.236137, -77.941537) are the ones I've tried for this shot. Next week, I've got to narrow in on where the Unity World space will be set, and carve space out for my rooms. @Micah brought up a relevant question pertaining to the beginning of our experiences. Namely, should we have a static load screen or should there be a kind of load-in area where the experience starts? For now, I'm trying to make the menu scene a bit like the user is looking at Earth from a satellite and is then able to select a location.
Several efficiencies were found after conferring with a more skilled artist than me regarding making discrete objects (i.e. CDs on a desk) from a scene created in Google Blocks. For flow purposes, it's much better to create a scene all in one Google Blocks session. So for example, if I have a TV remote sitting on a table it's a part of the overarching mesh that I export from Blocks. This more experienced artist showed me how to use Maya to go into the living room scene, select the TV remote using either the edge or face tool and then follow this subset of steps which is mostly the same for both.
1. Use Mesh>Extract, while focal object is selected
2. Go into object mode and select the separated geo
3. Go into Mesh>Combine
4. Select the newly created mesh
5. Modify>Center Pivot
6. Rename
With WRLD, I've found the certain long. and lats. don't offer as much in the way of buildings as other coordinates (34.236137, -77.941537) are the ones I've tried for this shot. Next week, I've got to narrow in on where the Unity World space will be set, and carve space out for my rooms. @Micah brought up a relevant question pertaining to the beginning of our experiences. Namely, should we have a static load screen or should there be a kind of load-in area where the experience starts? For now, I'm trying to make the menu scene a bit like the user is looking at Earth from a satellite and is then able to select a location.
This week we focused on finalizing and fine-tuning gameplay
scripts. We tested deployment to the Gear VR and using the VR remote. We improved
several of the gameplay scripts to increase performance and maintain a solid
framerate. We also polished some game assets with improved shaders and built
out the functionality for our Start Screen/Hub.
Me testing the game in the headset.
We also worked on improving the user experience through the
reticle. We created assets for our reticle to be responsive to the type of
action taken when interacting with an object. We are hoping our reticles help
guide the players’ actions in a non-verbal way.
The game was really lacking audio feedback, so I found
several sound clips for our train from www.soundbible.com
and purchased a few clips from www.audiojungle.com.
Satisfied with what I found, we implemented those sounds, really improving the
feedback and feeling of the game. There were still a few sounds we needed, so I
recorded some sound clips using my Yeti mic. This was fun and new for me. My
cats were really intrigued when I made recordings and kept trying to meow into
the mic.
Now that we’re confident in the gameplay loop, we’re focused
on polishing our assets. The biggest feedback suggestion from Oculus was to
focus on the asset quality, so we are going to spend next week enhancing our art
assets.
Week 7!
This week was mainly focused on the assets that will be used in the final product. The pipeline to create the assets and import them into the editor to work with our homebrew "lighting" workflow is interesting and we've been trying out a few tests.
Here is an example of the lighting working correctly using the material functions.
https://youtu.be/qTDwJ9hd4Zs
Also, there was a big focus on the art side of the assets, and concept art was explored, discussed, and now we are on to building the assets themselves.
This week was mainly focused on the assets that will be used in the final product. The pipeline to create the assets and import them into the editor to work with our homebrew "lighting" workflow is interesting and we've been trying out a few tests.
Here is an example of the lighting working correctly using the material functions.
https://youtu.be/qTDwJ9hd4Zs
Also, there was a big focus on the art side of the assets, and concept art was explored, discussed, and now we are on to building the assets themselves.
Neil D. Short
Project Title: Achilles Heel
Type: 360 Video Series
Genre: Sci-Fi Sitcom
OLP17
- Blog 7
Like
many other OLPers I was traveling this week to the mountains of Idyllwild,
California and a day in L.A. to attend the screening of a film I helped produce in the L.A. Shorts Film Festival. I mostly did remote work via email and Google Hangouts with my production
designer and concept artist.
I
received feedback on my proposal this week and it was good to get questions now
that I can address while still in this development stage. One of the big
hurdles for the demo is the cost to hire the actors I will want and to build a
physical set. With this in mind, and the feedback I received I have decided to
use concept art to create a virtual version of the set in 360 and shoot a few
scenes with temporary actors on green screen and composite them in. This will
hopefully be a way that is “cost/time friendly to deliver the concept clearly”.
It is risky, since I want to use practical sets, but not having the time and
money at this point, I feel this is my best option.
We
have a good floor plan for the main room the series will take place in and my
production designer is creating it in Sketch Up to hand over to the concept
artist to flesh out the look.
Another
question in the proposal feedback was regarding interactivity and how to
provide the viewer with a feeling of inclusivity (agency). My initial plan was
to have the experience mostly passive, but considering the feedback I’ve
decided to add an element of interactivity.
The
viewer’s role is observer through a journalist’s camera placed in the center of
the room. The action takes place around the room in various places and my
initial thought was to block the actors in such a way that they move closer and
further away from the camera based on if they are to be part of the immediate
“scene” or conversation. I’ve decided to add strategically placed 180-degree
ship cameras, built into the wall like security cameras, that are accessed by
the ship’s computer. As characters move away from the 360-camera, I want to
give the viewer the ability to use gaze selection to choose a ship camera near
the characters to spy on their conversation. By doing this, the viewer will
have knowledge of information revealed to one or two characters that is not
shared with the whole group, which can change the perspective of a
conversation. For example, if Drew says something off topic and silly, and his
younger supervisor Samantha tells him to “shut up and go find something to
clean” and he immediately complies, it may seem like he is just following the
orders of his boss. Now, if we jump to a ship camera and listen in on Drew as
he explains to Taylor that the only reason he has a job is because Samantha
tolerates and protects him from their superiors, then the viewer will have a
different context to future exchanges between these characters. At the same
time, if the viewer does not switch camera views they don’t loose out on the
story, they just don’t have as full an experience as is possible.
Next
week’s focus will be on camera tests Unity integration. I’ve been learning on
my own, but like so many others doing 360Film the learning curve vs. time vs.
to-do list is proving to be a challenge. I’m going to reach out and see if I
can find a Unity developer to join the team.
Another short one
this week as I was out a bit in the start of the week…
I took a step back
from a lot of the conversation detail this week and focused on working with my
team on the atmosphere. It's really important that this scenario feel grounded
in reality. If this comes across as comic-booky or arcade-y, a lot of the point
is missed.
The character
(customer/ guest) is now looking really good. We got a great hair sim in there
(using cloth, not hair, but it works!) and played around a lot with shaders to
really give her skin and clothes a polished look.
The environment is
near complete. It's close enough to start playing with the lighting aspects.
Lighting is always tricky in VR. Too many dynamic lights gives some huge
performance problems. Too few lights leaves your scene unappealing. And baked
lighting can take a long time. The goal is to use all dynamic lights, but
sparingly. We'll bake it when we're happy with the result of the dynamic
lights. Luckily, our NPC isn't going to move much, so we don't have too much of
a concern about shadows, etc.
As for the
conversation, I've played around with that some, but honestly just didn't have
the time to do as much there as I wanted to this week.
Sneak peek!:
Week 7: Games For Change & Next Steps
Week 7 has been a bit of a whirlwind for me. I know that I'll have around 6 weeks left to polish the potential submission of our game Spark a Memory. Most of which comes from demoing and presenting the game to the rest of the VR For Change day, receiving good feedback and interest. Wick and I took the stage to represent out team at the VR for Change to show what we were able to accomplish with Spark a Memory and field a few questions from 5 judges.
One of the most challenging questions was about our next steps and what would our team want to do if we would continue to develop our prototype. I remember explaining that most of the nodes (Neurons) for now are procedurally generated and random. So far we had made the game so it could primarily show how neurons connect to each other and send signals. In doing so the player would complete a neural net that would unlock a memory which is essentially a 360 video. However in order to make the game something full and engaging we would encourage exploration.The first step would be to work the system we have now to design a small series of puzzles to present a challenging element rather than 'ooo this is cool but what's the point'. Implementing this might prove to be an obstacle. I've never built active puzzles before but that's what game jams are about, learning new things. I'll have to spend some time sketching out simple puzzles that play with a neurons ruleset of being able to heat up or cool down the connected cell based on the connection's signal and strength. The current system allows us to demonstrate this but is pre-formed neural net every time the game starts. Once I'm able to place them and have them connect to each other this way level design will actually be a possibility. We'll see what's feasible.
During development I pushed for the game to be able could run on both the Oculus Rift and the Gear VR for it to be accessible to a larger population of players not limited by their systems. It's easy to go hey let's just run this on the rift but it's important that we think about our potential market with this game. As it stands right now our main hardware guy Scott made a great effort to make the game as comfortable as possible in terms of the game running at 60fps on both platforms. He was excited that our play experience had no hick ups and our demo didn't make anyone sick. One problem is the Unity Video Player make our Gear version crash after the video scene ends. So we'll be buying an video player asset for that. The other issue is compressing the 360 video diminished the quality of it completely on Gear. It was a quick fix to avoid large apk files but considering that the video went from 1080p to 480p you can imagine that production value has also diminished. This is a priority on the list of things to do as well we add memories and level design.
This project combines the VR with 360 video in a unique way. There's a little bit of science, exploration, and wonder in it as well constructing the kind of impact we want our game to have. We've been able to accomplish so much in just a weekend, we'll have to see what we can do now that we're not all huddled in the same room for a couple days. I look forward to giving you more updates. If anyone has any solutions to the video compression issue I would be very grateful for any advice.