08-08-2018 07:03 PM
08-12-2018 05:56 PM
Happy Mid-Point!
This week I continued working on the prototype. Starting to really feel the pressure now.
That's it for now, I'm off to finalize my mid-point check in submission.
08-12-2018 06:06 PM
This week I'll start with a joke...
What do you call a
VR project that won't run? You call it a V-aren't 🙂
There has been
considerable time devoted to testing and programming this week. It's particularly exciting to start adding elements to the prototype that have only
been in our imagination. It's definitely fun to finally experience it!
Overall it has been a busy
week for sure, balancing working on the prototype/project and the Midpoint
Check-In document - but overall very rewarding!
Until next time, remember...
Keith Patterson
"Make Every Day A Beach Day!"
08-12-2018 06:55 PM
Blog Post # 6
This week I finished texturing the daughter and placed her
in the hospital scene. My partner has been tweaking the animations and working to make them feel more realistic. The scene feels very surreal and takes me back to that
final moment. It’s discomforting watching the two characters breathe in the
scene, as if watching from the outside and intruding on a personal moment.
Additionally, I finalized the mid-point check-in and budget for the
experience. I realized I need to start thinking about sound effects and get
some sound in the experience. I also need to start recording my voice-over
narration. I worked on fleshing out the narrative script more, so the memories
flow better and have a solid foundation.
The past few weeks have had a few learning curves, but now
that we have a stable workflow, I think the last leg of this journey is going
to by fast!
08-12-2018 08:10 PM
08-12-2018 08:25 PM
First things first
Before we started shooting our “Fire Together” VRE, we thoroughly researched which cameras to use. We fantasized about using Funique (after taking a sneak peek at “Your Spiritual Temple Sucks”). But unfortunately, that would have been as affordable as a trip to the moon. Reading online reviews and talking to industry veterans helped us narrow down our list to the following:
Camera test day
Our friends at Radiant Images were able to accommodate a camera test day for us, and we brought some of our talent and prototype wardrobe to get an idea of what our footage would look like when captured by different cameras. The most important things we needed to consider were as follows
1. Ability for stereo capture. We believe that 3D is absolutely essential to make viewers feel like they’re actually on stage with our artists.
All cameras can do stereo, but we found the effect from the V1 Pro to be the most convincing, which we attribute to its ideal distance between sensors. Even the giant Jaunt has so many sensors that the distance between adjacent ones is closer than are human eyes, which makes reconstructing realistic pictures quite challenging.
2. Overall picture clarity and color reproduction. This is important for any camera; VR cameras, having built-in ultra-wide-angle lenses, suffer in this department quite a bit. We also considered shooting on green screen, which requires really good color reproduction to work. We subsequently decided that investing in a set on Chicago’s biggest stage was a more affordable.
The V1 Pro was the winner here, too. The Jaunt was just slightly behind, and the Kandao R was significantly less clear and had a lot more noise due to smaller sensors. We actually preferred the picture clarity of Insta360 Pro, which was half the price.
3. Resolution and frame rate. Like a fine wine, we wanted our VRE to age well with the emergence of new technologies, so it was important for us to shoot in 6K or 8K. In addition, most photographers agree that 50-60 FPS is the lowest you can go with fast movement. But as frame rate increases, so does your post-production cost, especially if your workflow includes rotoscoping.
All the cameras could do 8K resolution, but only in certain combinations with frame rate. The limit, as we learned from talking to manufacturers, comes from the bandwidth of the DSPs. The Kandao and V1 Pro could only do 4K at 60 FPS stereo. The Jaunt, having been built from the ground up, didn’t have this limitation. However, we never had a chance to experience 120 FPS 8K capture due to issues in their cloud offering.
4. Cost. We are mostly self-funded; this was a major factor.
Cost varied quite a bit between the cameras. Renting the cameras typically costs around 10% of the camera cost per week. The V1 was about 3 times as expensive as The Kandao, and the Jaunt was 3 times the cost of the V1 pro (plus you have to pay for their cloud stitching to get good results, and that grows linearly with the frame rate). The Jaunt quickly fell out of our budget, but we were able to accommodate the cost of the V1 pro.
What we learned
We learned that a perfect camera solution doesn’t exist. Even putting cost aside, going with top-end cameras like Jaunt or Funique locks you into their post-production workflow, so you lose a lot of flexibility in your stitching workflow. As we’re working with Image Flux on stitching, we introduced a lot of processing on pre-stitched footage, and we don’t think we could get results of this quality without those steps.
Results and inevitable compromises
We’ve just received our first samples of what a clean stereo stitch would look like. Did we get what we wanted? Mostly, yes. Stereo quality is very believable, and unlike a stitch with consumer tools, you don’t have to struggle to focus on different objects. The overall picture quality is great (better than many other pieces shot on the same camera). At 6K, the resolution still provides crisp details with headsets like Samsung Odyssey or HTC Vive Pro (we don’t have a Pimax to test). The cost will comprise a significant chunk of our budget.
The biggest compromise came in the frame rate. And that will be a topic for next week’s post!
08-12-2018 08:40 PM
08-12-2018 09:46 PM
08-12-2018 09:57 PM
#MURDER- Blog 6
By: Nir Netzer
This week was busy preparing for the Mid-Point Submission, but we were working towards it for 3 weeks now. In addition to the Launchpad template, we are submitting a pitch deck and a script/flow-chart for the prototype. The pitch deck is telling more in-depth about the characters, the story, and the concept. The Script/flow chart is an interactive PDF presentation. Each slide tells a scene in a paragraph and has buttons with options to continue. We will use the flowchart as our guideline script when filming the prototype in a couple of weeks.
Overall I am proud of what we are submitting. Azra and I met this week to work on the Launchpad doc together. I think Katherine and I had about 5 hours on the phone in total this week. Working on the flow chart. It was a lot of work to flesh out the concept but I think we did a good job.
Next week Azra and I will start looking for a villa for shooting the prototype. I also have a meeting with someone from Samsung. They have a new VR camera and they would loan it to us in Making360 if we will try it and upload tutorials to our Youtube Channel. I will test it to see if I would want to use it to shoot the prototype. I was shooting with the Kandao again when I was in Taxes last week on the CBS production. The camera overheated a lot, which made me doubt it for this project, because we will have to film a lot all and for long periods of time. I have access to the Insta360 pro from VRenetic, but this camera has internal fans which can disturb the dialogue. Therefore, I’m looking forward to check out what that new Samsung cam has to offer. I also want to budget the prototype next week. It will be an investment out of my pocket but it’s totally worth it. I was looking forward to an opportunity like Launchpad for a long time and I am excited for it.
08-12-2018 10:23 PM
Still splitting my time between these two passion projects!
It feels precisely like a midpoint in both!
LIVE THEATER IN VR
We completed the last workshop session with actors Carl
Hendrick Louis (Broadway credits: 1984, The Cherry Orchard) and Kareem Lucas
(Black Is Beautiful But It Ain’t Always Pretty). These are two enormously
talented actors, so getting to work with them was a real treat. Louis and Lucas
chose a scene from The Wire. We hadn’t worked with a TV scene yet in VR and I
was curious how this would play out. I suggested to my team that instead of
running the scene in the Theater, we run it instead in a more realistic
environment; an apartment which might feel more intimate.
By the second time we ran the scene we knew the environment
was wrong. We began hopping around the apartment to exterior environments.
First a patio deck off the apartment; next a deck area; and finally a rooftop.
Oddly enough, this last environment we landed in, the rooftop felt the best to
me and the actors agreed. I felt that the actors’ work resonated best in this
fairly bare environment.
Some interesting reflections the actors made regarding how
they felt embodying their avatars. “I didn’t
stop to look at who I was…I only cared about what my scene partner looked like.”
Carl Hendrick Louis. Another highlight of this session was watching Kareem
discover flying. Despite the heavy drama of this scene; there was a lot of
laughter during our session!
METROPOLES
I spent a lot of time with my UX designer this week. We introduced
some new models to the street that we’ve been working on. Feedback is critical
for the player to receive which game designers obviously know; but I am still
discovering how to implement this and to do so in the lyrical way for how I see
it fitting with the entire game. Therefore we tried some different types of
lighting tricks for selection. I’m really excited about how this looks and
feels. We also spent time working on some of the aesthetics and colors in the
game. I really believe color is important to how the player will feel. While
this felt like a tiny task and we have big pieces to accomplish, I’m excited
about every detail getting revisited.
Next up for METROPOLES: Since we’ve started over we eliminated
some elements that are important for any game and we need to reintegrate them.
One of these elements is teleporting. Another element is the mechanic of adding
a property. We are trying to do away with another button on the controller to keep
the learning curve as simple as possible so that the player can focus on the
story.
08-12-2018 11:18 PM