Our 176-cameras bullet-time studio, 2022 edition!

*Disclaimer: part of this project is sponsored by Intel/Asus. And yes, they helped us achieve better performance with our system via the ASUS Zenbook Pro 16X OLED

Over the past 12 years, we went from 24 to 176 cameras, with a total of over 400 cameras at the studio. This 360 degree station is made of 22 clusters of 8 cameras each (=176 cameras total). Each cluster contains two raspberry Pis (4 cameras on each), one switch (network relay) and a bunch of bricks to power up everything. There are only two cables coming out of each camera: a usb cable for the data, triggering and settings; and another one for the power.

Full equipment list

  • 176x Canon SL1

  • 44x Raspberry Pi (3b and 3b+ mix)

  • 22x small network switches (1GB)

  • 1x master network switch with 10GB uplink)

  • 1x USB-C 5GB network interface

  • Network cables

  • Power adapters

  • 10x LED strips

  • 1x ASUS Zenbook Pro 16X OLED

  • Xangle Cameras Server software (of course!)

(see our gear page and budget calculator for alternative configurations)

What’s the process?

When I press the button (a wireless powerpoint bluetooth clicker), all the cameras are firing at the same time. Then the files are transferred to the main computer (ASUS Zenbook Pro 16X OLED), and the software crunches the files to apply digital calibration and generate a mp4 file ready to display within a few seconds. This is all automated of course.

2022 performance improvement

We’ve been measuring the performance of each process (files transfer over the network, jpg extraction from raw files, digital calibration, resizing, mp4 generation) for a while and a big challenge for us is to always improve the timing of each step. We’ve been able to shave a few seconds recently by improving our network components, but, our biggest upgrade this summer is this new ASUS Zenbook Pro 16X OLED. It drastically improved our studio work for 3 main reasons.

  1. Performance

    Did we just go from 40 to 30 seconds in the overall processing time of 176 RAW files? That is a massive improvement. The faster the results, and the better we can keep our creative flow from one shot to the other. This computer is powered by a 12th Gen Intel® Core™ i9 processor and is able to crunch the raw files and apply digital calibration within just a few seconds.

  2. Design
    But a fast CPU doesn’t do much if it’s throttled down due to overheating, which is something we have experienced way too many times over the last years. And that’s a major design feature we have in our hands now. We constantly spike at 100% CPU during the calibration process but the overall temperature is well maintained as the Active Aerodynamic System Ultra generates optimal airflow for max cooling.

  3. OLED Screen
    When we review the images, we need to make sure that the details are great, that the focus is right and that the colors are accurate. As we have all of that on this computer 16-inch 4K OLED screen, we save a bunch of time by doing every step, including the post-production on a single computer.

Here’s the final result in 4k:

Additional pictures from the project

Created in Montréal by Kim Henry, Eric Paré and the Xangle team, 2022

Previous
Previous

Canon R50 trigger test for bullet-time / photogrammetry

Next
Next

Basic support for Nikon cameras