On Monday the 11th of may we gave a presentation to several DVTG students, Evert Hogendoorn and video artist Kaisu Koski. Having presented the first working prototype on friday we tried to push for extra features like being able to switch 230W power on and off with an programmable 4piece power-switch. Also we installed a Proximity sensor with an Arduino interface to give an alternative to optical tracking.
The presentation is a start of a six week project in which the students will use the Performance engine to create a theatrical piece that will be presented on the 25th of june.
The total presentation is about 30 minutes.We are working on a shorter and subtitled version!
The order of the presentation is:
- Introduction of research context and Performance engine
- Example of controlling Video and Audio with WII-mote
- Example of controlling Video and Surround Audio with color-tracking
- Example of using a Proximity sensor for controlling Video
- Example of using Reactivision ficeudal marker tracking with Unity 3D engine to control virtual 3D object
- Example of same technology controlling the virtual camera
- Example of same technology controlling scale and ‘time’
- Explanation of the general Performance engine setup
- Explanation of the Statemachine
- Giving the assignment to the students
We’re working on implementing the proper protocol to all the systems, both for the INIT-list and in- and output streams. The aforementioned 255 item per list bug is now solved by sending several packets that never exceed this 255 item maximum. This has however again be implemented to the Audio-, 2D- and 3D engine. Also the <TAB> separated structures are vulnerable to mistakes, which results in messy input lists in the Statemachine.
Continue reading Testingdays Performance Engine — Day 8
Today at three we gave a presentation for a few students of the Theatre Education department, Evert Hogendoorn the head of DVTG and Sabine Dehne temporary head of the Actors department. Tracking seems to work stable but making connections through the Statemachine is not yet presentable. Surround sound and 2D engine do their job but Unity 3D and the adapted Miyu patch seems to crash about every 10 minutes!
Continue reading Testingdays Performance Engine — Day 7
A traditional setback in the development process has happened. Connecting all the systems together shows some fundamental problems in the way they communicate. Sil has designed a custom protocol for sending and receiving data over UDP, we have named SIL.UDP. Every system communicates in two distinct steps; first it sends a INIT file explaining the amount and names of all the variables that will be sended, secondly it will start streaming these variables. The structure of this information uses <TAB> based separation for hierarchical ordering. We have discovered however that MaxMSP is becoming unstable as soon as the INIT or stream contains more than 256 characters. Workarounds in MaxMSP for this don’t seem to work. So we need to find a solution for this.
Continue reading Testingdays Performance Engine — Day 6
Slow day today. Everything seems to come together but of course as soon as they are new bugs appear.
Ferdy has started to test again the Reactivision tracker. We have made a quick object with several markers to fool around with. Tracker works fine under optimal lighting conditions. Point is that Reactivision claims a webcam, so we need to have one reserved for this particular tracking only. For this reason we have removed the IR filter from one of the top-view camera’s. One can be used for the Reactivisiontracker and the other for the rest of the trackers. The IR-led tracking we might do only from a frontal position.
Continue reading Testingdays Performance Engine — Day 5
Had one day off because of Queensday, a national celebration day. It’s now friday and the last day of the first testsweek.
We have worked on setting up the video hardware setup. We unfortunately have to use two analogue EDIROL video-mixers (v4) because of financial reasons. Having a lot of computer as video input sources we have bought 4 Konign (cmp-telview2) Digital to Analogue video converters. The result is very poor quality. Needing to show also computer desktop GUI’s with readable text we now have a serious problem. A digital video-mixer with VGA/DVI input (prefereably with 8 inputs and at least 3 outputs) is completely unafordable. We are now working on a PAL resolution (720×576) with CVBS color (4 pin S-video). We have to solve this later, for now it is ok to test with.
Continue reading Testingdays Performance Engine — Day 4
Today started to hang the beamers and camera’s on the ceiling. We have now a projection on the wall and one on the floor (effective surface about 3,5 x 2,5 m). The camera’s are two webcams (Logitech Quickcam 9000 Pro) with powered 3 meter extension cables. we put an Infrared filter in front of one of these for the IRLed tracking. This filter is nothing else but a layer of foil form an old floppy disk and a developped black foto negative, have a look at Instructables for this cheap DIY IR filter!
Roderick G got his surround audio working! It’s currently with four speakers but it will be six. With Ferdy’s colortracker we could already position sound on the basis of the position of a person, cool!
Continue reading Testingdays Performance Engine — Day 3
All work-stations are installed and running. We made an inventariastion of things to do, mostly on the software level. We used the Perfromance engine VUE-map to have a clear idea which components still need te be developed.
The optical tracking module needs still a lot of work. Ferdy will get some help from Tim tomorrow. Next to colortracking we try to make use of IR leds. In theatre however there is a lot of Infrared light coming from the theatre lights. We have found IRleds with a very bright output and a wonderful 140 degrees emitting-range. Roderick V has soldered one of them to a AA-battery and simple on/off switch.
Continue reading Testingdays Performance Engine — Day 2
From the 27th of april till the 11th of may we’ll be building and testing our first prototype of the performance engine! For this test we have the opportunity to work in Huis aan de Werf. Today (27th) we mostly transported all the equipment from several locations to our studiospace and builded the basic layout of all the computerworkstations.
Continue reading Testingdays Performance Engine — Day 1