Differences

This shows you the differences between two versions of the page.

--- g1_lighting [2020-02-13 18:36] – control alfie
+++ g1_lighting [2020-02-13 22:49] – [Eye candy] alfie
@@ Line 14: / Line 14: @@
 ===== Eye candy =====
+==== Off-the-shelf ====
+There are lots of weird cheap DMX lighting fixtures available from ebay/aliexpress. They're nowhere near suitable for any kind of "real" lighting, but for making the lab pretty/impressive they're worth a punt.
+  * **[[https://www.ebay.co.uk/itm/Mini-Stage-Lighting-Beam-LED-Moving-Head-DMX-Spotlight-RGBW-Disco-DJ-Party-Light/352496170626|£40 moving head beam thingy]]** No gobos, no focus, no zoom, doesn't do colour mixing (each colour channel's beam comes out of the fixture differently aligned, probably due to separate colour chips) but it's fourty quid!
+==== DIY ====
+We're hackers, we've made pretty lights before, we can make pretty lights again!
 ===== Control =====
+==== Hardware ====
 DMX is the standard lighting control protocol. Art-net and sACN are two protocols for pushing DMX over a network. Allie has some designs for cheap sACN to DMX nodes based around the ESP8266, which have proven to be pretty reliable so far.
-Control software will be required. [[https://www.qlcplus.org|QLC+]] is open source and reasonably alright.
+==== Software (sACN Router/Proxy/Mapper/whatever) ====
+The core of the new lighting will be a control engine (probably in python) which is responsible for the following:
+  * Generating control data based on cues (presets/scenes) fired over MQTT ("office mode") or by other automations (time-of-day control, perhaps? need to figure out boundary of what this software should be responsible for)
+  * Receiving sACN raw control ("party mode") data
+  *  Intelligently(-ish) mixing/fading between office mode data and party mode data, either based on MQTT commands or priority/precedence rules.
+  * Mapping the input channel space to the output channel space (channel 73-82 of control universe 3 might map to channels 1-10 of output universe 1 of unicast sACN device 10.200.10.3, for instance)
+  * Sending sACN to the lab's lighting fixtures
+  * Exposing some form of interface/configuration language to configure the above
+This is not an insignificant amount of complexity - I would assume it would end up including some sort of DSL for creating complex architectural lighting cues, as well as maybe some form of web interface/etc.