can't decide on con...
 
Notifications
Clear all

can't decide on controller

9 Posts
3 Users
1 Reactions
1,243 Views
Posts: 22
Topic starter
(@ketas)
Eminent Member
Joined: 19 years ago
[#58394]

i can't decide on what method to use to control my future setup. the reason to build is that those tiny inline cabled ws2812b's are getting so cheap

could get maybe 1000pcs for a start

but can't decide how to control them at all. wifi is out, i'll hardwire it. my options are:

1) usb serial and sw, drawbacks

2) usb serial with mpsse, strange

3) uart to uc, and fastled, but serial bw bottleneck?

4) ethernet using some phy, some uc, running fastled, i like it the best somehow

just can't decide on hw eh. i searched too. 11y ago, i loved those ohm2013 light controllers, just ethernet ic and atmega, protocol was artnet

those, similar, controllers i found a bunch too. not a hard task even, too, just bridge it

is wt hw currently best for it? also, i can split it up too, don't need to run it all from single host

as for display sw, i'm unsure, i could do that later, just maybe someone else has had this issue too

clearly i know things, just decisions are somehow hard : )

i can flash things on my own too, i don't need to buy ready made hw, unless someone somewhere here has that in a large batch or so

i see things for wifi, but not ethernet, sadly. i would prefer that and wifi config is a hassle too. i even considered repurposing bt controllers that sometimes come from those but that'll bite my ass later i assume

so, any ideas on how to choose?

it's luckily not very expensive mistake tho

oh and i searched here too, but the sheer amount of information here is... good, but that doesn't help. it's tmi, eh


8 Replies
Posts: 6728
(@p-short)
Illustrious Member
Joined: 19 years ago

What's wrong with WiFi? If you're worried about interference and dropouts, there a few solutions that involve storing the pixel data in the controllers and using WiFi to tell the controllers to start playing.


Reply
dirknerkle
Posts: 10310
(@dirknerkle)
Illustrious Member
Joined: 18 years ago

Instead of buying hardware now, why not figure out what you want your display to look like, where the lights are, what kind, etc. and then buy the equipment that works with the lights you choose? For example, if you planned to buy an electric car, would you buy a couple thousand gallons of gas for storage in a tank?


Reply
Posts: 22
Topic starter
(@ketas)
Eminent Member
Joined: 19 years ago

it's just easier to wire it to be honest, for me. i'm wiring mains there anyway. it's not like people actively interfere with wifi in my 1100m2 garden
but apparently wifi is strangely more common even in fixed install. times, eh
among other things i would need to properly extend range there and i would also want to segment it somehow. mind you, this is home network with 13 vlans in it

i recall things like
https://wiki.tkkrlab.nl/tkkrlab.nl/wiki/OHM_LEDS.html

one can easily repeat it and i guess like it. perhaps with other hw, perhaps not. noone else really does it?


Reply
Posts: 22
Topic starter
(@ketas)
Eminent Member
Joined: 19 years ago

i have no idea how display will look like. i would just try one pine tree. there's only one pine. it's highest in garden. it's near one main road that crosses right in the middle of private houses area of my smal 6k people town. would like some slow effects there so it's fun to walk by. white blinking stars, etc. it would be orders of magnitude better than string of 100 of supermarket 8 mode leds left to random mode. this also means i don't really spend much. as you see i'm member for 17 years here. i've wanted to have display that long. now i can build even better one. i've wanted music sync too but seq is super hard job


Reply
Page 1 / 2
Share: