16bit mode in DMX512/1990

[Adrien Gary Lucca]

Hi everyone,

I have some of these PCB DMX drivers from a chinese company called G&G:

https://www.alibaba.com/product-detail/4-channel-constant-current-350mA-700mA_1600315235915.html?spm=a2700.12243863.0.0.38403e5f4nmueS

These drivers have interesting specs:

4 channels

12-36V input

(fixed) constant current per channel 350mA, 700mA or (selectable) 800~1500 mA

manufacturer says it has 16bit greyscale capacity per channel

However, the manufacturer says that the PCBs will only use 4 DMX addresses, which in my experience makes it an 8bit driver (255 values per channel)

But the manufacturer keeps saying that this is only true for an older version of the DMX protocol and that it is possible to operate the board in 16bit mode using only 4 DMX addresses... which is something I have never heard of...

Does any of you has some experience with a way of using DMX drivers in 16bit per channel mode with only 1 DMX address per channel?

Best regards,

Adrien Lucca, Belgium

[Peter Newman]

You can download the E1.11 standard (there's an updated version which is mostly just a re-ratification of the initial one):

http://tsp.esta.org/tsp/documents/published_docs.php

But you're right, it's up to 512 8 bit slots and a start code.

I think there have been non-started extensions over the years with twice as many slots, and possibly with double-width slots and Avolites definitely used to output two universes on the XLR5s on some of their desks, but none of these would match what they say with "official" standardised DMX.

Indeed they claim:

"Support US ITT DMX512/1990 and extended DMX512, transmission speed 250/500Kbps adaptive;"

So given they mention 500Kbps I assume they're talking about some magic 1024 slot universe.

Also be mindful of:

"Meanwhile, it is equivalent to a signal repeater (with signal shaping function), eliminating the trouble of adding repeaters and eliminating the need for coding lines"

Which would mean if a unit failed the ones downstream will also die (as opposed to normally when it just tees off to the driver and you only lose the broken unit).

[Adrien Gary Lucca]

Hi Peter,

thanks for this excellent answer, I was getting crazy thinking I did not understand the standard DMX protocol....

Do you have any idea how to use this "magic 1024 slot universe" if this is indeed what they mean?

Best regards

[Rowan Maclachlan]

The 16bit grayscale is maybe the resolution of the output PWM signal, the DMX is out is 8bit per channel.    

So 0.255 is gamma corrected presumably to a curve that outputs 0..65535, that is nice for LEDs/eyes.

Eye is less sensitive to small 'bright' changes than small 'dim' changes... presumably they are trying to say "LEDS dims smoothly" rather than  suddenly are on.

--
The Open Lighting Project: open-l...@googlegroups.com, #openlighting (irc.freenode.org)
To unsubscribe from this group, send email to open-lightin...@googlegroups.com
For more options, visit https://groups.google.com/groups/opt_out?hl=en
---
You received this message because you are subscribed to the Google Groups "open-lighting" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-lightin...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-lighting/25b10c30-15e3-449a-a191-620c4d7a668cn%40googlegroups.com.

[Glenn Meader]

Never heard of a way to provide 16bit per channel using only one DMX address per channel.
There is no mention of that in the DMX specs. There is no mention of that by Googling.

Probably a Chinese to English language translation problem.  
DMX is only 8bit per address. There is no "new" or older version of DMX  that differs regarding this.

 
Ask the manufacturer what console they recommend that supports 16 bit per DMX address. You will find none.

What happens if you hook one of these drivers to an ordinary console and try using 4 addresses and then try using 8 addresses?

--

[E.S. Rosenberg]

Your product is from alibaba.

From previous experience it would seem the majority of manufacturers there don't know the meaning of "standard" except when it suits them and saves money.

That being said in their fantasy land with "DMX-512" running at 500kbps each conventional "slot time" would be sufficient to broadcast 16b/2 slots so maybe that is the source of their claim.

The 3980Hz refresh rate is of course also completely unrelated to DMX (though granted that may just refer to the levels being sent to the LEDs).

Op za 27 nov. 2021 om 23:51 schreef Glenn Meader <gme...@gmail.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/open-lighting/CAHUyLYx4CpayQJXaqkQp9xpJbdQTr4UEZFgitsa0yzcjqeYUig%40mail.gmail.com.

[Adrien Gary Lucca]

Hi Hippy,

I see, so you are suggesting that the output of the driver would be non-linear (gamma corrected). I am going to check that immediately.

[Adrien Gary Lucca]

Okay.... mystery solved! Hippy was right, the device is an 8bit, gamma corrected driver, here's the plot of DMX input VS luminance, showing the non-linear pattern of the light intensity:

It's a pretty accurate cubic function.

Thanks a lot, now I finally understand the specs of this device...

Adrien

On Saturday, November 27, 2021 at 3:42:33 PM UTC+1 Hippy wrote:

[Peter Newman]

I think it's both/either; as Keeper of the Keys and I both mentioned, they also talk about 500kbps data stream, which would imply either 16 bit slots, or 1024 slots per frame/universe, or two universes with alternate frames down one cable.

They'd need to tell you which of those it is, and ideally some more specs. In theory it might be possible to hack OLA to do it, depending on hardware used and capabilities of the same, but it won't be very useful as it would probably only work with those drives, so just accepting the dimmer curve may be the way to go for now.

[Rowan Maclachlan]

That's the hog "overdrive" or something doubling the baud.... nobody does that in the wild except Hog model X something i believe?

To view this discussion on the web visit https://groups.google.com/d/msgid/open-lighting/3854c371-a48a-473c-8c4a-16a9a13bd799n%40googlegroups.com.