[argyllcms] Re: MYIRO
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Mon, 19 Oct 2020 15:30:43 +1100
DP N/A wrote:
Does your SDK documents how to interact w/ MyIRO device w/o using your own
driver so that
Graeme can write his own driver with whatever API his ArgyllCMS code needs ?
something
like this document probably
=
http://www.sdg-net.co.jp/products/x-rite/products_detail/pdf/spectroeye-serial-interface-ed2_en.pdf
As I understand it, no.
There are two classes of instrument interfaces. One has relatively high level
protocols, where the instrument itself processes its raw measurements into
spectral and color values. Things like the JETI 1211, Klein K10A,
other high end instruments and the old X-RITE DTP instruments fall into this
category.
The second class is where the protocol communicates low level information
such as raw sensor readings, and the processing into spectral and color
values is done by the driver software running on the computer.
This tends to be the case with lower end instruments, such as the
entry level colorimeters and the entry level spectrometers such
as the i1Studio and i1Pro. I'm assuming that this is also the case
with the Myiro.
Manufactures tend to be secretive about this level of protocol,
probably due partly to practical issues (documentation may
be sketchy, they don't want to provide support for people
trying to use such an interface, there are lots of ways you
can mess up the operation of the instrument and
compromise its accuracy or operation, or even brick it
if you don't know what you're doing etc.), as well as
the belief that "secret sauce" would be revealed.
[ I find this a little ridiculous - outside the manufacturers
themselves, I probably know more than anyone else about
writing such drivers. Many of the techniques I've pioneered
such as FWA compensation and manual transmissive strip measurement etc.
have subsequently been adopted into the instrument manufacturers
products. So I'm hardly a hobbyist, and would be able to
critique the protocol and color science. ]
I'm not that convinced by the "secret sauce" argument - I've yet to come
across anything truly novel in any of the instrument protocols I've
RE'd, and in any case competitive advantages in hardware like
instruments is typically protected by Patents, not secrecy.
[ I'm a little surprised that this low level approach is still being used.
Embedded micro's have become very powerful. Something like the ARM
chip used by the Teensy 4 even has double precision floating point
and signal processing capability. I though this might prompt a switch
back to more open protocols for color measurement instruments. ]
There are exceptions to the above. Some of the Ocean Optics modules
use a documented low level protocol that they even provide example
code for.
Cheers,
Graeme Gill.
Other related posts:
