[argyllcms] Re: Colprof -b key and B2A table computing issue
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Thu, 24 Jul 2008 12:38:18 +1000
Nikolay Pokhilchenko wrote:
Hi! It seems to me, that the either -bn and -bl keys of colprof (V1.0.1_Win32
without patching)
doesn't affect the B to A table computing process. I have to wait a long time
every time I'm do
preliminary profiling for choosing the black generation curve parameters.
The -bn key don't disable B2A table in the resulting profile.
See the documentation - it's not meant to. It only has an effect on input
profiles,
as output profiles are not ICC compliant without a B2A table.
I'll change it to be the same as (the undocumented) -B for output profiles
though,
which is the closest thing to not having a B2A table.
The -bl key don't decrease computing time with the
-qm or -qh or -qu keys.
time colprof -ni -no -qm 1150
real 0m47.524s
time colprof -ni -no -qm -bl 1150
real 0m17.774s
time colprof -ni -no -qm -B 1150
real 0m8.986s
Of course the effect depends on how much time is spent
on the A2B table, and computing the per channel input curves
takes some time (ie. not disabling them with -ni -no).
Graeme Gill.
Other related posts:
