Fork of HCFR Started - What's needed? - Page 11

+1

I just saw this, and have to admit I've never understood the purpose of measuring 0 IRE. I can understand if you're doing a contrast measurement, or if this was an option you could toggle (for those who MUST know what their absolute black level measures at), but forcing this measure every time is redundant.

The default operation is that they calculate gamma between black and white (with black compensation), and if you have a display with a considerable Y measurement at black it can affect the gamma calculation to some extent.

Even if we assume some value in measuring IRE 0, we shouldn't measure the same value more than once (per series). At least on the mac version if you do a full set of measurements, you end up doing both IRE 0 and 100 a couple of times, which seems silly.

Same here. I never was given an option to use it anyway like XP used to do. I have not tried to use unsigned drivers in a long time.

I did look into it and found this,
http://www.killertechtips.com/2009/0...ing-windows-7/

I'll give it a try tonight and see what this does.

Thanks for the reply,

I think that's because I failed to ship some required files with the release, the drivers should be signed I think they fail because of the missing files.

Try again with the next version which should be out this week.

John

I think there is method in the apparent madness, at the moment there is no workflow support and so no assumptions about which points won't move can be made. This not going to change quickly but we'll try and move it in the direction of much more flexibility in what readings are taken for a given run, first on the windows side will be reapeating a single measurement from within a run.

John

What do I have to download for the display LT and the i1pro? Having a little trouble with the new version

there should be nothing else to dwonload, as above theer was an issue withe installer and I missed a few files, I'd wait for the next version and try again.

John

Ok, Ill install the older version

My display has a considerable Y at 0% (.15 cd/m2) - is it better to use black compensation or not in this case?

There isn't an easy answer to this, none of the currently implemented formulas take account of raised black in a sensible way. It's an open research topic basically, I have a few ideas that I'll add support for at some point but that's some way off. Also I'm not convinced the code does black compensation correctly anyway, so I'd not worry about it.

the biggest issue with raised black is missing shadow detail so don't be afraid of raising the 10% and 20% points a bit if you feel that your blacks are getting crushed.

John

Thanks. To be sure I understand correctly, raising 10% and 20% would result in a lower gamma at those levels, right? I would assume we're talking about a subtle adjustment, maybe shoot for 2.15 at 20% and 2.1 at 10% if my overall gamma target is 2.22?

Yes exactly, then try it out with material with known shadow details preferably that you've seen on a calibrated display with better blacks and confirm that the result is acceptable.

John

Will do John.

Thanks for all your work!

Hi John ,

Is it your intention to update your first Post , pg1 , to reflect new releases and possibly some Release Notes ? This will help all of us from repeating questions and requests for items already released or not released . Just a thought .

BTW , I'd be really interested in the long or short story of how you ended up with the HCFR Codes and the ability to start upgrading this old favourite . We know the original Team stopped development a while back but how did you get involved ?

Scott..................

I looked at this tonight and thought I'd share my observations.

Here are my grayscale readings from a recent calibration that was performed with black level compensation OFF:

I changed my settings to "Display gamma with black level compensation", and HCFR recalculated all of the gammaY values:

As you can see, if I were doing the calibration run with black level compensation enabled, HCFR would have me raise Y, just as you suggested. The only difference being that HCFR is suggesting changes to all gray levels starting with large-ish amount at 10%, gradually decreasing to almost no change at 90%. The gamma chart changes as would be expected based on these re-calculated values.

So my observation is that the code is indeed doing something currently, and it seems somewhat reasonable based on your suggestion to me. Hope this is helpful.

I'd prefer to use the project's wiki as the main source of information and this post will point to that but yes there will be updates in the usual places when there are releases.


I forget exactly how I got hold of a copy of the source, I've had the zip in a safe place for a few years now, I remember it being a pain to get as the original devs were wierd about posting it publically and very slow to responding to requests for it, so when I saw a source zip posted briefly it I grabbed it. I've never had any contact with them apart from Jerome who was very helpful in getting us the mac code.

As to why now, for my cms product I need a tool to help people do pre-setup and also to verify that it's working correctly and no real desire to build something from scratch. I'd been thinking about the best way forward for a while. Doing a bit of digging here and there to find out who supported what.

The issue with HCFR has always been the meter support has been a bit dodgy from a GPL standpoint and I was watching the argyll cms project and seeing the meter source was gpl and then it was just a question of doing a cut and shunt job on the code, that and not being able to use my colormunki photo in HCFR and having a bit of free time to play meant that I just went ahead and did it.

John

Oh OK thanks for that, it may well be ok on the greyscale page but I don't think the black handling is consistent throughout.

My issue with black compensation is just that it doesn't go far enough, with a raised black level even with black compensation the near black points are too close together to be distingishable. With black compensation is better than nothing.

John

Thanks John ,
That clears up a lot of questions . Hopefully we can all supply you with the info you need to make HCFR the best free product we have .

Scott.............................

Hi folks,

first of all THANK YOU. It's a miracle, and it happened

Another Colormunki spectrophotometer user here ... I could help testing the Win and Mac versions (but the Mac version is coming?)

Thanks again!!!
Andrea

The mac version is being worked on... although I have been somewhat distracted this weekend... new iPad! hah.

Anyway, the current mac version works in that its possible to get readings, however, there is currently no way to select which device to use if the argyll library detects more than one. I've also been taking a bit of time, because, longer term, it would be nice to share the capability that exists on the windows version of having multiple meters used concurrently, and the GUI interface will be heavily related to how we implement this.

TL;DR version: the current mac version works but is rudimentary, and still needs some work before I'd consider a release.


Hope this helps.

eheheh
thanks for your effort, guy! we appreciate it.

Andrea

JohnAd, not entirely related, does Upsilon Mixer make use of of secondary xyY measurements? For example, yCMS takes primary xyY's only.

Cheers

It does during the refinement phase where it finds out how non-linear the display is at various points in the cube.

John

3.0.1.0 is now up on sourceforge

+ Update Argyll code with fixes for i1d3 and i1d2 and support for Spyder 4
+ Add support for getting spectrum with Argyll meters
+ Add extra Argyll tools to install package
+ Add support for new display type function within Argyll
o Improve handling of multiple meters
o Fix issues with loading file with disabled meters
o Add missing driver files to installer

Could i1d3 and spyder 4 users try and use the new command line utilities to extract ccss files for their meters.

Thanks

John

Hi John,

The sourceforge link is not working for me for the new version.

Thank You,
Jason

great timing John. i am getting a new lamp for my projector tomorrow so should be able to do some more calibrating later this week.

does one need to uninstall the previous version?

thanks

I've been really busy for the past couple of weeks, and unfortunately I haven't had a chance to test anymore. I've popped in today, and I noticed the new version. I'll see if I can test during the weekend (link and install worked fine for me). In the mean time, I've attached the ccss files from my i1D3. I hope they're helpful.

 

ccssFiles.zip 82.439453125k . file

Odd, here's the direct link
http://sourceforge.net/projects/hcfr...p.exe/download

All the mirrors may not have uipdated when you tried.

John

No just install it over the top of the old one.

John