1
1
<!DOCTYPE doctype PUBLIC "-//w3c//dtd html 4.0 transitional//en">
4
<meta http-equiv="Content-Type"
5
content="text/html; charset=ISO-8859-1">
6
<meta name="author" content="Graeme Gill">
7
<meta name="description" content="Root of Argyll CMS documentation">
9
content="Mozilla/4.73 [en] (WinNT; I) [Netscape]">
10
<meta name="Author" content="Graeme Gill">
11
<meta name="Description"
12
content="Summary of changes for the most recent release.">
13
<title>Argyll CMS changes since last release</title>
16
<h1> Summary of Argyll CMS Changes since last release</h1>
17
<h3>For a <span style="text-decoration: underline;">complete</span>
18
and more detailed list of changes, please see the log.txt file.</h3>
19
<h1>[V1.3.2 -> V1.3.3] 13th May 2011</h1>
23
<li>Fixed compiler dependant bug in Eye-One pro and (possibly) Munki
24
high res. spectral wavelength calculation.</li>
25
<li>Add support for install variables DESTDIR and PREFIX in Jamtop.
26
These can be set on the command line using "jam -s"</li>
27
<li>Added targen -N parameter to allow adjustment of neutral axis
28
patch density emphasis, as well as increasing the default. This should
29
improve the result without needing to add explicit grey test patches.</li>
30
<li>Added spectro/instlib.ksh script to assemble all the files needed
31
for a standalone instrument library. Changed licence to GPLv2 for the
32
files included in the instlib.zip file that is thus created. See
33
spectro/instlib.txt for more details.</li>
34
<li>Fix Jambase so that recent MingW compilers don't need extra .dll's</li>
35
<li>Change Linux serial code to test ports using O_NONBLOCK</li>
36
<li>Modify xspect & illumread to improve realism of UV spectrum
38
<li>Fixed profile/txt2ti3 so that a sample name that looks like an
39
integer is treated as text. (Fixes problem with latest ProfileMaker
41
<li>Added LCh option to spotread.</li>
42
<li>Fixed numerical issue in scanin/scanrd.c, where large input
43
rasters would cause fitting to fail.</li>
44
<li>Modified colprof input chart white patch detection to slightly
45
favour patches that are close to D50 neutral.</li>
46
<li>Increase the default XYZ PCS A2B profile default smoothness.</li>
47
<li>Improved cLUT input -u black & white point
49
<li>Improved black point determination for devices that have
50
extremely narrow gamuts doe to the use of custom inks.</li>
51
<li>Added -Z option to colprof, to allow setting ICC attribute flags.</li>
52
<li>Fix CIECAM02 to better match forward and backwards, to fix
53
perceptual table white point.</li>
54
<li>Add code to override X-Rite's new OS X drivers for ColorMunki and
55
EyeOne. Note new installation instructions
56
<http://www.argyllcms.com/doc/Installing_OSX.html>.</li>
57
<li>Added -R flag to colprof, which restricts the range of the white,
58
black for better compatibility with other programs.</li>
59
<li>Fixed typo bug that prevented flash measurement mode from working.</li>
60
<li>Replaced spectro/average with a new version that is more general.</li>
61
<li>Fixed bug in printcal not working with spectral only files.</li>
62
<li>Added extra verbose output to printcal in which it computes an
63
ideal power-like value to apply to the test chart values in targen.</li>
64
<li>Modify the way that XYZ cLUT B2A tables are indexed, so that the
65
white point is at the top corner of the grid.This should solve
66
Photoshop CS4/CS5 complaining that XYZ LUT profiles are 'defective'.</li>
67
<li>Added option in xicc/xicclu to plot an arbitrary slice.</li>
68
<li>Expand the number of i1 Display OEM devices that can be used.</li>
69
<li>Made some changes to help compile on FreeBSD.</li>
70
<li>Added another intent, "pa", Perceptual Appearance, which is the
71
same as perceptual except that the grey axes are not forced into
72
alignment, allowing the appearance parameters to have full affect,
73
including altering the chromatic mapping.</li>
74
<li>Fixed bug in txt2ti3 - it wasn't creating an iRGB colorspace file
75
for output device RGB files, causing warnings warnings and failures
76
when mixed with other iRGB tool sequences. </li>
77
<li>Added pathological case fix for target/ofps where the ink limit
79
<li>targen was failing to proceed when fixed points happened to be
80
numerically just over the total ink limit.</li>
81
<li>Added more navigation options for chartread patch by patch mode.</li>
82
<li> Fixed bug in "chartread -r -H" that caused resume of i1Pro
83
high res to fail with "The resumed spectral type seems to have changed".</li>
84
<li>Modified profcheck so that it prints patch location if it is
85
present in the .ti3 file.</li>
86
<li>Changed dispcal and dispread -K option to -J. Added -K option to
87
dispcal as an alternate way of profiling a calibrated display, and also
88
added a -K option to dispcal. </li>
89
<li>Increased ColorMunki emissive auto scaling target "over" margin
90
from 5% to 10% to allow more room for instrument drift during
92
<li> Fix bug in winusb + i1Display, where dark CRT measurements
95
<h1>[V1.3.1 -> V1.3.2] 4th November 2010<br>
98
<li>Turn off debugging that was accidentally left on in FWA code. Add
99
gcc 3.3 PPC optimizer bug workaround to FWA code in xicc/xspect.c</li>
100
<li>Change shaper/matrix profile back to using power curve as 0th
101
order shape. Improve it with input & output offsets and straight
102
segment at zero. Make cLUT input -u black & white point
103
extrapolation use pure shaper curves with special tweaks.</li>
104
<li>Increase dispcal native white target weighting from 10 to 50 to
105
encourage white to be device 1.0,1.0,1.0 more strongly.</li>
107
<h1>[V1.3.0 -> V1.3.1] 26th October 2010<br>
110
<li>Fixed MSWIN Vista/Win7 problem where having Task Manager running
111
would stop display test window updating. Also fixed plot library to
112
avoid the same problem.</li>
113
<li>Swapped dispwin -E and -D flags, to make -D debug consistent
114
throughout tools.</li>
115
<li>Changed the ARGYLL_NOT_INTERACTIVE mode so that all return and
116
line feed characters are ignored, so that they can be used freely to
117
flush stdin without triggering anything.</li>
118
<li>Fixed endless loop problem with chartread -r -p on fully read
120
<li>Added -S option to chartread, that suppresses wrong strip and
121
unexpected value warnings.</li>
122
<li>Fix dispcal and spotread so that color temperature takes into
123
account any non-standard observer (ie. the color temperature is the
124
closest point on the spectrum locus as determined by the chosen
125
observers interpretation of the Plancian or daylight spectrum.)</li>
126
<li>Fix bug in libusb1 triggered on systems that support bulk
127
continuation (Linux)</li>
128
<li>Added 1964_10c observer to spectro/dispcal, to better allow
129
comparison to the default numbers.</li>
130
<li>Added recognition for Huey built into Lenovo W series Laptops.</li>
131
<li>Fixed chartread/dispsup/spotread etc. so that -N isn't fatal if
132
the instrument doesn't support it.</li>
133
<li>Fixed dispcal to disable black & white drift tracking during
134
interactive adjustment.</li>
135
<li>Added -s option to ccmxmake to allow the number of test patches
138
<h1>[V1.2.0 -> V1.3.0] 8th September 2010<br>
141
<li>Added option to <span style="font-weight: bold;">dispcal</span>
142
and dispread that attempts to counteract instrument black drift and
143
display white drift (-I option). This may help with instruments that
144
haven't properly acclimatised to the measurement location, and LCD
145
displays that also take some time to stabilise. The is a short
146
discussion <a href="Scenarios.html#PM6">here</a>.<br>
148
<li>Added option to <span style="font-weight: bold;">dispcal</span>
149
to allow specifying a non 1931 2 degree observer if a spectrometer is
152
<li>Added new utility spectro/<span style="font-weight: bold;">ccmxmake</span>,
162
+ Display combination, using a Spectrometer as a reference.
163
The resulting <span style="font-weight: bold;">.ccmx</span> file can
164
then be used with <span style="font-weight: bold;">spotread/dispcal/dispread</span>
165
(-X option) to improve the accuracy of the colorimeter on that
166
particular display. See a discussion <a href="WideGamutColmters.html">here</a>
167
and <a href="Scenarios.html#PM6">here</a>.<br>
169
<li>Fixed bug in spotread's handling of emissive measurements. If
170
the XYZ was computed from spectral, it was using a D50 white instead of
171
no white reference.</li>
172
<li>Fixed bug in i1pro normal resolution wavelength calibration,
173
introduced in V1.2.0.<br>
175
<li>Changed libusb V1.0 name to libusb-1.0A, so as not to clash with
176
any official but different libusb V1.0 installation. [This may
177
necessitate re-installing device drivers on MSWin.]</li>
178
<li>Added support for HP DreamColor version of the i1 display.<br>
180
<li>Fix problem with ARGYLL_NOT_INTERACTIVE - reading from
181
instruments was not actually possible, because polling for input was
183
<li>Adjust ColorMunki dark threshold to reduce misread reports. Add
184
inconsistent data to debug output. Fix bug in adaptive mode - the
185
integration time was sometimes too short. Set adaptive emissive
186
target at 95% to allow a little more margin to saturation.<br>
188
<li>Fix problem with ColorMunki reporting erroneous inconsistent
189
measurement errors. This shows up on display calibration.</li>
190
<li>Fix some minor compiler warnings.</li>
191
<li>Added direction indicators to xy values in dispcal
192
interactive monitor adjustments. </li>
193
<li>Fix bug in CIECAM02 viewing condition settings :- the enumerated
194
conditions after "mt" are displaced by 1. (ie. "mt" is really "pc",
195
"mb" is "mt", "md" is "mb" etc.) Added option -c:sn for auto surround
196
from the Lv parameter (-c:l).</li>
197
<li>Add option to illumread to average several readings. Fixed bug in
198
the way illumread displays available instruments.<br>
201
<h1>[V1.1.1 -> V1.2.0] 30 July 2010<br>
204
<li>Re-worked gamut mapping to improve perceptual intent saturation
206
as well as improve highlight and shadow contrast. Added fine tuning to
207
improve both smoothness and the precision with which the source is
208
mapped to the destination.</li>
209
<li>Added illumread, which allows measuring an illuminant and
210
estimating its UV content, for better accuracy with FWA compensation.</li>
211
<li>Use a modified/forked version of libusb V1.0, that supports Win2K
212
(libusb0.sys) back end by default. Supports 64 but MSWin using a
213
combination of WinUSB.sys and ptlibusb0.sys. [ The HCFR does not work
214
on Win 64 bit though, due to its buggy USB implementation. ] NOTE that
215
the included version of Libusb V1 has been carefully tested with all
216
supported instruments on all supported platforms, and includes
218
bug fixes needed for correct functioning. While bug fixes have been fed
219
upstream, not all have been adopted. In particular there is a nasty
220
race condition that has not, and may never be fixed upstream, as well
221
as missing critical functionality (clearep()).<br>
223
<li>Modified colprof -p to allow different abstract profiles to be
224
applied for each intent.</li>
225
<li>Added -I option (imitation) to printcal, so that an existing
226
devices response can be set as a target.</li>
227
<li>Increase target/ofps.c vertex intersection retries from 10 to 40
228
to give it a better chance of working with difficult profiles.</li>
229
<li>Fixed bug in plot that shows up on XP+, where the window isn't
230
dismissed by the first keystroke, but only after it has been moved or
232
<li>Changed CMYK black point to be natural, rather than the darkest
233
point in the same direction and K only. This may wreck K only to black
234
point matching, but it will stop printers with funny colored K ink from
235
messing up the black point.</li>
236
<li>Make Lacie Blue Eye colorimeter appear as an i1display.</li>
237
<li>Improved i1pro matching to Original Manufacturers Driver (see
238
doc/i1proDriver.html).</li>
239
<li>Improved i1pro/ColorMunki patch recognition for better uniformity.</li>
240
<li>Fixed bug in ColorMunki driver scan mode calibration when
241
instrument is more sensitive than usual.</li>
242
<li>Added EV calculation to spotread -a<br>
245
<h1>[V1.1.0 -> V1.1.1] 21 February 2010<br>
248
<li>Renamed the following tools:<br>
249
cb2cgats -> cb2ti3<br>
250
kodak2cgats -> kodak2ti3<br>
251
logo2cgats -> txt2ti3<br>
252
splitcgats ->
255
mpprof -> mppprof<br>
257
<li>Modified black curve to make sure that smoothed curve meets
258
target level at boundaries.</li>
259
<li>Added -M option to printtarg, for the case where the TIFF file is
260
to include the margin.</li>
261
<li>Fixed several build bugs in imdi code related to 64 bits.</li>
262
<li>Fixed profile/colprof -u so that it is applied to matrix profiles
264
<li>Changed tiffgamut to use one pass gamut hull finding and modified
265
the smoothing so as to end up with a more closely wrapped volume.</li>
266
<li>Fixed bug in input matrix profiles introduced by the XYZ cLUT
267
display matrix profile change, where the correct white and black point
268
weren't being written.</li>
269
<li>Added matrix only/linear algorithm option to profile/colprof, for
270
raw camera profiling.</li>
271
<li>Modified libusb/55-Argyll.rules for better compatibility with
272
systems that have ACL installed but no ConsoleKit. Also set ID_VENDOR
273
and ID_MODEL using. usb-db.</li>
274
<li>Modified target/targen to cope better with case where adding
275
nodes fails to determine vertex positions a lot of the time, causing
276
extreme slowdown.</li>
277
<li>Fixed colprof so that the per channel input curves for XYZ PCS
278
B2A tables are actually scaled correctly.</li>
279
<li>Changed link/collink to apply Y to L* curve if the input or
280
output space is XYZ. Fixed the Y to L* scaling to make sure it only
281
apples to XYZ space, and that the L* non-linearisation still applies to
282
Y like device spaces.</li>
283
<li>Modified scanin so that it ignores any alpha channels in the
284
input .tiff file.</li>
285
<li>Changed printcal so that it will create .AMP file with more than
286
4 channels. Also fixed up plotting to plot up to 10 channels.</li>
287
<li>Changed dispcal and dispread so that a request for projector mode
288
falls back to display mode if the instrument doesn't support a
290
<li>Updated ref/CMP_Digital_Target-3.cht as it seems that the
291
reference chart has columns labelled "2A - 2D" rather than the "AA -
292
AD" that is actually printed on the chart...</li>
293
<li>Altered xpsect FWA code to reduce overshoot artefacts due to
297
<h1>[V1.0.4 -> V1.1.0] 17th January 2010<br>
300
<li>Spyder3 and ColorMunki Design, Photo and Create instrument
302
<li>Added a complete printer calibration system. This can work either
303
with a print system that supports per channel print calibration curves,
304
or purely using ICC profiling mechanisms.</li>
305
<li>Default targen (OFPS) test point distribution has been re-written
306
to generate test points on the gamut surface, refine the point
307
locations when using a guide profile, and use a better error estimate
308
model to determine the test point locations. <br>
310
<li>Changed chartread strip reading mode to allow navigating about
311
the strips, saving a partially read chart, and resuming a partially
314
<li>Improved and re-tuned
315
gamut mapping. This is noticeably smoother and better retains source
317
<li>Re-tuned the cLUT profile creation smoothness vs.
319
<li>Fixed viewgam so that the number of gamuts that can be viewed is
320
unlimited. Also added error when computing intersecting volume if the
321
two gamuts are incompatible.</li>
322
<li>Improved CMYK black generation control and smoothness near the
324
<li>Improved collink special black and colorant handling so that the
325
gamut mapping is consistent with the special black and colorant
327
<li>Changed profile/colprof to generate matrix tags for Display XYZ
328
PCS cLUT profiles, to improve compatibility with other CMMs.<br>
329
By default (-ax) the matrix tags will be a dummy transform that swaps
330
red and green, while using -aX will create real matrix tags.</li>
331
<li>Added -V option to dispcal and dispread to allow use of i1pro
332
adaptive mode to give better low level consistency.</li>
333
<li>Changed dispcal to default to -f 1.0 (assume black is all output
334
offset) to make it work in more sympathy to a typical display response.
335
Also changed default gamma to 2.4 for OS X 10.6 systems.<br>
337
<li>Improved X11 XRandR CRTC detection.</li>
338
<li>Added spotread option to save spectral reading of an illuminant
340
<li>Added Color Rendering Index (Ra) to spotread measurement results.</li>
341
<li>Added i1pro & ColorMunki flash measurement
342
support. <br>
344
<li>Improve reliability of detecting backwards read i1pro/Munki
346
<li>Fixed bug in i1pro driver that subtly affected accuracy.</li>
347
<li>Modified xicc/cam02 to clip the blue to avoid crazy behavior
348
outside the spectrum locus.</li>
349
<li>Simplified the Linux installation instructions, particularly with
350
regard to USB and serial permissions.</li>
351
<li>Added working MSWindows 64 bit libusb drivers, which will work on
352
Vista 64 and MSWindows 7 64 bit. Because of Microsoft's driver signing
353
requirements though, they won't be usable unless a driver code signing
354
workaround is used.</li>
355
<li>Changed printtarg so that the TIFF output has the paper margin
356
subtracted from it. This is so that the resulting TIFF can be placed on
357
that sized paper without clipping or scaling. Set the margin to zero to
358
get a TIFF that exactly fits into the specified paper size</li>
360
<h1>[V1.0.3 -> V1.0.4] 30th June 2009<br>
363
<li>Modify icc/icclib to protect against integer overflow exploits,
364
and fixes to minor bugs. Bump icclib version to 2.11 to reflect this.<br>
366
<li>Fix bug in spectro/hidio.c that can cause a crash (bus error) on
367
OS X for any program that accesses the instruments.</li>
368
<li>Fix bug in xicc/xfit.c where too little memory was being
372
<h1>[V1.0.2 -> V1.0.3] 3rd September 2008<br>
375
<li>Added multi-TIFF and popularity filtering to <span
376
style="font-weight: bold;">tiffgamut</span>.<br>
378
<li>Modified gamut mapping in <span style="font-weight: bold;">colprof</span>
379
and <span style="font-weight: bold;">collink</span> to be consistent,
380
and have higher perceptual intent saturation.<br>
382
<li>Fixed timeout problem with the Eye-One Display colorimeter.<br>
384
<li>Fix segmentation fault in <span style="font-weight: bold;">dispread</span>.</li>
385
<li>Fix out of memory error in <span style="font-weight: bold;">colprof
394
</span>for systems with > 3Gig Ram.</li>
395
<li>Add support for the Eye-One Monitor spectrometer.</li>
396
<li>Added -L option to <span style="font-weight: bold;">printtarg</span>
397
to suppress the i1pro target holder clip margin.<br>
399
<li>Fixed bug in <span style="font-weight: bold;">dispcal</span>
400
when using -a with -t<br>
403
<h1>[V1.0.1 -> V1.0.2] 19th August 2008<br>
405
Various bug fixes, the main ones being:<br>
407
<li>Fixed some <span style="font-weight: bold;">colprof </span>performance
412
<li>Fixed issues with Eye-One Pro Rev B timeouts.</li>
413
<li>Added new option to collink -fk, that forces 000K input to K only
415
<li>Added device target value quantization option to <span
416
style="font-weight: bold;">printtarg</span>, as well as making it
418
TIFF output files.put files.</li>
419
<li>Fixed leak that was affecting <span style="font-weight: bold;">printtarg</span>.<br>
422
<h1>[V0.70 Beta 8 -> V1.0.0] 1st July 2008<br>
425
Apart from numerous bug fixes and many minor feature additions and
426
improvements, the main changes to this version compared to the previous
429
<li>Speedup in profile and device link generation (inversion code),
430
and better memory usage.<br>
432
<li>Support for embedded profiles in TIFF files.</li>
433
<li>Support for installing and uninstalling and loading of display
434
profiles for all operating systems, and a micro CMM system for X11/Linux<br>
436
<li>Improved display calibration and profiling, including ambient
437
light adjustment.</li>
438
<li>X11 XRandR 1.2 support added.</li>
439
<li>Raster test charts now supported, as well as PS and EPS.<br>
441
<li>Guidance for installing on a wider range of systems.</li>
442
<li>Fixed luminance and ambient calibration issues with various
444
<li>Renamed "profile" to "colprof", and "icclink" to "collink" to
445
avoid clashes that have arisen with other executable names.</li>
446
<li>Streamlined source code build system, for much easier building.<br>
448
<li>Installation archives now include a top directory, and gzip/tar
449
format for OS X and Linux.</li>
450
<li>Added B2A table to input device LUT profiles by default.<br>
453
As usual, a more detailed description of all changes is in the <b>log.txt</b>
454
file that accompanies the
457
<h1>[V0.60 -> V0.70 Beta 8]15th January 2008<br>
460
<li>Added quick display ICC profile creation as part of calibration.</li>
461
<li>Added support for the Huey, Spyder 2, DTP20, Eye-One Pro,
462
DTP22/Digital Swatchbook, Eye-One Display 1 and 2 instruments.</li>
463
<li>Changed to GPL Version 3 license.</li>
464
<li>Countless other bug fixes and feature enhancements.<br>
4
<meta http-equiv="Content-Type" content="text/html;
6
<meta name="author" content="Graeme Gill">
7
<meta name="description" content="Root of Argyll CMS documentation">
8
<meta name="GENERATOR" content="Mozilla/4.73 [en] (WinNT; I)
10
<meta name="Author" content="Graeme Gill">
11
<meta name="Description" content="Summary of changes for the most
13
<title>Argyll CMS changes since last release</title>
16
<h1> Summary of Argyll CMS Changes since last release</h1>
17
<h3>For a <span style="text-decoration: underline;">complete</span>
18
and more detailed list of changes, please see the log.txt file.</h3>
19
<h1>[V1.3.7 -> V1.4.0] 20th April 2012</h1>
21
<li>Modified spectro/ccxxmake so that a colorimeter can be used as
22
a reference to make ccmx files if two .ti3 files are used. Added
23
ref/ccxx.ti1 as convenient way of creating ccmx .ti3 files.<br>
25
<li>Added dither/screening support for 8 bit output of render, and
26
then made it available in target/printtarg.<br>
28
<li>Added JPEG file support to imdi/cctiff, xicc/tiffgamut and
29
xicc/extracticc. ICC profiles embedded in JPEG files can now be
30
used anywhere a TIFF file with embeded ICC profile can be used
31
as a source of an ICC profile.<br>
33
<li>Fixed memory leaks in usbio.c, xdg_bds.c & conv.c</li>
34
<li>Fixed double memory free bug in icc/icc.c when iccdump'ing a
35
profile that has a duplicate tag.</li>
36
<li>Changed license of xicc/ccmx.[ch] to GPL2.<br>
38
<li>Made display calibration and profile making deal with displays
39
without hardware calibration support (VideoLUT support) more
40
graceful. Added tutorial section covering this.<br>
42
<li>Added option to dispwin/dispcal/dispread/ccxxmake to redirect
43
the test patches to a web browser via a local web server. This
44
augments Argyll's existing local and remote display capability.<br>
46
<li>Fixed bug in spectro/i1d3.c which results in NAN if a low
47
level readings drops to zero at a particular time. Improved
48
refresh rate calibration accuracy. Fixed bugs in adaptive
49
measurement logic that caused a channel to be pre-measured when
50
it shouldn't. This seems to noticeably improve repeatability on
53
<li>Fixed bug in ucmm/jcnf where it was failing to locate the
54
correct profile for a display.<br>
56
<li>Fix bugs in ColorMunki Transmissive measurement mode
60
<h1>[V1.3.6 -> V1.3.7] 26th March 2012</h1>
62
<li>Fix regression in Spyder support - ccmx files were not being
63
handled (bug introduced in 1.3.6).</li>
64
<li>Fix packaging problem - Spyder4 MSWin .inf file was missing.</li>
65
<li>Change dispwin so that it will install a profile when there is
66
no access to the display VideoLUT if the profile has no vcgt.<br>
69
<h1>[V1.3.5 -> V1.3.6] 19th March 2012<br>
72
<li>Added Spyder4 support. Note the need for spyd4en for access to
73
a full range of Manufacturers calibrations. The Spyder4 can use
74
.ccss calibration files too. Speeded up all Spyder instrument
75
readings on brighter colors.</li>
76
<li>Experimental ColorHug support is compiled in, but is disabled
77
unless the environment variable "ENABLE_COLORHUG" is set. The
78
ColorHug currently doesn't seem to work reliably across all
79
platforms ArgyllCMS supports.<br>
81
<li>Changed and expanded display selection (-y flag) to be
82
instrument specific. This is to support the Spyder4 and
83
ColorHug, and adds a refresh display selection to the i1d3.</li>
84
<li>Tweaked i1d3 integration times and added accurate refresh
85
period calibration to the refresh display mode. Refresh display
86
measurement times are double non-refresh displays. Improved i1d3
87
period measurement logic to improve measurement speed and
88
accuracy for dark colors.</li>
89
<li>Changed i1disp measurement logic to try and make it more
90
robust against light to dark changes during a reading. This may
91
make it slightly less precise for LCD displays on bright colors
92
(equivalent now to Refresh display precision).<br>
94
<li>Added a -V option to spotread to allow tracking reading
96
<li>Changed ccxxmake to create default .ccss with just RGBW,
97
and not to weight W. This may give better matching. Made
98
corresponding change to CCMX, giving the white patch 1/4
99
weighting of sum of all other patches.</li>
100
<li>Fixed applycal so that it applies calibration to both A2B and
101
B2A tables, to preserve softproofing.</li>
102
<li>Fixed timeout in SpectroScanT reference transmission
103
measurement. (Someone kindly donated me a SpectroScanT to test
105
<li>Made DTP94 driver ignore with a warning any
106
NEEDS_OFFSET_DRIFT_CAL_ERR after a full reset. It seems that
107
occasionally a few instruments do this, and X-Rite don't appear
108
to be prepared to treat this as an instrument fault.</li>
109
<li>Added support for Datacolor SpyderCheckr (Thanks to Jos
111
<li>Improved the ability of spyd2en to cope with slightly
112
different setup.exe formats.</li>
113
<li>Add support for NEC SpectraSensor Pro version of the i1d3.<br>
116
<h1>[V1.3.4 -> V1.3.5] 24th October 2011</h1>
118
<li>Fix bug (crash) that affects ColorMunki design/photo display
119
measurement. This also stops it restoring a calibration (-N
121
<li>Add support for the OEM version of the i1d3.</li>
122
<li>Fix bug that stopped ccxxmake being able to make ccmx's.</li>
123
<li>Tweak gamut mapping to improve dark area mapping,
124
non-monotonic profile inversion, and contrast preservation to
126
<li>Kill i1ProfileTray.exe process if unable to open i1d3 on
128
<li>Fix DTP20 chart printing - TID was sometimes incomplete. This
129
shows up on a 4x6 chart.<br>
132
<h1>[V1.3.3 -> V1.3.4] 31st August 2011</h1>
134
<li>Added support for the X-Rite i1 Display Pro and ColorMunki
135
Display colorimeters. As part of this, added support for CCSS
136
calibration files for the instruments and added CCSS support to
137
ccxxmake (renamed from ccmxmake). Provide new tool i1d3ccss to
138
translating and installing CCSS files as well as the
139
manufacturers calibration files for these instruments. Added
140
non-default observer support for these instruments too.</li>
141
<li>Fix gamut code to ignore setting primary/secondary cusps that
142
are unlikely to be true. This avoids buggy gamut mapping
143
behavior for gamuts that are very small and odd shaped.</li>
144
<li>Changed Linux USB code to avoid doing a
145
set_configuration if possible, since the USB driver does
146
this by default. This then avoids triggering a bug in the
147
Spyder2, which allows it to work on Linux version without the
148
reset_ep fix, and may also allow the Spyder to work better with
150
<li>Change printtarg for DTP20 to allow for variable patch size.</li>
151
<li>Changed dummy display matrix table to have channels rotated
152
rather than R & G swapped, to make it more obvious.</li>
153
<li>Added option to colprof to allow setting the default profile
154
rendering intent.</li>
155
<li>Enhanced spectro/fakeread so that it will process a .ti3 file
156
that has been renamed to .ti1.</li>
157
<li>Fix bug in matrix input profile white point selection, + add
158
in slight neutral bias code used in clut profiles.</li>
159
<li>New profcheck -I wasn't working - fix option parsing.</li>
161
<h1>[V1.3.2 -> V1.3.3] 13th May 2011</h1>
165
<li>Fixed compiler dependant bug in Eye-One pro and (possibly)
166
Munki high res. spectral wavelength calculation.</li>
167
<li>Add support for install variables DESTDIR and PREFIX in
168
Jamtop. These can be set on the command line using "jam -s"</li>
169
<li>Added targen -N parameter to allow adjustment of neutral axis
170
patch density emphasis, as well as increasing the default. This
171
should improve the result without needing to add explicit grey
173
<li>Added spectro/instlib.ksh script to assemble all the files
174
needed for a standalone instrument library. Changed licence to
175
GPLv2 for the files included in the instlib.zip file that is
176
thus created. See spectro/instlib.txt for more details.</li>
177
<li>Fix Jambase so that recent MingW compilers don't need extra
179
<li>Change Linux serial code to test ports using O_NONBLOCK</li>
180
<li>Modify xspect & illumread to improve realism of UV
181
spectrum estimation.</li>
182
<li>Fixed profile/txt2ti3 so that a sample name that looks like an
183
integer is treated as text. (Fixes problem with latest
184
ProfileMaker file).</li>
185
<li>Added LCh option to spotread.</li>
186
<li>Fixed numerical issue in scanin/scanrd.c, where large input
187
rasters would cause fitting to fail.</li>
188
<li>Modified colprof input chart white patch detection to slightly
189
favour patches that are close to D50 neutral.</li>
190
<li>Increase the default XYZ PCS A2B profile default smoothness.</li>
191
<li>Improved cLUT input -u black & white point
193
<li>Improved black point determination for devices that have
194
extremely narrow gamuts doe to the use of custom inks.</li>
195
<li>Added -Z option to colprof, to allow setting ICC attribute
197
<li>Fix CIECAM02 to better match forward and backwards, to fix
198
perceptual table white point.</li>
199
<li>Add code to override X-Rite's new OS X drivers for ColorMunki
200
and EyeOne. Note new installation instructions
201
<http://www.argyllcms.com/doc/Installing_OSX.html>.</li>
202
<li>Added -R flag to colprof, which restricts the range of the
203
white, black for better compatibility with other programs.</li>
204
<li>Fixed typo bug that prevented flash measurement mode from
206
<li>Replaced spectro/average with a new version that is more
208
<li>Fixed bug in printcal not working with spectral only files.</li>
209
<li>Added extra verbose output to printcal in which it computes an
210
ideal power-like value to apply to the test chart values in
212
<li>Modify the way that XYZ cLUT B2A tables are indexed, so that
213
the white point is at the top corner of the grid.This should
214
solve Photoshop CS4/CS5 complaining that XYZ LUT profiles are
216
<li>Added option in xicc/xicclu to plot an arbitrary slice.</li>
217
<li>Expand the number of i1 Display OEM devices that can be used.</li>
218
<li>Made some changes to help compile on FreeBSD.</li>
219
<li>Added another intent, "pa", Perceptual Appearance, which is
220
the same as perceptual except that the grey axes are not forced
221
into alignment, allowing the appearance parameters to have full
222
affect, including altering the chromatic mapping.</li>
223
<li>Fixed bug in txt2ti3 - it wasn't creating an iRGB colorspace
224
file for output device RGB files, causing warnings warnings and
225
failures when mixed with other iRGB tool sequences. </li>
226
<li>Added pathological case fix for target/ofps where the ink
228
<li>targen was failing to proceed when fixed points happened to be
229
numerically just over the total ink limit.</li>
230
<li>Added more navigation options for chartread patch by patch
232
<li> Fixed bug in "chartread -r -H" that caused resume of
233
i1Pro high res to fail with "The resumed spectral type seems to
235
<li>Modified profcheck so that it prints patch location if it is
236
present in the .ti3 file.</li>
237
<li>Changed dispcal and dispread -K option to -J. Added -K option
238
to dispcal as an alternate way of profiling a calibrated
239
display, and also added a -K option to dispcal. </li>
240
<li>Increased ColorMunki emissive auto scaling target "over"
241
margin from 5% to 10% to allow more room for instrument drift
242
during measurement.</li>
243
<li> Fix bug in winusb + i1Display, where dark CRT
244
measurements timeout.</li>
246
<h1>[V1.3.1 -> V1.3.2] 4th November 2010<br>
249
<li>Turn off debugging that was accidentally left on in FWA code.
250
Add gcc 3.3 PPC optimizer bug workaround to FWA code in
252
<li>Change shaper/matrix profile back to using power curve as 0th
253
order shape. Improve it with input & output offsets and
254
straight segment at zero. Make cLUT input -u black & white
255
point extrapolation use pure shaper curves with special tweaks.</li>
256
<li>Increase dispcal native white target weighting from 10 to 50
257
to encourage white to be device 1.0,1.0,1.0 more strongly.</li>
259
<h1>[V1.3.0 -> V1.3.1] 26th October 2010<br>
262
<li>Fixed MSWIN Vista/Win7 problem where having Task Manager
263
running would stop display test window updating. Also fixed plot
264
library to avoid the same problem.</li>
265
<li>Swapped dispwin -E and -D flags, to make -D debug consistent
266
throughout tools.</li>
267
<li>Changed the ARGYLL_NOT_INTERACTIVE mode so that all return and
268
line feed characters are ignored, so that they can be used
269
freely to flush stdin without triggering anything.</li>
270
<li>Fixed endless loop problem with chartread -r -p on fully read
272
<li>Added -S option to chartread, that suppresses wrong strip and
273
unexpected value warnings.</li>
274
<li>Fix dispcal and spotread so that color temperature takes into
275
account any non-standard observer (ie. the color temperature is
276
the closest point on the spectrum locus as determined by the
277
chosen observers interpretation of the Plancian or daylight
279
<li>Fix bug in libusb1 triggered on systems that support bulk
280
continuation (Linux)</li>
281
<li>Added 1964_10c observer to spectro/dispcal, to better allow
282
comparison to the default numbers.</li>
283
<li>Added recognition for Huey built into Lenovo W series Laptops.</li>
284
<li>Fixed chartread/dispsup/spotread etc. so that -N isn't fatal
285
if the instrument doesn't support it.</li>
286
<li>Fixed dispcal to disable black & white drift tracking
287
during interactive adjustment.</li>
288
<li>Added -s option to ccmxmake to allow the number of test
289
patches to be set.</li>
291
<h1>[V1.2.0 -> V1.3.0] 8th September 2010<br>
294
<li>Added option to <span style="font-weight: bold;">dispcal</span>
295
and dispread that attempts to counteract instrument black drift
296
and display white drift (-I option). This may help with
297
instruments that haven't properly acclimatised to the
298
measurement location, and LCD displays that also take some time
299
to stabilise. The is a short discussion <a
300
href="Scenarios.html#PM6">here</a>.<br>
302
<li>Added option to <span style="font-weight: bold;">dispcal</span>
303
to allow specifying a non 1931 2 degree observer if a
304
spectrometer is being used.<br>
306
<li>Added new utility spectro/<span style="font-weight: bold;">ccmxmake</span>,
307
which makes Colorimeter Correction Matrices for a particular
308
Colorimeter + Display combination, using a Spectrometer as a
309
reference. The resulting <span style="font-weight: bold;">.ccmx</span>
310
file can then be used with <span style="font-weight: bold;">spotread/dispcal/dispread</span>
311
(-X option) to improve the accuracy of the colorimeter on
312
that particular display. See a discussion <a
313
href="WideGamutColmters.html">here</a> and <a
314
href="Scenarios.html#PM6">here</a>.<br>
316
<li>Fixed bug in spotread's handling of emissive measurements. If
317
the XYZ was computed from spectral, it was using a D50 white
318
instead of no white reference.</li>
319
<li>Fixed bug in i1pro normal resolution wavelength calibration,
320
introduced in V1.2.0.<br>
322
<li>Changed libusb V1.0 name to libusb-1.0A, so as not to clash
323
with any official but different libusb V1.0 installation. [This
324
may necessitate re-installing device drivers on MSWin.]</li>
325
<li>Added support for HP DreamColor version of the i1 display.<br>
327
<li>Fix problem with ARGYLL_NOT_INTERACTIVE - reading from
328
instruments was not actually possible, because polling for input
330
<li>Adjust ColorMunki dark threshold to reduce misread reports.
331
Add inconsistent data to debug output. Fix bug in adaptive mode
332
- the integration time was sometimes too short. Set
333
adaptive emissive target at 95% to allow a little more margin to
336
<li>Fix problem with ColorMunki reporting erroneous inconsistent
337
measurement errors. This shows up on display calibration.</li>
338
<li>Fix some minor compiler warnings.</li>
339
<li>Added direction indicators to xy values in dispcal
340
interactive monitor adjustments. </li>
341
<li>Fix bug in CIECAM02 viewing condition settings :- the
342
enumerated conditions after "mt" are displaced by 1. (ie. "mt"
343
is really "pc", "mb" is "mt", "md" is "mb" etc.) Added option
344
-c:sn for auto surround from the Lv parameter (-c:l).</li>
345
<li>Add option to illumread to average several readings. Fixed bug
346
in the way illumread displays available instruments.<br>
349
<h1>[V1.1.1 -> V1.2.0] 30 July 2010<br>
352
<li>Re-worked gamut mapping to improve perceptual intent
353
saturation levels, as well as improve highlight and shadow
354
contrast. Added fine tuning to improve both smoothness and the
355
precision with which the source is mapped to the destination.</li>
356
<li>Added illumread, which allows measuring an illuminant and
357
estimating its UV content, for better accuracy with FWA
359
<li>Use a modified/forked version of libusb V1.0, that supports
360
Win2K (libusb0.sys) back end by default. Supports 64 but MSWin
361
using a combination of WinUSB.sys and ptlibusb0.sys. [ The HCFR
362
does not work on Win 64 bit though, due to its buggy USB
363
implementation. ] NOTE that the included version of Libusb V1
364
has been carefully tested with all supported instruments on all
365
supported platforms, and includes many bug fixes needed
366
for correct functioning. While bug fixes have been fed upstream,
367
not all have been adopted. In particular there is a nasty race
368
condition that has not, and may never be fixed upstream, as well
369
as missing critical functionality (clearep()).<br>
371
<li>Modified colprof -p to allow different abstract profiles to be
372
applied for each intent.</li>
373
<li>Added -I option (imitation) to printcal, so that an existing
374
devices response can be set as a target.</li>
375
<li>Increase target/ofps.c vertex intersection retries from 10 to
376
40 to give it a better chance of working with difficult
378
<li>Fixed bug in plot that shows up on XP+, where the window isn't
379
dismissed by the first keystroke, but only after it has been
380
moved or resized.</li>
381
<li>Changed CMYK black point to be natural, rather than the
382
darkest point in the same direction and K only. This may wreck K
383
only to black point matching, but it will stop printers with
384
funny colored K ink from messing up the black point.</li>
385
<li>Make Lacie Blue Eye colorimeter appear as an i1display.</li>
386
<li>Improved i1pro matching to Original Manufacturers Driver (see
387
doc/i1proDriver.html).</li>
388
<li>Improved i1pro/ColorMunki patch recognition for better
390
<li>Fixed bug in ColorMunki driver scan mode calibration when
391
instrument is more sensitive than usual.</li>
392
<li>Added EV calculation to spotread -a<br>
395
<h1>[V1.1.0 -> V1.1.1] 21 February 2010<br>
398
<li>Renamed the following tools:<br>
399
cb2cgats ->
401
kodak2cgats -> kodak2ti3<br>
402
logo2cgats -> txt2ti3<br>
403
splitcgats
406
mpprof ->
409
<li>Modified black curve to make sure that smoothed curve meets
410
target level at boundaries.</li>
411
<li>Added -M option to printtarg, for the case where the TIFF file
412
is to include the margin.</li>
413
<li>Fixed several build bugs in imdi code related to 64 bits.</li>
414
<li>Fixed profile/colprof -u so that it is applied to matrix
416
<li>Changed tiffgamut to use one pass gamut hull finding and
417
modified the smoothing so as to end up with a more closely
419
<li>Fixed bug in input matrix profiles introduced by the XYZ cLUT
420
display matrix profile change, where the correct white and black
421
point weren't being written.</li>
422
<li>Added matrix only/linear algorithm option to profile/colprof,
423
for raw camera profiling.</li>
424
<li>Modified libusb/55-Argyll.rules for better compatibility with
425
systems that have ACL installed but no ConsoleKit. Also set
426
ID_VENDOR and ID_MODEL using. usb-db.</li>
427
<li>Modified target/targen to cope better with case where adding
428
nodes fails to determine vertex positions a lot of the time,
429
causing extreme slowdown.</li>
430
<li>Fixed colprof so that the per channel input curves for XYZ PCS
431
B2A tables are actually scaled correctly.</li>
432
<li>Changed link/collink to apply Y to L* curve if the input or
433
output space is XYZ. Fixed the Y to L* scaling to make sure it
434
only apples to XYZ space, and that the L* non-linearisation
435
still applies to Y like device spaces.</li>
436
<li>Modified scanin so that it ignores any alpha channels in the
437
input .tiff file.</li>
438
<li>Changed printcal so that it will create .AMP file with more
439
than 4 channels. Also fixed up plotting to plot up to 10
441
<li>Changed dispcal and dispread so that a request for projector
442
mode falls back to display mode if the instrument doesn't
443
support a projector mode.</li>
444
<li>Updated ref/CMP_Digital_Target-3.cht as it seems that the
445
reference chart has columns labelled "2A - 2D" rather than the
446
"AA - AD" that is actually printed on the chart...</li>
447
<li>Altered xpsect FWA code to reduce overshoot artefacts due to
451
<h1>[V1.0.4 -> V1.1.0] 17th January 2010<br>
454
<li>Spyder3 and ColorMunki Design, Photo and Create instrument
456
<li>Added a complete printer calibration system. This can work
457
either with a print system that supports per channel print
458
calibration curves, or purely using ICC profiling mechanisms.</li>
459
<li>Default targen (OFPS) test point distribution has been
460
re-written to generate test points on the gamut surface, refine
461
the point locations when using a guide profile, and use a better
462
error estimate model to determine the test point locations. <br>
464
<li>Changed chartread strip reading mode to allow navigating about
465
the strips, saving a partially read chart, and resuming a
466
partially read chart.<br>
468
<li>Improved and re-tuned gamut mapping. This is noticeably
469
smoother and better retains source image detail.</li>
470
<li>Re-tuned the cLUT profile creation smoothness vs. accuracy.</li>
471
<li>Fixed viewgam so that the number of gamuts that can be viewed
472
is unlimited. Also added error when computing intersecting
473
volume if the two gamuts are incompatible.</li>
474
<li>Improved CMYK black generation control and smoothness near the
476
<li>Improved collink special black and colorant handling so that
477
the gamut mapping is consistent with the special black and
478
colorant mapping</li>
479
<li>Changed profile/colprof to generate matrix tags for Display
480
XYZ PCS cLUT profiles, to improve compatibility with other CMMs.<br>
481
By default (-ax) the matrix tags will be a dummy transform that
482
swaps red and green, while using -aX will create real matrix
484
<li>Added -V option to dispcal and dispread to allow use of i1pro
485
adaptive mode to give better low level consistency.</li>
486
<li>Changed dispcal to default to -f 1.0 (assume black is all
487
output offset) to make it work in more sympathy to a typical
488
display response. Also changed default gamma to 2.4 for OS X
491
<li>Improved X11 XRandR CRTC detection.</li>
492
<li>Added spotread option to save spectral reading of an
493
illuminant to a .sp file.</li>
494
<li>Added Color Rendering Index (Ra) to spotread measurement
496
<li>Added i1pro & ColorMunki flash measurement
497
support. <br>
499
<li>Improve reliability of detecting backwards read i1pro/Munki
501
<li>Fixed bug in i1pro driver that subtly affected accuracy.</li>
502
<li>Modified xicc/cam02 to clip the blue to avoid crazy behavior
503
outside the spectrum locus.</li>
504
<li>Simplified the Linux installation instructions, particularly
505
with regard to USB and serial permissions.</li>
506
<li>Added working MSWindows 64 bit libusb drivers, which will work
507
on Vista 64 and MSWindows 7 64 bit. Because of Microsoft's
508
driver signing requirements though, they won't be usable unless
509
a driver code signing workaround is used.</li>
510
<li>Changed printtarg so that the TIFF output has the paper margin
511
subtracted from it. This is so that the resulting TIFF can be
512
placed on that sized paper without clipping or scaling. Set the
513
margin to zero to get a TIFF that exactly fits into the
514
specified paper size</li>
516
<h1>[V1.0.3 -> V1.0.4] 30th June 2009<br>
519
<li>Modify icc/icclib to protect against integer overflow
520
exploits, and fixes to minor bugs. Bump icclib version to 2.11
523
<li>Fix bug in spectro/hidio.c that can cause a crash (bus error)
524
on OS X for any program that accesses the instruments.</li>
525
<li>Fix bug in xicc/xfit.c where too little memory was being
529
<h1>[V1.0.2 -> V1.0.3] 3rd September 2008<br>
532
<li>Added multi-TIFF and popularity filtering to <span
533
style="font-weight: bold;">tiffgamut</span>.<br>
535
<li>Modified gamut mapping in <span style="font-weight: bold;">colprof</span>
536
and <span style="font-weight: bold;">collink</span> to be
537
consistent, and have higher perceptual intent saturation.<br>
539
<li>Fixed timeout problem with the Eye-One Display colorimeter.<br>
541
<li>Fix segmentation fault in <span style="font-weight: bold;">dispread</span>.</li>
542
<li>Fix out of memory error in <span style="font-weight: bold;">colprof
559
</span>for systems with > 3Gig Ram.</li>
560
<li>Add support for the Eye-One Monitor spectrometer.</li>
561
<li>Added -L option to <span style="font-weight: bold;">printtarg</span>
562
to suppress the i1pro target holder clip margin.<br>
564
<li>Fixed bug in <span style="font-weight: bold;">dispcal</span>
565
when using -a with -t<br>
568
<h1>[V1.0.1 -> V1.0.2] 19th August 2008<br>
570
Various bug fixes, the main ones being:<br>
572
<li>Fixed some <span style="font-weight: bold;">colprof </span>performance
577
and memory usage issues.</li>
578
<li>Fixed issues with Eye-One Pro Rev B timeouts.</li>
579
<li>Added new option to collink -fk, that forces 000K input to K
581
<li>Added device target value quantization option to <span
582
style="font-weight: bold;">printtarg</span>, as well as making
583
it default for TIFF output files.put files.</li>
584
<li>Fixed leak that was affecting <span style="font-weight:
585
bold;">printtarg</span>.<br>
588
<h1>[V0.70 Beta 8 -> V1.0.0] 1st July 2008<br>
591
Apart from numerous bug fixes and many minor feature additions and
592
improvements, the main changes to this version compared to the
593
previous one are:<br>
595
<li>Speedup in profile and device link generation (inversion
596
code), and better memory usage.<br>
598
<li>Support for embedded profiles in TIFF files.</li>
599
<li>Support for installing and uninstalling and loading of display
600
profiles for all operating systems, and a micro CMM system for
603
<li>Improved display calibration and profiling, including ambient
604
light adjustment.</li>
605
<li>X11 XRandR 1.2 support added.</li>
606
<li>Raster test charts now supported, as well as PS and EPS.<br>
608
<li>Guidance for installing on a wider range of systems.</li>
609
<li>Fixed luminance and ambient calibration issues with various
611
<li>Renamed "profile" to "colprof", and "icclink" to "collink" to
612
avoid clashes that have arisen with other executable names.</li>
613
<li>Streamlined source code build system, for much easier
616
<li>Installation archives now include a top directory, and
617
gzip/tar format for OS X and Linux.</li>
618
<li>Added B2A table to input device LUT profiles by default.<br>
621
As usual, a more detailed description of all changes is in the <b>log.txt</b>
622
file that accompanies the source code.
623
<h1>[V0.60 -> V0.70 Beta 8]15th January 2008<br>
626
<li>Added quick display ICC profile creation as part of
628
<li>Added support for the Huey, Spyder 2, DTP20, Eye-One Pro,
629
DTP22/Digital Swatchbook, Eye-One Display 1 and 2 instruments.</li>
630
<li>Changed to GPL Version 3 license.</li>
631
<li>Countless other bug fixes and feature enhancements.<br>
added
removed
doc/ChangesSummary.html
