Skip to content

ChromIQ v4.3.50

Choose a tag to compare

@github-actions github-actions released this 20 Nov 06:34
· 95 commits to master since this release

For everyone updating from 4.3.49: “Strip may be misaligned” now also finds a strip whose white paper at one end was read as a patch, a misread the check could not see before. It asks you, as it always has, and it never refuses a strip: Keep it keeps the reading in one click. For owners of an X-Rite i1iO 2 table who try the beta option “X-Rite i1iO support”: a real i1iO 2 has now read one complete page with ChromIQ 4.3.49, patch by patch, and this version changes what that first real page showed. The arm is brought home at the end of a measurement, the check at the third mark is gone, one plain question comes before the arm goes forward, and with the dual scan the table keeps M1 and M2 for every row that has them. Reading in rows, the dual scan on the table and a second page have still not run on a real table: please read the known issues at the end. With the beta options off, ChromIQ measures as 4.3.49 did, except that “Strip may be misaligned” asks about more misread strips, and the charts it makes, your measurements and your profiles do not change.

New

  • “Strip may be misaligned” also finds a strip whose white paper was read as a patch. When you read a strip by hand, ArgyllCMS finds the patches in your one long movement by the change of colour from one patch to the next. When it takes the white paper in front of the first patch (or after the last patch) for a patch of its own, every reading after it is filed one place along, and the check could not see this when the first reading was a blend of paper and patch. Now ChromIQ also compares the first two or three readings of a strip, and the last two or three, with the colours one place along. When they fit those clearly better, it asks the same question as before, in words of its own (“The start of strip A was probably read one patch out of place”), with Re-measure this strip, Keep it and Stop. A first or last reading that looks like the white paper gets its own question (“The first reading of strip A looks like the white paper”), because one reading cannot tell a misread from a patch that did not print: look at the patch on the print, and choose Keep it if it is white or very pale. How often it finds a strip filed one place along: we moved the readings of real, correctly read strips on purpose, with the white paper read as patch 1. It found a shift of one patch at one end in about half of them (47 %), of two patches in about four of five (78 %), of three in 86 %, and a whole strip in 98 % (before: almost never for one patch, about half for two or three, the whole strip as before). With an i1Pro 2 it finds a little less (45 %, 70 %, 78 %, 95 %). On 1,497 real strips that were read correctly it asked nothing. The check is on by default. To switch it off, untick “Warn me if a strip looks misaligned” in Preferences ▸ Measurement: this switches off every question about misaligned strips.
  • The app log now tells what happened in a measurement, and keeps what is private out of it. When something goes wrong, we will ask you for the app's log, and it now answers most of our questions by itself. At each start it names this version, your system, the build of ChromIQ's chart-reading helper and every beta option with its state. Each measurement writes what was in effect at its start (instrument, chart, patches, strips, patch size, filter, dual scan, checks), one line for each strip as it is decided (kept, asked about, read again, and why), every question a window asked with the button you pressed, the dual scan's and M0/M1/M2's start and end and where each file was written, and, with an i1iO table, each step of the sitting. The lines of the tabs' own logs are there too, in English. Before a line is written, a privacy filter replaces your home folder, your account name, your first and last name (also in device names such as “Annas AirPods” or “Anna's iPhone”) and your computer's name, and it replaces instrument serial numbers, device ids and Bluetooth addresses by short codes that stay the same on your computer but cannot be turned back into the number. Two lines that named your devices now say only what kind of device it was (the sound output at the start of a measurement) and how many paired Bluetooth ports were skipped. The log you already have is passed through the same filter once, at the first start of this version (with its older copies), and a line in the log says so; the crash log is cleaned at every start. What the filter does not remove: project, target and file names, printer names (unless the name contains your own), and anything you typed into a description or a note. Please look through the log before you send it if one of those is private. A name that is also an ordinary word (an account called “max” or “test”) is replaced only where it can only be a name, in a folder path or a device name. To find the file, press the new button “Show the log file” in the Help window: it shows chromiq.log in Finder or Explorer.
  • Every measurement file ChromIQ writes says which version wrote it. The .ti3 of a measurement, its M0/M1/M2 files, a merge or an average, and now also a measurement from the i1iO table carry two lines in their header, CHROMIQ_VERSION and CHROMIQ_HELPER_BUILD. Your readings are not changed, and ArgyllCMS and older versions of ChromIQ read such a file as before.
    Only with the beta option “X-Rite i1iO support” on:
  • The arm goes home at the end of a measurement. After Save and stop or Discard and stop, ChromIQ now brings the arm to its resting place as part of the ending. The page “The arm now goes back to its resting place” asks you to take your hands away first, and the arm moves when you press “Bring the arm home, my hands are away”. When your sheet reached far towards you, ChromIQ first moves the arm a little back along the way it came. “Not now” moves nothing, and the way home stays offered while the table window is open. Please do not move the arm while you reach for the sheet: ChromIQ brings it home only from the place where it left it, and when the arm has been moved, it says so and moves it no more.
  • With the dual scan, the table keeps M1 and M2 for every row that has them. A row whose second pass (with the UV light) failed keeps M0 only, and its own page says so (“Row 5 has M0 only for now”). Every other row keeps its M1 and M2, as when you measure by hand, and Continue Measurement with the dual scan also reads those patches again. In 4.3.49 the table kept M1 and M2 only when every patch had them.
  • Its own page when the table answers a small step towards the front with an error, or does not answer it (“The table stopped during a small step towards the front”): what happened, that ChromIQ sends nothing more, where the arm was and was sent, and what to do, with the power switch first.

Changed

Only with the beta option “X-Rite i1iO support” on:

  • No check at the third mark. After your three aims, the arm no longer moves the measuring opening to the third mark to ask “Is the opening on mark 3?”. Nobody can see the opening under the instrument, and i1Profiler has no such step: its first row follows the aims directly, and ChromIQ's does too now. This was decided with the table owner who read the first real page, as a first answer that may still be revised. The checks of the three aims themselves stay. The slow steps towards the front now go only as far as the patches need, not to the mark (on the first real page, 9.5 mm less).
  • One plain question before the arm goes forward. The front-zone page is now “The arm has not been this far forward on this table yet”: why ChromIQ asks, then two buttons, “Go forward carefully, my hands are away” and “I will move the sheet”, then what happens if the arm cannot go further. The second question before the slow steps is gone; the first button is the hands-off confirmation.
  • In row mode, the first row starts right after the aiming, from the arm you have just aimed by hand, after its own hands-off question, as i1Profiler starts its first row.
  • The time limit for very small moves. On the real table a move of a fraction of a millimetre was answered in 0.016 s. For a move only a little longer, 4.3.49's lower limit would have called such an answer too early and could have stopped a measurement for nothing. Moves of up to 1 mm now have no lower limit, and the time a longer move may take follows its straight-line length. Every one of the 1,102 real and recorded moves lies inside its window.
  • The question that starts the drive is asked again once (“ChromIQ is about to drive your i1iO table”), because what has and has not run on a real table has changed.
    With “Strip may be misaligned”:
  • The questions of the new check say one thing. When a question can also mean a patch that did not print, it no longer also says that the strip should be read again. The questions name the buttons (“choose Keep it”) and the option that switches the check off.

Fixed

  • 4.3.49 left the table in its aiming mode when a measurement was saved with the arm out (with “X-Rite i1iO support” on). ChromIQ switches that mode off only when it brings the arm home, so after Save with the arm left out over the table, the next START MEASUREMENT stopped, and its page named the wrong cause. Now the arm goes home as part of the ending, the last page says so if you leave it out, and the next START says “The table is still in its aiming mode” and what to do. If this happened to you with 4.3.49: press End measurement, stay beside the table with its power switch in reach, open i1Profiler 3.8.6 or older once and let it find the table (it should bring the arm home), then close i1Profiler and press START MEASUREMENT in ChromIQ again.
  • The page for an unknown answer of the table at START now begins with “Press End measurement”. Until you press it, ChromIQ still holds the table's connection, and i1Profiler cannot find the table.
  • No page promises any more that i1Profiler brings the arm home from a place where no recording shows it doing that. Those pages now say that it should, and to stay beside the table with the power switch in reach.

Known issues

  • A real i1iO 2 table has read one complete page so far. With ChromIQ 4.3.49 it read 528 patches patch by patch, on a Mac, with the aiming and the slow 5 mm steps towards the front, and “Bring the arm home” after a stopped measurement has worked once on a real table. The changes in this version come from that page and have run only on a simulated table.
  • Not yet run on a real table: reading in rows (in this version the first row starts from the arm you have just aimed by hand: please keep your hands off the arm when you press Go), the dual scan on the table, a second page, bringing the arm home from the front of the table and the way back before it, and the end of the arm's travel towards you, which the slow 5 mm steps are meant to meet gently. Nobody has yet seen what a real table does at that moment; if the motor keeps working, switch the table off at its power switch. Keep the power switch in reach.
  • If you close the table window with the arm still out, the table stays in its aiming mode, and ChromIQ cannot start a new measurement on it until the arm is home. ChromIQ cannot bring the arm home from a new measurement. i1Profiler 3.8.6 or older should bring it home, but no recording shows it doing that from the front of the table or from this mode: press End measurement in ChromIQ first, stay beside the table with the power switch in reach, and close i1Profiler before you press START MEASUREMENT again. ChromIQ does not start a table that was just switched on while the arm is out over the table, because that start has never been recorded.
  • The dual scan of the i1Pro 2 has been tried on a real instrument twice. A tester made two real dual scans by hand: the pass with the lamp is sound, and M1 and M2 have not yet been checked against a reference. The other limits of 4.3.48 still hold (how close M1 and M2 come to X-Rite's values with the standard factor, the ruler by hand, both passes read again when one fails), and “High resolution spectral mode (-H)” and the dual scan still do not go together.
  • “Strip may be misaligned” cannot find every shifted strip, and it can ask about a correct one. It does not find a moved strip whose first patch is itself white or nearly white, because that looks the same as a light print. On a print whose colours are far from the chart's design (for example one with green and blue swapped), it can ask about one strip per chart or more; each time, Keep it keeps the reading. A patch that did not print at either end of a strip is asked about every time.
  • “Show the log file” on Windows. The new button “Show the log file” in the Help window has not been checked on Windows when the name of the log's folder contains a space. If it does nothing there, you find the log file in Preferences, under Paths, at “Log file:”.
  • The new and changed texts of this version are shown as proposed texts and may still be reworded.

Installation (Linux, beta): Download the tarball for your architecture (x86_64 for most desktops/laptops, aarch64 for ARM systems like Raspberry Pi 4/5). Extract with tar xzf ChromIQ-Linux-*.tar.gz, then run ./ChromIQ/ChromIQ. Requires ArgyllCMS — install via your package manager (sudo apt install argyll) or from argyllcms.com/downloadlinux.html.


☕ ChromIQ is free and always will be — if it saves you time or ink, a coffee is a kind way to say thanks. Completely optional, and the app stays fully featured either way.

Support ChromIQ on Ko-fi

Source of libusb (LGPL): ChromIQ now includes libusb 1.0.30, unmodified, under the GNU Lesser General Public License 2.1 or later, loaded as a separate library. Its source, libusb-1.0.30.tar.bz2, exactly as published by the libusb project, is attached to this release.