El Aura

IMAP, Mail and the iPhone

Thu, 21 Aug 2008 23:05:00 +0200

I finally switched to IMAP for my main Gmail account on the occasion of getting an iPhone. I was diligent and did all the ‘recommended’ extra configuration:

In Mail.app’s account preferences, checked: ‘Store sent messages on the server’
Selected the ‘Sent messages’ folder in the GMAIL folder in the Mail sidebar and via Mailbox->Use this Mailbox for->Sent

However, no email sent from my Mac or from my iPhone using the gmail account showed up in the sent folder in gmail. Nor was any email sent from my iPhone stored anywhere. After a total of my maybe eight hours I had eventually figured it out.

The problem is Mail.app and even more so the iPhone.

The point of IMAP is to have synchronized read counts (and deleted mails, etc.) and synchronised folders in multiple locations including the server. The most import folder apart from the inbox might arguably be the Sent Messages folder.
With multiple email accounts if one has not consolidated both the receiving and sending in one (IMAP) account (or does everything locally in Mail.app) one usually wants to have control which server sends and stores the sent messages (since sending to a server which is accessed via POP won’t store the messages anywhere).
That is why in Leopard, Mail acquired the possibility to send emails from a given account only with a specified smtp server.

And that is where things break down. If Mail cannot reach the specified smtp server, it asks after a short while whether it should try a different server. I run into this issue almost daily since at work only my work smtp server is accessible and at home my work smtp server is not accessible. The problem is that Mail than sticks to that manually selected alternative server until that one fails AND there is no possibility to know which smtp server Mail is using at any given moment (apart from sending an email to oneself and examining the header). And on top, there is no simple way to reset the smtp server (my solution is to make an smtp server inaccessible and then see whether Mail fails, if yes Mail is currently using this server for this account, in case of failure I can also select an smtp server manually).

What is even worse that this alternative smtp server is now (without you knowing it) ‘synched’ to the iPhone. And if you examine the account on the iPhone, the smtp you originally selected is still listed even though the iPhone using a different one. And even if you fixed things in Mail, synching won’t fix it on the iPhone you have to reset it.

And Gmail? Well, it just got the beating because it is so darn convoluted that everybody seems to give you different directions of how to configure it. Who would have thought that the smtp listed for an email account on the iPhone is not the one actually used.

Capture NX - Why bother?

Sun, 17 Aug 2008 14:29:00 +0200

This quote from the Aperture User’s Network very much sums it up:

Personally I don’t care if a photographer uses Aperture or Lightroom or PhotoMechanic or even Bridge to do selects and ratings, but it pains me on a nearly physical level to see photographers opening up countless files in CS3 (or CS2 or CS, or 7, I’ve seen it all today) just to zoom in and see if they’re sharp.

(http://www.apertureprofessional.com/showthread.php?p=22282#post22282)

I got the same feeling after trying to use Capture NX (I bought vs. 1.3 one week before 2.0 was announced, bought the update now as well and only now started using it). When I first started using raw, I dabbled a bit with ACR, dragging images out of iPhoto on PS’s dock icon, but ended up doing most of my adjustments in PS proper. Only when Aperture came out, doing image adjustments went from being a chore to becoming a delight (I have used Aperture since 1.0.0).

I also own Lightroom (still on 1.4, waiting for the performance bugs to be ironed out in 2.0). Even though I clearly prefer Aperture, using LR for adjustments and web galleries is a lot of fun as well. In fact I do most of my web galleries in LR, exporting 50% jpegs out of AP and importing them into LR.

But Capture NX? If you only process a handful of images at a time, maybe. Or if you rely on the Picture Styles and things light Active-D Lightning, possibly. I bought Capture NX for its Auto-CA removal, which works great on raw files (not so on tiffs), but what is the point if your image has no CA (which you at any rate you only would see at 100%) but otherwise sucks? Not that Capture NX would not be able to produce nice pictures, but I simply would never arrive there. Firstly, the speed is abdominal, just switching the CA control on (or off) takes five seconds (on a Core 2 Duo MBP 2.33 GHz, 3 GB RAM, 7200 rpm drive). One cannot even call this a lagging response, this is in a completely different time zone. But foremost, it’s no fun using Capture NX. I don’t make a living of photos, so I won’t spend time on things that I annoy me.

And now imagine a non-destructive (ie, real-time) plugin to Aperture (or LR) that would be as slow as Capture NX. It would completely destroy the user experience of Aperture.

Why Spaces will never work

Sun, 01 Jun 2008 17:47:00 +0200

In priniciple, Spaces is already superfluous. The real problem with Spaces is that hiding applications + minimizing windows does basically the same thing. Hiding applications allow for the creation of any combination of virtual spaces by unhiding a set of applications (and de-minimizing windows). That is what most users, particularly power users, do constantly. Unfortunately, minimizing does not really work. One cannot easily enough discern the content of minimized window s and similarly to switching spaces, OS X de-minimizes windows when and if it feels like. Moreover, I at least tend to forget about minimized windows and usually open a new one with the same content instead.

An alternative solution:
What is needed is a sort of a shelf for currently un-used applications and windows. The Dock does a good job for applications but a less good one for individual windows. Something like Expose for minimized (or shelved) windows, perhaps optionally restricted to those within an application should complement it. Spaces already has this Expose-like view of your different shelves, pardon, spaces.
Therefore I think, Spaces should be replaced with Shelf. Shelf should allow you to move windows to different shelves (like Spaces does). One could simply select any shelf to be the current desktop. Hidden applications would not be assigned to any shelf but would simply be added to your desktop when unhiding them. If a hidden application has windows distributed over different shelves, unhiding it would bring up the Expose view of Shelf. Hiding applications would hide their windows on all the shelves as well. This would mean that one would have to arrange any ‘desktop’ one app and one window at a time (if the app only has only one window or all windows in the same shelf, that simply would mean clicking the app’s Dock icon; if the app had windows on multiple shelves, one would have to drag multiple windows). The only real disadvantage compared to Spaces (or a true multiple desktop feature) would be that it would not be possible to quickly switch between multiple window configurations, ie, the positioning of windows (if an app has windows on more than one shelf) has to be re-done every time. The only tricky question is how to differentiate between switching to an application and unhiding it. One solution could be to always unhide and switch only (or a open a new window in applications configured for that) upon shift-click.

The only other alternative would be completely separate desktops with no switching except for intentional switching.

Projects and albums in Aperture

Sat, 19 Apr 2008 23:38:00 +0200

The primary organisation of images in Aperture is via projects (and blue folders containing projects). Every image belongs to one and only one project. Albums provide a secondary organisational structure which is independent of projects but can be freely combined with them.

The main difference is that the same image can appear in several albums. That means that dragging an image from one album to another, ‘copies’ the image, it does not move it (dragging from project to project moves the image). The same image can therefore appear in albums within different projects. And albums themselves can appear outside of projects. Pressing alt, while dragging from album to album moves the image, but only from album to album, the master still remains in its original project.

My advice would be to that any organisational criteria that is exclusive (eg, holiday locations or trips, customers) should be implemented via projects. Any organisational structure that is topical (eg, people vs. landscapes) should be done via (smart) albums. Any output oriented structure (eg, web site, book, presentation) is also best done via (smart) albums, with the build-in web galleries being just a special kind of albums.
Since every image has to be associated with a project, it can be that when topical and exclusive criteria lead to the same structure, a structure based on projects already provides the necessary topical division.

To summarize, first think what ‘exclusive’ organisational structure you want, ie, a structure that is unambiguous, that says this picture should be here and not anywhere else, and implement that structure using projects and blue folders.
Secondly, think about how else you want to access your images and use albums for that (and group with folders when necessary).

A lot of people choose the date as an unambiguous criteria. That is the lazy and unimaginative way (but it naturally works). I think it is worth to make the effort to create a more descriptive structure.

Spaces - I don't get it

Thu, 20 Mar 2008 02:39:00 +0100

In Leopard, Apple introduced Spaces, a multiple desktop tool. So far, I have not been able to figure out what it is really good for. Most often when my monitor(s) become(s) cluttered, it is because I have too many Finder windows open, some needed for one task and one application, some for another. I can distribute these windows over several spaces but as soon I close all windows in one space and then click back into the Finder to open a new one, I’m switched to another space where there are still open Finder windows. So, Spaces is useless for the task of distributing Finder windows.

In contrary to ubiquious apps like Finder, Terminal, Mail, or Camino, some apps are only used for one task. They could be assigned to one space. Or a combination of two apps could be assigned to one space. Yes, that actually works, but instead of switching between spaces I can just hide and unhide apps, thereby creating my own virtual spaces. What is the point of Spaces if clicking on one app moves you to that app’s space and therefore hides all other apps? Geez, I can do this with alt-click on the app as well. If you could distribute different windows of the same app over multiple spaces, without being transported back to the space with the last-used open window of that app when in fact you want to create a new window for that app in your current space, Spaces would be usefull. Right now, it is pretty useless and just provides a fancy way of hiding and unhiding groups of apps.

After the dust has settled

Sun, 09 Mar 2008 17:11:00 +0100

After a week with Leopard surprisingly the machine seems to settle down (and I as well, I presume). The hiding issues are becoming less frequent. Mounted servers no longer ask for a password but still sometimes take very long (20 seconds) to mount. The spinning beachball has become less frequent also. Still, fixing and researching all these small issues and exploring all the new functinality takes a lot of time. These are pretty minor issues but can easily take one to several hours to figure out. Here is my list of what I had to do after the upgrade (necessary even with perfect hindsight):

Necessary installs:

Developer Tools
Fortran compiler
Acrobat plug-in
VPN client
Latex subsystem
Onyx
Cocktail
GnuPG (?)
GPGMail
LaTeXit 1.14.3 (1.14.4 hangs)
latexdiff

Required reconfiguration:

repoint svnX
repoint server alias
recreate IP printers and obtain their PPDs
Xcode 3 preferences & shortcuts (took some time to figure things out here)
redo the Dock (described earlier)
fix set-up of SMTP server in Mail
re-adding of ignore-locations for Spotlight
add the full path back to the Terminal prompt

Don't, really don't

Wed, 05 Mar 2008 00:02:00 +0100

Do not upgrade to Mac OS X 10.5, at least not now, unless you have really nothing else to do. Wait maybe another six months and make sure to take a week off then, you will need it. Not that 10.5 is in any way a particular problematic update but you will probably find about half a dozen minor issues and each will take about a day to fix. That is why you need that week, and a box of tranquilizers. And hopefully they have fixed one or two of these half dozen issues by then.

Apple may manage to fix in the order of about 100’000 bugs in a major OS release but they also may introduce about new 100 bugs (remaining after the .2 release) and out of these 100’000 fixed bugs one notices possibly ten, but out of these 100 new bugs one also notices about ten. That is the truly amazing part. To give you just a few examples. In Tiger, only a few Adobe apps would not hide with cmd-h (but they would hide if alt-clicked on another app). In Leopard, half of the time when you try to hide an app it will not hide, not by any method. Restarting the app will bring back the hiding capabilities. Or take mounting a server, in Tiger one authenticated ones and that was it for the rest of the year, in Leopard one has to authenticate about every second time, it is pretty random. And the server used to appear in the sidebar, now it may appear or it may not.

All these are rather minor issues but one spends hours first mounting and unmounting servers trying to figure out what is happening and then scouring the net for any indications of what is going on. That is why you need a week to install Leopard. And if during every month waited one issue is fixed then every month waited might save you one day of your life (and possibly even a few weeks of your life, considering the frustation that is avoided).

Not realizing your strong points

Tue, 04 Mar 2008 02:56:00 +0100

Macs had a really great Finder. Or at least they had one feature which made working in the Finder very neat when one had to deal with folders containing large numbers of files. Something I regulary run into whenever I start ‘coding’ projects. This includes larger Latex documents and all kind of programming languages where you keep amassing files, source code, supporting files (eg, images in Latex documents), documentation files, output and results files etc.

Sure, one could create subfolders for every kind of file, but as the project went along, ever more files were created, ever more subfolders were necessary and one constantly had rethink the folder structure to keeps things tidy. And then re-adjust the code so that it looked for and placed its files in the right subfolders. And one kept running into trouble because some older code thought a file should be in subfolder A, but now subfolder A was within subfolder B.

On Macs, the easy solution was to simply use a flat folder structure with subfolders only for files which were generated in large numbers. Keeping files listed by recently modified brought a lot of files already into an easy reach. But the coolest thing was that whenever one could not immediately find a file, one just typed a few letters of its name (or content) into the search box and instanteneously the selection of visible files narrowed down to much more practical selection.

I said ‘was’. Because in 10.5, Apple decided that this was a fairly unimported feature and that what people wanted when they typed something into the search box was most of the time finding a file that they had no idea where to find and that therefore should be looked for on the whole drive. Alas, what I considered one of the coolest features of the OS, was considered secondary by Apple. In 10.5 the default behaviour when typing something into the search box is search the whole computer. Sure, one can click on a button and restrict the search to the current folder. But that easy, instanteneous narrowing down of the current folder view is gone. Maybe somebody finds a plist to reverse that behaviour but most people using Macs will longer experience this great feature.

What I am missing with Spotlight in Leopard: Quickly seeing the...

Mon, 03 Mar 2008 22:21:06 +0100

What I am missing with Spotlight in Leopard: Quickly seeing the top five results for different categories. Dito for seeing images as thumbnails and the rest as a list…

How to make the Leopard Dock look more acceptable. Here is part...

Sun, 02 Mar 2008 22:13:00 +0100

How to make the Leopard Dock look more acceptable. Here is part of the how-to , sampling the desktop colour is the key.

svnX and file conflicts

Tue, 11 Dec 2007 20:35:17 +0100

svnX is an open source subversion client (svn) for Mac OS X. It requires the installation of the svn command line back end (Leopard includes it already). When it is presented with file conflicts of binary files, it might not be obvious how to resolve them.

In short, pressing ‘Resolved’ retains the local version, pressing ‘Revert’ retains the remote version.

'Replace all' droplet using Platypus

Sat, 17 Nov 2007 16:42:00 +0100

Using Platypus (http://sveinbjorn.org/platypus), a development tool for Mac OS X, all sorts of shell or other scripts can be converted into proper applications. The Perl script to do a ‘Replace all’ presented in a previous post has to be modified somewhat to handle the handover of arguments and file paths.

The code below has to be saved as a .pl file. In Platypus the path of the script has to be selected and the ‘Is droppable’ option has to be chosen. The resulting application can be used as follows. Dropping the text file in which one wants to replace a number of words, characters (to be precise, strings) onto the application. It takes the two lists in a file ‘list-of-changes.txt’ and searches for all instances of the string in the left column and replaces it with the string in the right column and saves the file with the prefix ‘new-‘. The file ‘list-of-changes.txt’ has to be in the same folder as the ‘replacetext.app’.

To be more flexible with the list about what to replace, one can drag two files, the file one wants to modify and the file with the list of changes onto the app as well. This overrides the hardcoded ‘list-of-changes.txt’ file. One, however, has to make sure to select first the file to be modified and then the file with the list of changes.

#! /usr/bin/perl

# Select file to be modified (if no cmd line input present, take file below)

use File::Spec::Functions;

my $dir = catdir $ARGV[0], qw/Contents Resources/;
chdir $dir;

$filename = @ARGV[1];
$changelist = @ARGV[2];
@ARGV = $filename;

# Read file to be modified
while (defined($line = )) {
$content .= $line;
}

# Read list of changes
@ARGV = qw# ../../../list-of-changes.txt #;
if (defined($changelist)) {
@ARGV = $changelist;
}
while (defined($line = )) {
$cta += 1;
@old[$cta] = substr($line, 0, index($line, “\t”));
@new[$cta] = substr($line, index($line, “\t”)+1, index($line, “\n”)-index($line, “\t”)-1);
if (substr(@old[$cta],0,1) eq ‘\’){
@old[$cta] = substr(@old[$cta],1, length(@old[$cta])-1);
@new[$cta] = substr(@new[$cta],1, length(@new[$cta])-1);
}
}

# Apply changes
foreach (1 .. $cta){
$content =~ s/@old[$_]/@new[$_]/g;
}

# Get filename
my $nameonly = substr($filename, rindex($filename,”/”)+1, length($filename) - rindex($filename,”/”));

# Save file
my $name = “../../../new-” . $nameonly;
open my $NEWFILE, ‘>’, $name or die
“Unable to open $name for output: $!”;
truncate $NEWFILE, 1;
print $NEWFILE $content;
close $NEWFILE;

Batch 'Replace all' Perl script

Mon, 05 Nov 2007 21:59:00 +0100

Copy the text below into a text file and save with the extension .pl. It expects the replacement text to start with ‘' (ie, a backslash).

#! /usr/bin/perl
# Does a ‘Replace All’ on file1, using the list of changes supplied in file2
# Usage: replacetext.pl file1 file2
# If no arguments are supplied the two hard-coded files (file-to-be-modified.txt and list-of-changes.txt) are used
# If only one argument is supplied, the list of changes are taken from the hard-coded file
# The file with the list of changes has to contain a tab-separated list:
# Left column expression to be replaced, right column new value

# Select file to be modified (if no cmd line input present, take file below)
$filename = (“file-to-be-modified.txt”);
if (defined(@ARGV)) {
$filename = “@ARGV”;
if (@ARGV == 2) {
$changelist = substr($filename, index($filename, ” “)+1, length($filename));
$filename = substr($filename, 0, index($filename, ” “));
@ARGV = $filename;
}
} else {
@ARGV = $filename;
}

# Read file to be modified
while (defined($line = )) {
$content .= $line;
}

# Read list of changes
@ARGV = qw# list-of-changes.txt #;
if (defined($changelist)) {
@ARGV = $changelist;
}
while (defined($line = )) {
$cta += 1;
@old[$cta] = substr($line, 0, index($line, “\t”));
@new[$cta] = substr($line, index($line, “\t”)+1, index($line, “\n”)-index($line, “\t”)-1);
if (substr(@old[$cta],0,1) eq ‘\’){
@old[$cta] = substr(@old[$cta],1, length(@old[$cta])-1);
@new[$cta] = substr(@new[$cta],1, length(@new[$cta])-1);
}
}

# Apply changes
foreach (1 .. $cta){
$content =~ s/@old[$_]/@new[$_]/g;
}

# Save file with prefix ‘new’
$name = “new-$filename”;
$namestr = “>” . $name;
open NEWFILE, $namestr;
truncate NEWFILE, 1;
print NEWFILE “$content”;
close NEWFILE;

A Disappearing Desktop

Mon, 13 Aug 2007 20:01:08 +0200

Today, I was suddenly faced with a Desktop that had disappeared. Instead of being a representation of the Desktop folder, it was just a background picture. No icons on it, nothing could be dragged onto it. A second user was not affected. I happened after a couple of Fast User Switches (FUS), after one switch, the Desktop simply remained blank.

I tried almost everything. Eventually I discovered that somehow the ‘Hide the Desktop’ setting had been switched on. It is accessible via Cocktail (Interface > Finder). But I certainly had not been using Cocktail during these FUS. So what triggered that changed setting, I can only guess.

All along while the Desktop remained blank, the Finder kept (silently) crashing during presumably the log-in. There were no symptoms of this (apart from the blank Desktop), but console.log and system.log dutifully reported it. Additionally, during my troubleshooting attempts, DiskWarrior found the following problem (and fixed it): Disk: “Main”, Repaired the Text Encodings Bitmap of the Volume Information, Location: “Desktop”.

That, however, did not fix the issue. Only un-checking the ‘Hide the Desktop’ in Cocktail did. It also stopped the Finder crashes. Re-enabling it, does hide the Desktop, but does not bring back the problem of the crashing Finder.

Most likely, therefore, something probably related to the FUS both changed the ‘Hide the Desktop’ setting and corrupted it or the file it is stored in (or related files). Switching it off again via Cocktail caused this corruption to disappear (as the settings file was re-saved, possibly).

Rendering intents, soft proofing and Aperture

Mon, 23 Jul 2007 00:10:34 +0200

Choosing different rendering intents allows to influence how images (and other elements) will be converted to different output spaces. Soft proofing is meant allow one to judge how the material will look on other output devices than our own screen. Soft proofing, however, is no panacea, most often it cannot really fully represent other output media. Understanding what it can show us and what not is important to enabling us to predict the result on different media.

For this is it necessary to understand the differences between the different rendering intents and soft proofing options. The relevant choices are relative colorimetric, absolute colorimetric, and perceptual. The figure posted below illustrates how the conversion for the luminosity values between the working space (ie, the space the image is defined in) and the output space is carried out.

Absolute colorimetric rendering intent simply maps all values into the new color space, cutting off everything that does not fit into it. Relative colorimetric rendering intent, on the other hand, maps the source profile’s white point to the output profile’s white point and scales all other values relatively. Out of gamut colours are still cut off, however. Adding black point compensation (BPC) to relative colorimetric additionally maps the source profile’s black point to that of the output profile and scales everything accordingly, ensuring that no luminosity values are being cut off. Perceptual rendering intent acts very similarly to relative colorimetric with black point compensation, but additionally also maps all pure colors from the source to the output profile. It also employs a non-linear gamut compression, that is a custom function of the profile and normally leads to higher compression close to the boundary of the gamut.

When soft proofing, the resulting compressed gamut has to be mapped back to the screen profile using (I guess) relative colorimetric with black point compensation. This seems to make the whole exercise rather pointless for this intent as the normal conversion from image to screen is also rel. col. w/ BPC (also it might be perceptual as I said a while back). This however is not fully the case, a point that is still a bit mysterious to me. One can however, in Photoshop at least, specifically switch off black point compensation for that back-to-the-screen conversion using the function ‘Ink Black’, which allows to see the true reduced gamut of the output profile. Additionally, the function ‘Paper White’ can achieve the same for the white point of the output profile, essentially using absolute colorimetric rendering intent for the back-to-the-screen conversion.

To finally come to Aperture, comparing the rendering of simple black and white gradients, it seems that soft proofing in Aperture means converting into output profile using perceptual and converting it back to the screen using at least partially the ‘Paper white’ option.

Rendering intent and soft proofing.

Mon, 23 Jul 2007 00:08:52 +0200

Rendering intent and soft proofing.

Latex & Elsevier - a few pitfalls

Thu, 28 Jun 2007 21:50:38 +0200

Submitting latex code to Elsevier online? Here are two issues I ran into: The pdf the website generates is letter-sized. Nothing in the documentation, or even in the documentclass elsart.cls itself, makes that clear. Long tables or figures might be cut off by the change from A4 to letter.

The eps files should be saved as ASCII (not ASCII85 or binary). At least it rejected the eps files saved with PS CS2 using ASCII85 as being compressed files it could not read; the Acrobat 8 produced eps files using ASCII were accepted however.

Subversion and Samba

Fri, 08 Jun 2007 17:26:59 +0200

In short, never ever add an empty file to a subversion repository if you have working copies on a Samba share and you are running the subversion client on Mac OS X (10.4.x). Here is a more detailed post to the comp.subversion.user group with links referencing other people’s postings on this issue.

What they forgot to tell you about CM

Sat, 26 May 2007 01:27:19 +0200

Every advice about color management starts with the declaration that without a calibrated screen things will never quite work out. Well, hard to argue with that. And if screens were able to show all colors that the human eye is able perceive, or at least the subset that can be captured by digital cameras and scanners, starting out with a calibrated monitor should indeed make it possible to predict and control how our photos will look on other media (than our monitor).

Useless to say, most monitors have a gamut roughly that of sRGB, with cheap LCDs or laptop LCDs often even noticeably smaller than this and only the best monitors approaching Adobe RGB. Moreover, monitors are notoriously bad at displaying very dark blacks and have various other limitations (eg, maximum contrast or even 6 bit panels).

All this is well known, and color management is well equipped to deal with the shortcomings of display media, or rather it enables the users to deal with the shortcomings. It does this mainly by converting colors between different color spaces, most notably between that of an image (largely sRGB or Adobe RGB) and that of an output media (for anything on paper often of the CMYK type). It offers different rendering intents and on top the option of black point compensation (BPC) and a great deal is written about what this conversion entails.

A second type of conversion is usually given much less attention, the conversion from an image’s color space to that of the monitor. With raw conversion becoming ever more common, the mismatch between the color space in which images exist and are processed and the monitor color space can get quite large. However, information about how this gap is bridged is hard to find. By chance, I just stumbled upon a perfect test case which should provide us with some clues.

At http://regex.info/blog/2007-05-24/467, Jeffrey Friedl posted very recently an impressive photo of some very colorful flowers. Viewing the photo in a CM-aware application (eg, Photoshop, Preview, Aperture, etc.) an my 15” MacBook Pro (MBP) or my 20” Cinema display (CD) using the default Apple-supplied profiles shows strong colors with smooth gradients and almost no banding.

Using the profiles created by my monitor calibration device, the Huey, and CM-aware apps, reveals strong banding in the violet and pink flowers. Keeping the Huey profile, but using a non-CM-aware app (eg, Camino, Firefox), however, show very little banding. Comparing my Huey monitor profiles with sRGB (the one Jeffrey’s photo was defined in), reveals that both my MBP and CD have a smaller color space in respect to violet and pink tones than sRGB.

What is probably happening when using the Huey profile, is that the color management system (CMS), using relative colorimetric rendering intent, is clipping the out-of-gamut violet and pink tones, causing the posterization. Or, less likely in my view, using perceptive intent with a very agressive handling of out-of-gamut colors, CMS is compressing them into a very small range of tones which creates the appearance of posterisation and clipping.

The Apple-supplied profiles are for the violet and pink tones considerable larger than the Huey profiles and therefore do not require any or only very light clipping, assuming relative colorimetric rendering intent (and/or they entice the CMS to use perceptive). All profiles in question, one should note, have the tag ‘perceptual’ as preferred rendering intent. Using a non-CM-aware app, the monitor color profile is not used for any conversion, and the RGB values are sent straight to the graphic card, which also avoids the posterization.

Summarizing my conclusions and resulting questions:
1) At least with the Huey, color conversion from an image’s color space to the screen is done via relative colorimetric.
2) The Huey profiles have a smaller gamut than the default Apple ones. Any good reason why?
3) If true, working with images in huge color spaces (ie, when doing raw conversion) could on most screens very fast lead to posterization, although viewing the same photo on a high-gamut screen would not show it (and calibrating the screen with the Huey makes this worse). Or do raw converters work only internally with a huge color space but convert it down using some kind of perceptive rendering intent to a smaller one and then hand the result over to the OS’s CMS?
4) If the conversion from image to screen is done using relative colorimetric, is BPC enabled? I would guess so, but I would have to think that over (and hopefully come up with a test case to check it).

Things to find out about CM

Sat, 12 May 2007 17:51:55 +0200

About Aperture:

- What is the internal color space of Aperture? Presumably something big (ie, maybe similar to Adobe RGB).

- What is the rendering intent used [by Aperture] when exporting images and can it be changed? Apparently, Aperture uses the ‘preferred rendering intent’ of the printer profile when printing (as reported on the Aperture discussions forums at apple.com).

- Now, how does one find out what the ‘preferred rendering intent’ of a profile is? ColorSync Utility might show it but I could not find anything.

About ‘Working Color Spaces’:

- Having a monitor with a very wide gamut probably is rather helpful. Any image defined in a color space larger than one’s display’s gamut (and that is using the full range of possible colors) will be converted into the display’s color space for displaying purposes. Presumably using perceptive?

… to be expanded.