We both use and evaluate many monitor calibration and profiling instruments. This page summarizes our findings. Profiling systems are tested on a range of LCD monitors. These tend to be higher end displays, as it is both easier to evaluate profiling and calibration performance on these systems, and they are representative of monitors used by professional photographers. We included a few lower end displays as well. For a full listing of the displays used, see our monitor profiling details page.
We group the performance of the monitor profiling sysems by the type of monitor used. Some calibration systems perform better on certain classes of display than on others. The exact brands of monitor are not critical; what is current today will be outdated in short order. The ovberall characteristics of the monitors are what is most important. All monitors tested had 8-bit or higher internal LUTs. Cheap monitors designed only for gaming response times and having 6-bit internal processing are not capable of sufficiently good color performance that profiling and calibration are useful.
The majority of the test systems ran Windows 7 x64. Additional tests were performed on OSX 10.6. No significant differences were seen in performance or usability between Windows and Mac monitor calibration systems. Our reviews hold equally well for both platforms. Some calibration software supports multiple measurement devices. Which measurement puck was used made significant differences in overall performance depending on the monitor screen technology. We rated each calibration system using the best performing among the (relatively) affordable devices supported: Although some systems support $6500 spectroradiometers, we ranked them based on their performance with colorimeters primarily in the $250 and below range. For those interested in the details, we also compared the performance and accuracy of monitor profiling measurement devices on different display types.
The screens were calibrated using hardware adjustments when possible. Grayscale gradients and screen neutrality were checked both within and outside color managed applications. Spot checks of black, white, and neutral tones were made using a Photo Research PR-730 spectroradiometer. Extensive visual checks were then performed on images viewed from within Photoshop CS5.
Any monitor evaluations are necessarily subjective. Notes were compared from at least three sets of eyes. We ranked the various systems based on three criteria. First was overall performance. This is based on accuracy and aesthetics for all types of images. Synthetic images were used as well as a wide range of real world imagery. The next evaluation is of grayscale neutrality and tonality. Obviously, the more neutral the better. Of equal importance, however, was smooth tonality and minimal calibration-induced shadow banding. The final ranking is of shadow performance. We looked for an elusive combination of shadows that were as dark as possible while still revealing detail, open but not washed out, accurate color and neutrals, and smooth gradients as colors faded to black.
To help highlight differences between calibration software and sensors, the following color codes are used:
Outstanding performance. Visible improvement over other systems. |
Very good performance. Subjective measure: 90%+ of outstanding. |
Acceptable performance for non-critical work. |
Not recommended. |
The above subjective rankings are consistent for individual monitors only. Performance ranked acceptable on a Sony Artisan is far better than anything a laptop monitor can produce with any calibration system. In other words, the color-coded guide groups the performance of the calibration systems on a monitor by monitor basis.
Monitor | 1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
Sony Artisan (CRT) | G |
J |
F |
C |
D |
A |
E |
B |
|
LaCie Electron22Blue III (CRT) | J |
F |
C |
D |
A |
E |
B |
I |
|
HP p920 (CRT) | J |
F |
A |
D |
E |
C |
B |
||
Eizo CG-21 (LCD) | J |
F |
H |
A |
D |
C |
E |
B |
|
LaCie 321 (LCD) | J |
C |
D |
F |
A |
E |
B |
||
IBM T41p (laptop) | D |
J |
A |
C |
F |
E |
B |
||
Dell 8600 (laptop) | D |
E |
J |
F |
A |
B |
C |
Notes: The difference between the top four calibrators on the LaCie, Eizo, and HP monitors was small. All do a commendable job. The rankings of these instruments were largely determined by the spectroradiometer measurements of absolute neutrality. On the Artisan, the dedicated Sony puck gave results in a class by themselves. The Monaco XR, Eye-One 1, and Spyder2 were excellent but not at the same level. Strangely, this is one monitor where the new Eye-One 2 produced visibly less neutral results. On the laptop monitor, the top two calibrators were very close in performance, with a large gap to the next group.
Monitor | 1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
Sony Artisan (CRT) | G |
J |
C |
F |
D |
A |
E |
B |
|
LaCie Electron22Blue III (CRT) | J |
F |
D |
C |
E |
A |
I |
B |
|
HP p920 (CRT) | J |
F |
C |
D |
A |
E |
B |
||
Eizo CG-21 (LCD) | J |
D |
F |
E |
C |
H |
A |
B |
|
LaCie 321 (LCD) | J |
D |
F |
C |
A |
E |
B |
||
IBM T41p (laptop) | F |
J |
C |
D |
E |
A |
B |
||
Dell 8600 (laptop) | D |
F |
J |
E |
A |
B |
C |
Notes: On all displays the top three calibrators were close in performance (with the exception of the Artisan where the Sony puck led the way again).
Monitor | 1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
Sony Artisan (CRT) | G |
J |
A |
F |
D |
C |
E |
B |
|
LaCie Electron22Blue III (CRT) | J |
F |
C |
A |
D |
E |
I |
B |
|
HP p920 (CRT) | J |
A |
D |
F |
C |
E |
B |
||
Eizo CG-21 (LCD) | J |
F |
A |
D |
E |
H |
C |
B |
|
LaCie 321 (LCD) | J |
C |
A |
D |
F |
E |
B |
||
IBM T41p (laptop) | A |
J |
F |
D |
C |
E |
B |
||
Dell 8600 (laptop) | J |
D |
F |
A |
E |
C |
B |
Notes: The top four systems on the LaCie and Eizo monitors all showed very good performance. On the HP CRT and both laptops, the ColorVision Spyder2, GretagMacbeth Eye-One 2, and Monaco Optix XR all showed much better shadow detail and tonality than the rest of the field. The Spyder2 was particularly impressive on the HP p920 and IBM LCD.
Several results were system dependent. Nothing beats Sony's own specially tuned puck and software on the Artisan system. Not surprising, as the color filters are matched to the display phosphors. Neither Eizo nor LaCie's systems, both using non-modified hardware in their dedicated systems, were up to what third party software could provide. The poor showing of the GretagMacbeth Spectrolino on the Dell laptop came about because the Spectrolino's weight deformed the screen slightly, causing erroneous readings. The original ColorVision Spyder simply is not competitive in this bunch.
ColorEyes Display: This software combines stellar performance on high quality monitors with a non-intuitive user interface. As can be seen in the comparison tables above, the shadow performance is particularly impressive. With version 3.10, ColorEyes Display adds DDC control for many monitors. This allows the software to automatically adjust the monitor hardware. The benefit is most pronounced on high quality LCD screens, where the finely tuned gamma curve results in extremely good tonal gradients. Owners of Matrox video cards should be warned that the DDC functions are very slow with these cards, but they do work (uninstall any older versions first!).
The overall performance of ColorEyes Display is excellent on good monitors. Improvements to version 3.1 boosted performance on laptop screens and less than top-end LCD monitors as well. ColorEyes Display is available as software-only, which is compatible with many standard sensors, or as a bundle with the X-Rite DTP-94/Monaco XR/ColorEye sensor. The tests made above used the DTP-94. Performance with the Eye-One Display 2 is good, although not up to the DTP-94. Whether this is a software or hardware difference, we do not know. ColorEyes also supports a range of other sensors. Of these, the Eye-One Pro (spectrophotometer version) worked well. Other sensors, including the Sequel Squid and Eye-One Display 1 did not produce as good results.
The ColorEyes Display software is powerful, but has an unfinished feel. The user interface is non-intuitive, difficult to navigate, and distinctly designed for those both already familiar with monitor calibration and profiling. The flow through the software is not well thought out. For example, ColorEyes can use the instrument to guide setting screen luminance levels, although this requires starting the calibration with a CRT monitor selected, even if you use a LCD. You need to abort the process and restart if you have a LCD after setting the display luminance. In a similar vein, there is no direct readout of the screen white point. You can calculate it from the displayed xyY values, but this is not useful for most folks. If you are willing to live with quirky software, ColorEyes Display gets our nod as the overall best performer.
Monaco Optix XR Pro: The Monaco Optix XR Pro package is well thought out and performs exceptionally well. ColorEyes Display has the edge on excellent monitors, but Monaco's software is more forgiving of lesser quality screens. Note the "Pro" version is required to gain access to LUT based profiles and using the instrument to set various screen parameters. The XR sensor, the X-Rite DTP-94, is superb and the software very good. Ignore its recommendations on where to set display luminance if it appears drastically too dark. This is a quirk that surfaces on some systems. The XR Pro package adds display trending and matching tools that can be of use in large offices. The previous 5 installation limit at a given site has been done away with. Highly recommended.
GretagMacbeth Eye-One Display 2: The new GretagMacbeth Eye-One Display 2 + Eye-One Match 3.0 is nearly the equal of the Optix XR. The Eye-One adds DDC capability (as of early 2004 a Mac-only feature) to automatically adjust the monitor settings, eliminating the bother of manual adjustments. Another automatic goodie is the software locates where the sensor is on the screen and positions the measurement window accordingly. It is fast, accurate, and does an impressive job. It also includes an ambient light measurement feature that can be used to verify the color temperature of your viewing light. Thankfully, the old Eye-One Match problem of not allowing the white luminance to be specified is fixed. The recommended luminance levels are often too high; drop down to a more sensible value if necessary. A quibble is that the possible luminance values are given in steps of 10. This precludes selecting useful values for CRT monitors such as 85 or 95 cd/m^2. Avoid the "Easy" mode, as this chooses the default, overly bright settings.
The Eye-One 2 produces a touch less shadow resolution than the Optix XR on most monitors. On good quality laptop screens, the tables turned, with the Eye-One 2 giving the best B&W marks. The midtone and highlight response is superb. If you are profiling a lesser quality laptop screen (typically 12" and under for both Mac and PC), be sure to check the release notes on making a single-gamma profile. This helps make these displays more neutral. Do not use the single gamma option for better displays, as overall color accuracy is reduced. The calibration options for CRT monitors that cannot be controlled by the software are limited, particularly the setting of the monitor black level. Highly recommended for most monitors.
Update: The Eye-One 2 uses tiny suction cups to attach the puck to CRT screens. This has the advantage of leaving less marks on the face of anti-glare coated screens. Our initial findings were that these cups lost their grip after a few month's use. This proved to be the result of having an early version of the Eye-One 2 to test. Newer versions use a different material for the cups and have held on tenaciously even after intentionally getting the suction cups dirty and dust-covered. After a quick wipe with a microfiber cleaning cloth, the puck stuck to CRT screens well.
ColorVision Spyder2: The ColorVision Spyder2 bears little resemblance to the original Spyder. That is a good thing. The tonality and accuracy of shadow colors is excellent. Grayscale accuracy is, overall, exceptional. The exception was on some laptop monitors. The Dell UltraSharp was one example. Here the Spyder2 was good but not outstanding. A strength of ColorVision has always been their feature-laden software. This still holds for the new offering. The Spyder2Pro software is flexible and powerful with most of the features I deem essential. One exception is not calibrating to the native gamma for CRT screens, but a draw your own gamma curve gives even more power.
The Spyder2 exhibits two quirks. First, it is slow. You can make a sandwich and read most of a daily paper in the time it takes to measure a monitor. Second, the exceptional shadow performance does not carry over into the highlights. The Spyder2 does not measure any light colors - only up to the primary (e.g. [255, 0, 0] for red) values. The profiles appear to reflect this, with more abrupt highlight transitions than those made by GretagMacbeth or Monaco equipment. If shadow detail and accuracy are paramount, the Spyder2 is the one to buy. Wedding photographers faced with the combination of bright lights and slightly off-white gowns should look elsewhere. The Spyder2 is not as sensitive to ambient light levels as the original, but it is still advisable to perform the measurements in a darkened room. Recommended. Would be highly recommended if the highlight performance matched that of the midtones and shadows.
GretagMacbeth Eye-One Display (original version): The Eye-One Match 3.0 software is compatible with both the Eye-One One and the Eye-One Two. Aside from silly naming conventions, the original Eye-One Display held its own. Overall, the original Eye-One Display rates a notch below the Spyder2. It generally surpassed the Spyder2 in dealing with extreme highlights as colors faded to white. Owners of the original Eye-One Display hardware should download the free upgrade to Match 3.0.
GretagMacbeth Spectrolino: At something over $6000 list, the GretagMacbeth Spectrolino + ProfileMaker Pro software is not realistic to use for monitor profiling alone. The ProfileMaker software affords more options and allows precise fine-tuning of the calibration parameters. This venerable instrument was used as a baseline for comparison. It is an expensive, heavy, slow, and superbly accurate spectrophotometer. Overall, the tide looks to be shifting towards colorimeters. The overall performance falls in the range of the Spyder2. A Spectrolino is still a fine instrument, but unless you need oddball printer profiling capability as well, there are better ways to spend your money.
ColorVision Spyder (original version): The original ColorVision Spyder simply was not competitive with the above products. ColorVision deserves much credit for bringing the first reasonably priced monitor calibration and profiling tool to market. The Spyder certainly remains better than calibrating by eye. Shadow resolution, grayscale neutrality, color accuracy — the original Spyder falls short in all these areas. There are better options.
LaCie Blue-Eye One: I can think of no reason to purchase the LaCie Blue-Eye system. It uses the same hardware as the original Eye-One Display, but the software does little to take advantage of it. The Blue-Eye costs more than the DDC-enabled Eye-One 2 and does less. Not recommended.
Eizo ColorNavigator: The Eizo ColorNavigator software comes for free with the monitor. You need to supply your own Eye-One. The grayscale was OK, and the gamma curve nearly perfect. Color response was not. Use the Eye-One with the new version of GretagMacbeth software.
On high quality monitors, ColorEyes Display gives the best balance of accuracy, smooth tonality, shadow resolution, and calibration flexibility. Using the software effectively takes careful reading of the user's manual. This is made more difficult because the installation program does not copy the help file to your computer. There is also no on-line help except for a few prompts which may or may not have anything to do with what is shown on-screen. ColorEyes offers automatic DDC control over the hardware settings for selected monitors. Once you have mastered its quirks, the calibration and profiling of good quality monitors is simply superb. Getting the most out of ColorEyes Display requires either using the bundled version with the Monaco XR/X-Rite DTP-94 puck or the GretagMacbeth Eye-One 2. Although the software supports older measurement hardware (e.g. Sequel Squid, Eye-One Display One), the profile quality can not compare.
You can not go wrong with the Monaco Optix XR Pro. It matches well with most monitors, and offers useful data analysis and trending capabilities. If you do not need display matching, performance trending, and other workgroup features, the added cost of the Pro package is significant. Unfortunately, the base version with its reduced feature set is not competitive.
The GretagMacbeth Eye-One Display 2 is a good performer. Overall color performance is excellent. The calibration options are not as extensive as those of the Optix XR or Spyder2. If your monitor is DDC-enabled, the Eye-One Match 3 software performs the necessary monitor adjustments automatically (Mac only for now - Windows capability is "coming soon"). The fast speed, particularly on CRT monitors, makes for quick and painless calibration and profiling.
The ColorVision Spyder2 is a mixed performer. The limitation is in how highlights are handled. If your photography requires excellent shadow response or superb B&W performance, the Spyder2 is ideal. The included software offers the most flexibility of any of the recommended systems. Wedding photographers, or others needing to see delicate highlight details will do better with one of the above systems.
© 2020, Dry Creek Photo. All Rights Reserved.