Color Muse

[Cornishpastythighs]

http://the-gadgeteer.com/2016/11/16/colormuse-review/

Has anyone had any experience of this product in a printing environment? It looks like its made for paint and fabrics but sells itself as a $60 Spectro. Wondered if anyone had tried one to compare ink colours
Cheers

 

[gordo]

I've reached out to the company about this and will let you all know if I get a response.

 

[Louis Dery]

I have one and made some test with it. As far as color comparison between 2 colors, it works pretty well. It also allows to get D50 Lab value readings. Made comparison with I1Pro2... not too bad ;-)
What is missing: does not allow to import your own color library made from colorimetric data. You can only make a library from mesurements.

 

[gordo]

Can you give a screen grab of what it displays when measuring a patch? And when it compares 2 patches? How large must the target patch be? Is it easy to get the device to measure a specific patch and not any surrounding areas?

 

[Deleted member 16349]

I looked at the web site https://colormuse.io/pages/support

In the technical section, they give some details. What I was interested to see was what the sensors were. In the technology section, they list the following,

Tri-stimulus XYZ sensor

I find this interesting because I didn't know a "tri-stimulus XYZ sensor" existed. Not sure what this really means. I just wanted to make sure they were not just using an RGB sensor of some kind.

 

[Orpheus42]

"I find this interesting because I didn't know a "tri-stimulus XYZ sensor" existed""...... I think its called a human eye .......sorry couldn't resist that

 

[JohnTheMathGuy]

The aperture of the Color Muse is about 1 cm and the Nix is about 1.4 cm.

The Color Muse and the Nix both use a sensor developed in Germany by the company MazeT. The sensor is small and cheap.

The tristimulus sensor leads to one limitation for both devices. Even if the sensor were an absolutely perfect match to the tristimulus functions, it can only see the sample under the lighting that the device has. If a spectro is used, you can predict D50 or illuminant A or F10, or whatever. With a tristimulus based sensor, you have to resort to the Bradford transform, so there is error introduced.

Both products use white LEDs as the light source, so that is the illuminant that you are measuring under. But, the white LED used in both is different from the white LEDs in a flashlight. They both use multi-phosphor LEDs that don't have a deadspot in the spectrum just below 500 nm. In other words, they see color better. Both are perhaps similar to D50, although the Color Muse is a bit closer.

The biggest issue that I saw when I evaluated them was that the lighting in not 45/0 in either one. In the Color Muse, they made an attempt at 45/0, but in reality, it is closer to a spherical instrument. The Nix is actually closer to 45/0, but they have a large amount of light coming in at close to perpendicular. What this means is that it is not possible to calibrate the instrument to match a standard instrument on samples with different gloss. It may measure a glossy surface well, but won't do so well with a matte surface. Or the other way around.

Another issue is that neither of the devices have any UV component to the lighting. That means that you have basically an M2 instrument.

I measured a collection of white / gray / black objects with various amounts of gloss and OBAs. I found errors of up to 4.0 DE76 for the Nix. Errors were fairly small when you measure a Pantone book, small enough that I suspect that they calibrated on a glossy stock with a moderately high OBA content. The Color Muse had errors of up to 7.0 DE76 on the same samples.

Is this acceptable for printing? Perhaps for monitoring color during the run, but certainly not for checking at makeready.

 

[cementary]

Is this acceptable for printing? Perhaps for monitoring color during the run, but certainly not for checking at makeready.

Good god, now i can read John's posts not only on blogspot
So, let's count some: Pritchard, Nikkanen, Seymour...
John, i think when Widmer will get tired from his retirement you should show him this forum. Then the four of you could make a really cool iso-printing-color-consulting company

 

[Deleted member 16349]

So, let's count some: Pritchard, Nikkanen, Seymour...
John, i think when Widmer will get tired from his retirement you should show him this forum. Then the four of you could make a really cool iso-printing-color-consulting company

Don't panic. It's just a dream, maybe a nightmare even. You will wake up and all will be well.

 

[Louis Dery]

Can you give a screen grab of what it displays when measuring a patch? And when it compares 2 patches? How large must the target patch be? Is it easy to get the device to measure a specific patch and not any surrounding areas?

You can download the app for iphone or Android et see yourself the UI. It is basic but you get Lab, RGB, LCH, etc.

 

[gordo]

You can download the app for iphone or Android et see yourself the UI. It is basic but you get Lab, RGB, LCH, etc.

I did. But it didn't show the Lab, RGB, LCH, values for the paint colors it had built in.

 

[Louis Dery]

I did. But it didn't show the Lab, RGB, LCH, values for the paint colors it had built in.

Hi Gordo,
I see what you mean. If you make measurement, you will get the color infomation of the color you measured. But from the "paint color" example, there is no way to see the Lab, RGB, etc.
Also, there is no way to share palette with others. The only way to share is via Adobe ASE file but cannot be aded to the library on the iphone or Android app. ;-(

 

[gordo]

The aperture of the Color Muse is about 1 cm and the Nix is about 1.4 cm.

The Color Muse and the Nix both use a sensor developed in Germany by the company MazeT. The sensor is small and cheap.

The tristimulus sensor leads to one limitation for both devices. Even if the sensor were an absolutely perfect match to the tristimulus functions, it can only see the sample under the lighting that the device has. If a spectro is used, you can predict D50 or illuminant A or F10, or whatever. With a tristimulus based sensor, you have to resort to the Bradford transform, so there is error introduced.

Both products use white LEDs as the light source, so that is the illuminant that you are measuring under. But, the white LED used in both is different from the white LEDs in a flashlight. They both use multi-phosphor LEDs that don't have a deadspot in the spectrum just below 500 nm. In other words, they see color better. Both are perhaps similar to D50, although the Color Muse is a bit closer.

The biggest issue that I saw when I evaluated them was that the lighting in not 45/0 in either one. In the Color Muse, they made an attempt at 45/0, but in reality, it is closer to a spherical instrument. The Nix is actually closer to 45/0, but they have a large amount of light coming in at close to perpendicular. What this means is that it is not possible to calibrate the instrument to match a standard instrument on samples with different gloss. It may measure a glossy surface well, but won't do so well with a matte surface. Or the other way around.

Another issue is that neither of the devices have any UV component to the lighting. That means that you have basically an M2 instrument.

I measured a collection of white / gray / black objects with various amounts of gloss and OBAs. I found errors of up to 4.0 DE76 for the Nix. Errors were fairly small when you measure a Pantone book, small enough that I suspect that they calibrated on a glossy stock with a moderately high OBA content. The Color Muse had errors of up to 7.0 DE76 on the same samples.

Is this acceptable for printing? Perhaps for monitoring color during the run, but certainly not for checking at makeready.

Nix is about 10 times the price of Color Muse. Is the performance equivalent?

 

[NathanD]

I measured a collection of white / gray / black objects with various amounts of gloss and OBAs. I found errors of up to 4.0 DE76 for the Nix. Errors were fairly small when you measure a Pantone book, small enough that I suspect that they calibrated on a glossy stock with a moderately high OBA content. The Color Muse had errors of up to 7.0 DE76 on the same samples.

John, when you mention these errors, what is the reference? A dE of 4 doesn't mean much if the reference is a measurement from another device that doesn't measure color accurately.

In a follow up post, I will share results of an experiment we conducted last year.

 

[Cornishpastythighs]

I've reached out to the company about this and will let you all know if I get a response.

gordo, did these people ever get back to you?
Cheers

 

[gordo]

gordo, did these people ever get back to you?
Cheers

Sort of. I contacted John Kowalski their VP of Marketing who got back to me - unfortunately he had just left the company. So now I'm reaching out to his successor.

 

[JohnTheMathGuy]

John, when you mention these errors, what is the reference? A dE of 4 doesn't mean much if the reference is a measurement from another device that doesn't measure color accurately.

In a follow up post, I will share results of an experiment we conducted last year.

I was comparing measurements to a Spectrolino. Admittedly not the most accurate device, but I have compared this unit against an eXact, a Spectrodens, and a 939, and the differences on white/gray/black samples was below 1.0 DE.

 

[JohnTheMathGuy]

Hi Gordo,
I see what you mean. If you make measurement, you will get the color infomation of the color you measured. But from the "paint color" example, there is no way to see the Lab, RGB, etc.
Also, there is no way to share palette with others. The only way to share is via Adobe ASE file but cannot be aded to the library on the iphone or Android app. ;-(

Somehow, I got the CIELAB values directly from the app. I don't have the unit with me at the unit, so I can't try it now. I recall that maybe there was a compare color mode... which computes a deltaE from a reference that you type in?

 

[JohnTheMathGuy]

Good god, now i can read John's posts not only on blogspot
So, let's count some: Pritchard, Nikkanen, Seymour...
John, i think when Widmer will get tired from his retirement you should show him this forum. Then the four of you could make a really cool iso-printing-color-consulting company

From his Facebook posts, I think Widmer is having way too much fun travelling to want to do anything serious again! Lots of gorgeous pics from all over the world.