Canon x300 Media Configuration Tool

[shewhorn]

I'm wondering if anyone a) knows the answer to this question and/or  has any suggestions...

So I have a bunch of papers that I'm adding via the Media Configuration Tool. You get to the point where it prints out the various test prints at different ink levels. I find the patterns that are printed to be barely useful so I'm trying to create my own test pattern so I can perhaps approach this from a more scientific standpoint. At the very least I would like to be able to measure a black patch. You'd think Canon would have put a patch of black in their standard test print that one could measure with a spectro but nope. I figure at the very least this can tell me whether or not laying down more ink is giving me blacker blacks. I intend to have some other test patches of saturated colors in there as well. I figure if I measure them all and I hit a point where laying down more ink isn't making a difference, then there isn't much of a point in laying down any more ink.

Okay... on to my problem. My test image is a 16 bit TIFF, 300 ppi @ 23"x6.333..." (6900x1900 pixels). When I go to select that image in the MCT it tells me I can't use that TIFF image but it gives no reason as to why. Is it bit depth, is it the size??? The manual makes no mention of this.

Cheers, Joe

[Scott Martin]

I find the patterns that are printed to be barely useful so I'm trying to create my own test pattern so I can perhaps approach this from a more scientific standpoint. At the very least I would like to be able to measure a black patch.

I appreciate your frustration and desire for something more scientific. I've been working with RIPs and tweaking optimal media densities since the mid 90s and you'd think there would be a more impressive implementation than this. There are ways of scientifically determining optimal ink limits without subjective visual analysis.

One thing you can do (if you have the full version of ColorBurst) is to use ColorBurst's Linearization target for the MCT test image. Measure your test prints in CB SpectralVision utility and load the saved Lin file in CB. From there you can graph out the results and look for the density plateau, or even better, the chroma plateau. Yeay for science!

I figure if I measure them all and I hit a point where laying down more ink isn't making a difference, then there isn't much of a point in laying down any more ink.

The maximum density usually isn't necessarily the optimal one. When we compare ink densities next to chroma values you'll find that they increase together up to a certain point at which the chroma will start to fall as the density continues to increase. This chroma spike usually occurs pretty close to maximum density but the difference is important.

I look forward to the day when we can stick a new piece of paper into the printer and have the printer itself determine the optimal total ink limits and linearizations based on actual measurements from the media. The technology is here, the printer implementation is not (not even on the HP Z series that don't go this far).

Okay... on to my problem. My test image is a 16 bit TIFF, 300 ppi @ 23"x6.333..." (6900x1900 pixels). When I go to select that image in the MCT it tells me I can't use that TIFF image but it gives no reason as to why. Is it bit depth, is it the size??? The manual makes no mention of this.

The size is definitely too large! The max size is something like 4x7 inches at 300ppi - consult the literature on this as I can't recall off the top of my head. Make a small 4x5 inch 300ppi file and save it as both 16 and 8 bits and print them both. Then you'll have your answer as to bit depth!

Check out the "Onsight Media Selection Image" at http://www.on-sight.com/downloads/ It's not bad for basic measurement and the subjective visual analysis of linearity, gamut, shadow detail, and ink bleeding and smudging.

[JeffKohn]

I think there are some limitations on the TIFF format that the MCT can read, although it's been a while since I did this so I don't remember the exact details. If your TIFF has LZW or ZIP compression, try re-saving it without any compression. I can't remember if it has to be 8-bit or not. I don't recall there being any limitation on size, although I suppose there may be some limit on total pixel dimensions that I never encountered.

My approach was to use a 1-page profile test chart, and then create a profile from each of the ink settings and look at gamut, black point, etc. Once I identified the best one I'd re-profile using a larger patch count.

[Scott Martin]

My approach was to use a 1-page profile test chart, and then create a profile from each of the ink settings and look at gamut, black point, etc. Once I identified the best one I'd re-profile using a larger patch count.

Ah yes - the Bill Atkinson approach! That's even slower and more tedious but works well. It's fun to watch the gamut shape morph slightly from one to another.

[shewhorn]

Ah yes - the Bill Atkinson approach! That's even slower and more tedious but works well. It's fun to watch the gamut shape morph slightly from one to another.

Thanks for the replies everyone. I don't have the Colorburst RIP so I'm going to have to get creative. It would be nice if Canon made the brute force method of making media types for every iteration of ink output a bit easier but they kind of want you to run the print feed calibration every time which just makes it a bit tedious. Canon support says there's no way around it. I understand that there's a different version of the Media Configuration Tool for paper manufacturers and I'm asking around to see how one might get a hold of that. So far Canon tech support doesn't know.

Cheers, Joe

[Scott Martin]

I understand that there's a different version of the Media Configuration Tool for paper manufacturers and I'm asking around to see how one might get a hold of that. So far Canon tech support doesn't know.

Got it - don't bother!  Stay away from this stuff if tedious doesn't equal fun.

[shewhorn]

Got it - don't bother!  Stay away from this stuff if tedious doesn't equal fun.

I can handle a bit of tedious if need be. Work is mostly fun... not always fun but mostly fun. If I didn't have to build media types for about 15 different papers I'd definitely go the brute force route but I'm still holding out hope for a somewhat more sophisticated approach.

[Scott Martin]

I can handle a bit of tedious if need be. Work is mostly fun... not always fun but mostly fun. If I didn't have to build media types for about 15 different papers I'd definitely go the brute force route but I'm still holding out hope for a somewhat more sophisticated approach.

I wonder if you could do the scientific approach with the demo version of ColorBurst? You wouldn't be printing through Colorburst, and spectralVision doesn't require a dongle.

[shewhorn]

I wonder if you could do the scientific approach with the demo version of ColorBurst? You wouldn't be printing through Colorburst, and spectralVision doesn't require a dongle.

I was actually browsing around their website and had that thought as well. One little detail though, it would appear as if Colorburst only supports Epson printers at the moment.

Cheers, Joe

[Scott Martin]

I was actually browsing around their website and had that thought as well. One little detail though, it would appear as if Colorburst only supports Epson printers at the moment.

Doesn't matter because you won't be printing out of ColorBurst at all - you'll be printing from the MCT tool and analyzing the results in CB.

[Mulis Pictus]

Doesn't matter because you won't be printing out of ColorBurst at all - you'll be printing from the MCT tool and analyzing the results in CB.

BTW, do you know if there's a RIP which support ipf8300 and is able to use all 7 colors (CMYKRGB)? I am also fighting a bit with the media configuration tool and would like to try demo version of such RIP to compare the process.

[Scott Martin]

BTW, do you know if there's a RIP which support ipf8300 and is able to use all 7 colors (CMYKRGB)? I am also fighting a bit with the media configuration tool and would like to try demo version of such RIP to compare the process.

You mean all 12 inks? THere are lots of RIPs that support Canon printers now including those from CGS, EFI, Ergosoft, GMG, ONYX, Wasatch, but I wouldn't bother. The driver is where it's at, especially on the x300 printers. If the MCT tool is frustrating you, you'll really be frustrated with a RIP.

[Mulis Pictus]

You mean all 12 inks?

Yup, 12 inks of 7 hues (pardon my English if it is not exact :-)

THere are lots of RIPs that support Canon printers now including those from CGS, EFI, Ergosoft, GMG, ONYX, Wasatch, but I wouldn't bother. The driver is where it's at, especially on the x300 printers. If the MCT tool is frustrating you, you'll really be frustrated with a RIP.

I hope to be not, but not sure as I didn't try it yet. What I would like to have is a linearized output on third party papers. One would think Canon/Epson/HP would provide it with driver software, when there are devices (spectrometers or densitometers) capable of it built in the printers. But it looks like it is their business strategy to drive us away from these papers as much as possible.

What annoys me on MCT, is that there's not much info on what are the ink limits for various presets, whether special 1-5 and 6-10 differ in total ink limit, or also in limiting black differently from other colors, etc. (I have tried sp6 to sp10 on Tecco matte paper here and it looks like the black is limited similarly in sp6 and sp9?) I have looked on few of the RIPs you mentioned, but it is not much clear from their web pages, whether they drive the printer ourself or thru the driver and whether they are capable of using the secondary colors (RGB in Canon's case). So far I have only found ColorBurst stating that it can use additional inks besides CMYK so I might try it on our Epson 7900 to get some experience. I was hoping to try it on ipf8300 though.

[shewhorn]

Doesn't matter because you won't be printing out of ColorBurst at all - you'll be printing from the MCT tool and analyzing the results in CB.

I might give that a try then. I've somewhat resigned myself to the fact that I'm going to have to just spend the time and money to create a bunch of different media types. It just wouldn't sit right knowing that there's a chance I could have done a better job but did not spend the time to do it.

Cheers, Joe

[jgbowerman]

I might give that a try then. I've somewhat resigned myself to the fact that I'm going to have to just spend the time and money to create a bunch of different media types. It just wouldn't sit right knowing that there's a chance I could have done a better job but did not spend the time to do it.

Cheers, Joe

I don't use, nor am I familiar with ColorBurst, but I vote for the Bill Atkinson approach if you build your own custom paper/printer profiles. Using ColorThink, one can then do an objective evaluation of the different ink settings in the x300 MCT.

My take on ink settings is "Medium High" looks to be a very good choice when using third party fine art papers. After implementing the MCT, choose fine art mat or fine art glossy/semi-glossy, click the assist button, input the paper weight and go with Canon's media choice, use the Automatic settings to start with, and configure with different ink levels (I'm doing two configurations, Standard and Medium High), then print the profile targets. Next, create the profiles with the necessary software and take the two profiles into ColorThink for an objective evaluation. Looking at the two target profile prints under a viewing lamp serves as a good subjective evaluation.

I have found the recommended Special 5 setting for H. fine art glossy papers to be inferior to the more customized approach. Sure it takes time, paper, and ink, but if you want the best (as it sounds you do), go for it. The MCT is a remarkable improvement with the x300 printers.

[Scott Martin]

I don't use, nor am I familiar with ColorBurst, but I vote for the Bill Atkinson approach if you build your own custom paper/printer profiles.

You vote to do it that way even though you've never tried it the other way which is purported to be faster and even more scientific? Why not give it a shot and then vote?

[jgbowerman]

You vote to do it that way even though you've never tried it the other way which is purported to be faster and even more scientific? Why not give it a shot and then vote?

I can only "vote" for something I have familiarity. I'm not opposed to trying ColorBurst, I might very well look into it further when time permits. However, I am trying to get by with what I already have. The more software that comes available, it is clearly getting to the point one has to make a choice of what not to try and be happy with what one already has. Simply keeping up with new versions of the same software is a time consuming chore in itself.

[Scott Martin]

I'm not opposed to trying ColorBurst, I might very well look into it further when time permits. However, I am trying to get by with what I already have. The more software that comes available, it is clearly getting to the point one has to make a choice of what not to try and be happy with what one already has. Simply keeping up with new versions of the same software is a time consuming chore in itself.

If time is an issue and you're wanting to determine the optimal media setting with the MCT tool then the above mentioned procedure with the demo version of ColorBurst will save you a bunch of time versus making profiles and analyzing them in ColorThink. There's no need to buy ColorBurst or actually print with it - it can be used for a scientific analysis of the different options the MCT tool provides. I sure understand your frustration with all this time consuming learning curve stuff!

[jgbowerman]

If time is an issue and you're wanting to determine the optimal media setting with the MCT tool then the above mentioned procedure with the demo version of ColorBurst will save you a bunch of time versus making profiles and analyzing them in ColorThink. There's no need to buy ColorBurst or actually print with it - it can be used for a scientific analysis of the different options the MCT tool provides. I sure understand your frustration with all this time consuming learning curve stuff!

Onsight, I checked out ColorBurst Pro and see it is only compatible with the Windows platform. I have a Mac, and I see ColorBurst X-Proof is an option, will it provide the tools I need to determine the optimal setting with the MCT?

[Scott Martin]

Onsight, I checked out ColorBurst Pro and see it is only compatible with the Windows platform. I have a Mac, and I see ColorBurst X-Proof is an option, will it provide the tools I need to determine the optimal setting with the MCT?

Yes, it's all there. What I'm not completely certain about is if the demo version let's you do it all without a dongle. I think so...