Canon IPF6400 resolution

[samueljohnchia]

The printer driver can be set to 300 PPI or 600 PPI.

More specifically, the option to set the driver (or plug-in) to 300dpi (the silly driver does not label it as ppi which it should) is only available for certain media settings like "Plain Paper". I would assume the OP is printing with a different media setting than that for the high-end photo books. So only the 600ppi setting is available. All the files will be resampled by driver, whether you like it or not, to 600ppi.

Anyway the resolution of the files has absolutely nothing to do with the dot pattern/screening. That is controlled by which pass mode is selected in the driver.

[Bart_van_der_Wolf]

More specifically, the option to set the driver (or plug-in) to 300dpi (the silly driver does not label it as ppi which it should) is only available for certain media settings like "Plain Paper". I would assume the OP is printing with a different media setting than that for the high-end photo books. So only the 600ppi setting is available. All the files will be resampled by driver, whether you like it or not, to 600ppi.

Indeed, and that's why it is important to first make sure that the input image data was resampled to 600 PPI (and output sharpened), because otherwise the printer driver will resample (with a lower quality resampling method) and not sharpen.

Anyway the resolution of the files has absolutely nothing to do with the dot pattern/screening. That is controlled by which pass mode is selected in the driver.

Correct, the dithering does not materially influence the resolution but rather the smoothness of gradients and the creation of (more) subtle color nuances. As your macro shot clearly demonstrates, a very close-up inspection will reveal a difference in smoothness, but it would be harder to see from a bit of distance. The real difference in resolution comes from the input image data, it's either there or it isn't. All we can do is to keep what's there and not squander it by using a sub-optimal workflow and driver settings.

Cheers,
Bart

[samueljohnchia]

Indeed, and that's why it is important to first make sure that the input image data was resampled to 600 PPI (and output sharpened), because otherwise the printer driver will resample (with a lower quality resampling method) and not sharpen.

Correct, the dithering does not materially influence the resolution but rather the smoothness of gradients and the creation of (more) subtle color nuances. As your macro shot clearly demonstrates, a very close-up inspection will reveal a difference in smoothness, but it would be harder to see from a bit of distance. The real difference in resolution comes from the input image data, it's either there or it isn't. All we can do is to keep what's there and not squander it by using a sub-optimal workflow and driver settings.

Cheers,
Bart

Absolutely, we should be seeing the big picture, and optimize the quality in every way we can. But from the way the OP wrote about his struggles with dotty output from the Canon, he has my sympathy. It does not sound like he is having less than perfectly sharp prints. Rather dotty looking output.

I am able to quite clearly see the differences in Epson and Canon screening patterns with my naked eyes (the prints must be well illuminated). Often I find it much easier to judge the degree of dottiness without a loupe. The Highest (max. no. of passes) 16-pass setting brings the latest X400 generation Canon printers close to the Epson X900s and X880s in dither smoothness, but no so in the highlights. Arguably I prefer the Canon dither in the shadows. The Highest setting, as shown in my picture looks way worse to the naked eye than the macro image reveals. Some of my non-photographer friends can easily see the difference in dot smoothness with just their eyes too.

Aligning the heads perfectly can be a little challenging on these Canon printers. My experiments have shown that only the Initial Head Alignment (automatically performed by the printer) is capable of adjusting the dot accuracy sufficiently for the 16 pass setting. All the other auto alignments result in suboptimal dot placement accuracy. The manual alignment is not perfect either, despite how careful I am. Manual alignment does not print out the yellow channel. Less than perfect alignment does increase the dottiness in smooth areas quite visibly. Unidirectional printing improves the dot placement accuracy a little. Marcin is also right about the head height setting. It should be set to the lowest possible height that won't impact the paper. For many roll fine art papers on the iPF6450 that I tested, setting to "Low" is fine. Beware the initial couple of inches of the roll - the paper path direction is counter to the natural curl of the paper, so the leading edge will be raised slighly as a hump. Sometimes head strikes will occur only on the leading edge and be perfectly fine after for very curly papers.

[Czornyj]

Here is a macro photograph (1:1) of the Highest and Highest (Max. No. of Passes) dot pattern in the smooth area of Bart resolution test target. Download it and make sure you view at 100% magnification so your display does not downsample the screen preview.

I actually meant is that there's no difference in raster pattern resolution, the print head is physically 1200dpi and neither the size nor pattern of dots doesn't change. What really changes in 16 pass mode is the precision of dot placement, but to my eye the difference in perceptible graininess is visible, but not significant.

In the above mentioned case there are two possibilities - something's wrong with print settings and paper calibration, or someone expects same results as in case of SPx880/890/900. The latter is impossible to acheive on iPF, mainly due to different channel curves - and this is exactly what you found disturbing in bright areas:


The only way to counteract this is to use a RIP and create channel curves that put more bright inks in bright parts of the images. TBW makes it possible for B&W works, unfortunately color RIP solutions are way less affordable...

[Bart_van_der_Wolf]

Absolutely, we should be seeing the big picture, and optimize the quality in every way we can. But from the way the OP wrote about his struggles with dotty output from the Canon, he has my sympathy. It does not sound like he is having less than perfectly sharp prints. Rather dotty looking output.

I'm not sure, because the OP stated:

I created file with native 600ppi resolution and set paper type to gloss and it worked, the image si much better now. Still have to investigate furtherr. But 600ppi files are huge, i never had to create native 600ppi files with Epson. 

This suggests, to me anyway, that he now first resampled to 600 PPI and got better results. That may mean that the prior image data was of lower PPI, and thus (lower quality) resampled by the printer driver. Maybe that resulted in blocking artifacts that contributed to the sensation of 'lower resolution'. That, combined with the somewhat more visible dithering pattern, may give a  sensation of graininess.

If it's the resampling that did the trick, then adding additional output sharpening would visually close the gap, if any, even further.
 
Cheers,
Bart

[DryAxE]

Well i haven't responsed some time beacuse i was busy testing and calibrating.

By any mean i dont think Canon print is bad, i just noticed it is slightly more grainy. Maybe that nonvariable dot seen on Czornyj post explains it. We are speaking in looking really close differences not important with wall prints but with photo book prints higher detail is just better.

But i must also explain that the high resolution print from Epson 3880 was made on Hahnemuhle gloss 260gsm paper and i do not have any left to do the same test print with Canon ipf6400. I was making comparable prints on "high resolution" but still really cheap 140gsm matt paper. I have already printed books with that 140gsm matt paper in past and with combination of Epson L1800 dye printer the prints were really detailed.

I have made a lot of printing from Epson 3880 (pigment) and Epson L1800 (dye) so i noticed that 3880 had more accurate colors (both calibrated with Colormunki), especially blacks were way more neutral. On the other hand colors on L1800 were more saturated and image detail was higher due to smaller dot (1.5 vs 3.5pl). Also on dye there was no bronzing or gloss differential issues and that is a big drawback for pigmented inks especially for book making. Dyes are also less prone to physical damage like scratches or fingerprints.

So i made upgrade on my pigmented printer (from 3880->ipf6400) but i also believe i will have to upgrade to some better dye printer. L1800 has great output quality but suffers from poor build quality and not the best inks. It also lacks roll and has poor feeding mechanism --> even if i manually help the printer to securely grab one large sheet it curves/warps the print anyway. Not visible on print but when you cut it in correct rectangle there is up to 2mm warp difference on 60cm large print. When you print the photos with even white band arround it that can become visible. Not to mention i have to make this manual help every 15 minutes (time between prints) and that printer suffers from head clogs so i have to make a lot of head cleanings..

So printer that would be ideal to my use would have 24" width, dye Claria or Ultrachrome D6 inks, superior image detail and colors. I know there is Epson D3000 dye drylab printer. It has all i have mentioned except width, but even 30,5cm would be fine. But cost of this printer is another story. Any other suggestions?  

My goal is to get the best possible prints for my photobooks -->resolution, color, no bronzing, no gloss differential, scratch resistant, longevity,...

I will still continue to research Canon ipf6400 printer drivers to learn and to find the best way to get the quality i want.