How can Hobbicolors supply only one set of dye inks for all Canons?

[ghwellsjr]

I know that there is a difference in Canon OEM magenta and cyan inks for BCI-6 and CLI-8 cartridges, so how can Hobbicolors offer just one set of inks (UW8) for all the newer cartridges and the older BCI-6 cartridges? They used to supply a set of inks (WC6) for BCI-6 cartridges but they have dropped that line stating that the newer ink is more vibrant and has an ultra wide gamut. This sounds like a universal ink approach to me which is widely condemned on this forum.

Any comments?

 

[turbguy]

I've noticed that Hobbicolors produce heavily green-toned B&W images on my i960, and a nice warm brown tone on the B&W images from my i9900, but the color prints are just great from both!

I guess the physical properties are what counts for 'working', and there's no substitute for profiling...

Wayne

 

[ghwellsjr]

I supply lots of refilled ink cartridges for many of my friends. I don't know how to profile and I would rather come off with a solution that doesn't require it of my friends. I didn't have this problem with Inktec ink.

 

[Grandad35]

Here is a comparison of the OEM Magenta and Cyan inks:

The shorter wavelengths are on the left and the longer wavelengths are on the right (lifted from a previous post). While both inks differ, the magenta is really different. Note that the change in the Magenta is not to make it "stronger" - it completely changed its color by absorbing less of the shorter wavelengths (actually making it "weaker"). This was probably done to allow a wider color gamut from the printer, but the internal printer RIP must know about the color change or it will throw off all every color that uses magenta (as already noted).

Since older printers have their internal RIP designed for the BCI-6 ink, your only option is to use a custom printer profile. Profiles are meant for relatively minor corrections, and can help in this type of situation, but don't expect miracles. It's like having a car that pulls strongly to the right, and "fixing the problem" by simply remounting the steering wheel so that it is straight when the car goes straight. You can drive it, but you won't want to make a high speed left turn (the tires also wear out faster, etc.).

Using profiles isn't all that bad - see this post.

 

[pharmacist]

Somehow Hobbicolors managed to manufacture quite a good universal Canon ink, fit for all types of Canon cartridges. I know the colours are just fabulous and extremely vibrant and without any noticable colour cast (except maybe the B/W prints). I have this ink refilled in my Canon MP780 (BCI-6), MX850 (CLI-8), IP6600D (CLI-8), i9950 (BCI-6) and even without a dedicated profile the prints are for those not used to handle printer profiles are very neutral to the naked eye and vibrant. All the aforementioned printers do use a different internal RIP to cope with the inks. So the trick of Hobbicolors UW-8 as a universal ink does its work very well.

 

[ghwellsjr]

Grandad, can you point us to the original post where these plots were lifted? I cannot find the results of the work you and canonfodder did years ago. Is that where you lifted these plots?

I'm afraid that getting my friends to use profiles, when I can't even get them to refill using the German method, is out of the question. But I don't blame them, even your simple explanation looks very complicated and I have no interest in attempting this myself.

 

[mikling]

After messing with various inks and hundreds of combinations for a project I had what i can say is that messing around with ink colors is not an easy task and it is easy to make conclusions that are wrong very easily. Many many times I thought that I had finalized my formulations because the images I was using looked good until further testing revealed a hole! Back to the drawing board again. Many times, a revision in one aspect producing better results in one area resulted in obvious flaws in different images weeks later.... only to result in revisions again.

Therein lies the problem, I don't have the resources of Canon or Epson and I can definitely fully appreciate how much work the engineers there go through to improve the performance of what looks to be just colored water. As a result I am highly skeptical that an all in one solution exists because 1. Canon would have gone that route to save costs. 2. If an improvement in gamut was easy to come by they would have done it to better position their product offering.

As shown in this thread: http://www.nifty-stuff.com/forum/viewtopic.php?id=5324

It is amazing to see where certain colors are placed and even more surprising how they can vary and why. I certainly don't know the answer but in that example you can see how Magenta is used in so many colors that it makes you really think. Now add to that the aspect that the light magenta light Cyan as well as black in BCI-6 and CLI-8 are also different and the complexity explodes.

I am in total agreement with Grandad35 in his statement " the internal printer RIP must know about the color change or it will throw off all every color that uses magenta" . Because I got caught inside that exact problem. and that is why I devised the method to see where what color was being used and then shock ensued!

Grandad35's statement " Profiles are meant for relatively minor corrections" is also something I fully agree on as well since I stepped into that hole many times during my experiments or development. I had gotten formulations to look very good even fully profiled only to find that there was some oddities occurring in certain ranges and all of this had to do with the mismatch between Cyan and Light Cyan and what it produced. On certain images the color tones were correct until you started looking at gradients critically and you'd see non linearities and this is exactly what Grandad35 had run into when there were mismatches. Again, the image looked nice to a casual viewer but then after having looked at literally hundreds of the same images over and over again, you'd pick up very quickly where the problems were. In fact, at the end of the exercise, I had reviewed what I thought was acceptable just two months earlier and I qualified it now as totally not acceptable.

Like fine wine, you develop a taste and then all hell breaks loose!

The best solution in my mind is to use the formulation that tracks the OEM inks fairly closely. With this solution profiling even with a small number of patches yields near identical results as profiling with multiple pages of patches with little problems as the RIP and ink combination is close and only minor corrections are required. Of course, the ideal is to use OEM paper and ink but then it would not be as challenging and somebody has to support paper companies with all the wasted photo paper used in profiling.

 

[Grandad35]

Grandad, can you point us to the original post where these plots were lifted? I cannot find the results of the work you and canonfodder did years ago. Is that where you lifted these plots?

I'm afraid that getting my friends to use profiles, when I can't even get them to refill using the German method, is out of the question. But I don't blame them, even your simple explanation looks very complicated and I have no interest in attempting this myself.

This is the thread where the original data was posted. This is the link to the spreadsheet. Yes, the spreadsheet is where I got the plots.

Profiles may look scary, but they're not really all that hard once you get over the fear of tackling something new. You obviously have good color vision, but I'm partially color blind and had no choice except to learn color management and "do it by the numbers". By implementing a color managed workflow, shooting RAW, setting the camera's color balance in tricky lighting situations with with a gray card and profiling my printer, I am able to consistently get good color.

After messing with various inks and hundreds of combinations for a project I had what i can say is that messing around with ink colors is not an easy task and it is easy to make conclusions that are wrong very easily. Many many times I thought that I had finalized my formulations because the images I was using looked good until further testing revealed a hole! Back to the drawing board again. Many times, a revision in one aspect producing better results in one area resulted in obvious flaws in different images weeks later.... only to result in revisions again.

That's why most profile targets contain so many colors - for example here are the 918 colors used by my spectro.

It is amazing to see where certain colors are placed and even more surprising how they can vary and why. I certainly don't know the answer but in that example you can see how Magenta is used in so many colors that it makes you really think. Now add to that the aspect that the light magenta light Cyan as well as black in BCI-6 and CLI-8 are also different and the complexity explodes.

Exactly

The best solution in my mind is to use the formulation that tracks the OEM inks fairly closely. With this solution profiling even with a small number of patches yields near identical results as profiling with multiple pages of patches with little problems as the RIP and ink combination is close and only minor corrections are required. Of course, the ideal is to use OEM paper and ink but then it would not be as challenging and somebody has to support paper companies with all the wasted photo paper used in profiling.

Agreed, but there is no harm and very little time involved in using more target colors just to make sure that you have all of the bases covered.

In the end, it all boils down to how fussy you are about your colors. What is terrible for one person might well be acceptable for another.

 

[leo8088]

I emailed Hobbicolors a long time (2 years?) ago and asked if I can use UW8 ink on ip6000D BCI-6 printers. Or I should order their WC6 ink. They still had the WC6 ink at the time. I was told that UW8 ink will work on BCI-6 printers flawlessly with no worries. I would get a surprise that the colors would be more vibrant. I wouldn't have to buy two different set of inks. The colors would be more consistent between my older and new printers. I tried UW8 ink on my ip6000D. I never went back to WC6 ink ever since. I gave away my ip6000D to a friend's kid eventually. I had both inks about 2 years ago. UW8 ink is definitely a better ink. Hobbicolors probably convinced its customers to switch to the new ink. The WC6 ink was then discontinued. My latest printer is a Pro9000. I have had it for 6 months. The UW8 ink works just fine in my Pro9000. Pro9000 is a "COLOR" photo printer. I don't use it for black and white or gray scale printing.

 

[ghwellsjr]

You obviously have good color vision, but I'm partially color blind and had no choice except to learn color management and "do it by the numbers".

Actually, I am also color blind. My eyes are less sensitive to red, in fact, my two eyes have different sensitivities. Sometimes, I can see different shades of red that normal people cannot distinguish but more often than not, I cannot see the difference between two different colors. I have learned that in terms of the shades produced by the three colors on a monitor, if either the blue or green components exist, I have trouble telling the difference between two colors that have different intensities in the red component. For example, I cannot tell the difference between pure green and pure yellow produced by 100% green and 100% red. They both look yellow to me. But with the green reduced to 80%, it looks green to me instead of yellow. I also have trouble telling the difference between white (all three colors at 100%) and cyan (red at 0%, and blue and green at 100%). So for me, I am just not motivated when it comes to color. I just want it to work without any twidling. You and I have two completely opposite approaches to deal with our color blindness.