OEM Versus Hobbicolors ink in Canon Pixma: THE TESTS
BD
Recently had a rather discouraging experience with my first attempt at
using Hobbicolors ink.
Background:
I have a second-hand Canon Pixma iP5000, which _had_ a fresh set of OEM
ink.
I also have a *new* iP4000, which is still on its first set of OEM ink.
Prior to the 5000 running dry, I was finding that all print results
between the two models were extremely comparable. I've been using
Epson, Canon and Kingston Glossy Photo paper.
Of the three paper types, I find that Canon paper sucks, as I can see
track marks from the pinwheels in the printer on the surface of the
paper. Kingston is the current preference, because the output is good,
and it's CHEAP.
So:
Swapped in some Hobbicolors ink over the past few days. I noticed
IMMEDIATE and, I dare say, dramatic differences in the output. Colors
do not appear as saturated, and the absorption (?) looks coarser on
close examination.
I grabbed a test image, and printed a Color and a BW printout on both
printers. Settings are all the same (Glossy Photo paper, taken from the
Cassette, with high quality - borderless).
I've tried to make this as 'clean' a comparison as possible, ruling out
all extraneous variables. One thing I have _not_ done is a straight
swap, putting the Hobbicolors in the 4000.
For what it's worth, I just did one more print with Canon paper (all
out of the Epson), and the results are the same as with the Kirkland.
So it would seem that in the context of Hobbicolors ink, Canon /
Kirkland paper is not a significant variable. There does appear to be
slightly less banding in the printout, but as far as the color
representation and 'graininess', it's the same.
ALSO: prior to running this test, I did at least 8 full page prints on
the new ink. That should satisfy any 'flushing' requirements for the
changeover from OEM to aftermarket.
Anyway.
The results can be seen in this zip file (it's about 1.2MB):
http:members.shaw.ca/robertrd/printtests.zip
I also included the source file I printed from.
TEST 1: Color
Printer: Canon Pixma IP4000 (new)
Paper: Kirkland Glossy Photo Paper
Ink: OEM
TEST2: Color
Printer: Canon Pixma IP5000 (used)
Paper: Kirkland Glossy Photo Paper
Ink: Hobbicolors
TEST3: Greyscale
Printer: Canon Pixma IP4000 (new)
Paper: Kirkland Glossy Photo Paper
Ink: OEM
TEST4: Greyscale
Printer: Canon Pixma IP5000 (used)
Paper: Kirkland Glossy Photo Paper
Ink: Hobbicolors
Observations:
-Colors in Hobbicolors print are less saturated.
-Absorption looks grainier on close examination
-GREYS LOOK GREEN!!!!!! WTF?!
I'd appreciate any comments.
BD
BD
art
>Hey, all.
>
>Recently had a rather discouraging experience with my first attempt at
>using Hobbicolors ink.
>
Your experience is not much different from mine. I have not tried
hobbicolors ink. I tried Inktec and got color match worse than your
hobbicolors experiment. G&G ink comes close to matching, but the ink
cartridges leaked in the printer and alotofthings.com never answered a
query to their support. MIS inks also come close to matching the
Canon ink. Your greenish grays are from the magenta not being
matched. For my test prints I use 25, 50, 75, and 100% saturated
cyan, magenta, yellow, and black. The gray test strips show the
mismatch most easily because at less than 80% saturation gray is
printed using only color inks. The other factor to consider is the
longevity of the ink. I did accelerated UV testing that shows that
the above mentioned inks fade up to 25 times faster than the Canon
Chroma Life 100. I have read similar fading problems with hobbicolor
inks. If you are doing throw away type printing the MIS ink is not
pretty good. If you don't want to reprint your photos on a regular
basis stick with Canon. If you are interested I can post my fading
tests.
frank
Hobbicolor. In fact, the Hobbicolor results were exactly what I saw on
the screen, not so with the Canon oem. So I'm much more pleased with the
truer results of Hobbicolor than with the Canon oem.
Frank
Yianni
The problem is definately the magenta. Colors are far away.
Also, if you notice, there is banding in the magenta with the hobbicolors
inks. I don't think banding comes from clogged nozzles, most possible the
magenta ink don't have suitable chemical/physical properties for this
printhead/cartridge combination. You can blame magenta.
Also the oem inks print with a remarkable magenta cast (or it's the
scanner?). If the oem printed well, the hobbicolors ink would be more far
away...
--
Yianni
i...@mailbox9.gr (remove number nine to send me email)
Yianni
see the same results as oem (at least for color accuracy).
Yianni
any charge, 40ml each color, anywere in the world you are. It's from a
totally different ink manufacturer. If yes, please reply directly to my
email (removing the number 9 after the mailbox).
--
Yianni
i...@mailbox9.gr (remove number nine to send me email)
"art" <nojun...@msn.com> wrote in message
news:dd5qc2lvft3rsn0bj...@4ax.com...
zakezuke
BD wrote:
>
> I'd appreciate any comments.
If you could link to the test images you used I think I might have an
idea.
But at present I can say with absolute certainty that the ip3000 and
mis ink did much better than this for greyscale. Unlike the ip4000 it
doesn't have a dye black and a mix of the three colors = black.
BD
zakezuke wrote:
> BD wrote:
> >
> > I'd appreciate any comments.
>
> If you could link to the test images you used I think I might have an
> idea.
I'd posted a link to the images:
http://members.shaw.ca/robertrd/printtests.zip
The source image is there, as are scans of several tests.
Yianni
differences to cyan too.
--
Yianni
BD
Yianni wrote:
> A mistake. After examining the two scans more carefully, it seems there are
> differences to cyan too.
It's all very frustrating.
Is my situation 'unique' here? Or am I just demanding more from these
inks than most people...? Seems to me that 'grey' should not come out
looking 'green...' I hear so much good press about Hobbicolors, and yet
on the first try I'm - really disappointed. ;-(
frank
for 13 x 19 borderless I use only Canon Photo Paper Pro.
The results will probably be different than those produced by Kirkland
paper.
I am very satisfied with my printing results using Hobbicolor refill ink
in my Canon i9900 printer. I have not purchased an oem cart for years
and don't even intend to do so.
Frank
BD
> paper.
Perhaps. But not different enough to compensate for what I'm seeing,
I'll warrant.
I do want to make aftermarket ink work here, so I may try the 'Pro'
Canon paper. But at this point I'm rather discouraged.
frank
Well, when I first got the i9900 (about 3 yrs ago) I was having a lot of
problems with color matching what I saw on my Sony GDM-F520 monitor and
what was being printed out. It didn't match. I tried many different ICC
to correct the problem and I tried other after market ink and was more
satisfied, but not entirely. Finally I settled on Hobbicolor as it was
giving me the very best match of screen to print.
Many who own i9900's have complained about the screen to print mis-match.
I found my solution.
Frank
BD
I still need to educate myself somewhat on ICC profiles. Oh well...
Tony
I cannot add much to this except to relate a recent experience.
We have a supplier of ink and paper, both aftermarket, excellent quality inks
(I don't use this forum to sell anything so will not disclose the
manufacturer). They formulate and make their own inks, cartridges and photo
paper and export to many countries. We recently had a problem with their ink on
their photo paper both formulated for a series of Canon printers. The result
was absolutely awful, on asking their advice they responded quickly (as they
always do). Apparently their paper is formulated for Canon OEM ink and is not a
match for their own ink, this is a deliberate marketing decision and has to do
with the thickness of the coating on the photo paper. Their inks on Canon paper
are great, their paper with Canon inks is great, their paper with their inks is
just dreadful. Interesting how complex the technology can be isn't it?
Tony
BD
Quite. ;-)
**One thing that I haven't really thought about here is ICC profiles. I
am still kind of in Primary School as regards such stuff. I wonder if
looking at making a new color profile for the printer when using the
aftermarket ink would be of any value.
Burt
"BD" <rober...@gmail.com> wrote in message
news:1154361660.4...@m79g2000cwm.googlegroups.com...
Grandad35, a participant on the Nifty-stuff forum, has discussed custom
profiles and has one for his i9900 Canon printer with Formulabs inks from
Alotofthings. You would need to calibrate your monitor accurately as well.
BD
> Grandad35, a participant on the Nifty-stuff forum, has discussed custom
> profiles and has one for his i9900 Canon printer with Formulabs inks from
> Alotofthings. You would need to calibrate your monitor accurately as well.
Yes, I'd been considering buying one of those Eye One devices or
something in that ilk.
What kind of gets me (and I concede I don't know the first thing about
how ink is formulated) is that... OEM ink is not a moving target. What
was BCI-6 cyan two years ago is BCI-6 cyan today. We're talking about
creating three different colors, and a black. So I am forced to wonder
what it is that makes it such a challenge to accurately emulate OEM
color gamuts. You'd think that trial and error over time could allow
ink formulators to 'dial in' each color over time, to near perfect
accuracy.
But I guess I'm looking at it from the same perspective as
color-matching paint... which, it would seem, someone at my local
hardware store can do pretty well, without a whole huge amount of
experience or education - just an 'eye' for the effects that a change
in a primary color can have on the final color balance.
So - asking this not as a facetious jab, but a genuine lack of
understanding of what's involved - what is it that makes it so hard to
balance or emulate three little colors?
BD
frank
how come when I first got my i9900, using the supplied canon oem ink and
canon paper, I could not get an accurate, color matched printout of what
I was seeing on the screen?
Frank
BD
> how come when I first got my i9900, using the supplied canon oem ink and
> canon paper, I could not get an accurate, color matched printout of what
> I was seeing on the screen?
Weeeellll... I guess the message there is that printer, ink and paper
aren't all the variables in the equation. And I have no disagreement
with that.
I'd be surprised if what I printed looked exactly like my screen did as
well - _but_ what I expected was, all other things being equal, the
hobbicolors print looking different from my screen in a similar sense
as the OEM print.
I recently spotted a thread on nifty-stuff about a green cast from
Hobbicolors ink. I think I'll post in there and see what comes of it.
art
>What kind of gets me (and I concede I don't know the first thing about
>how ink is formulated) is that... OEM ink is not a moving target. What
>was BCI-6 cyan two years ago is BCI-6 cyan today. We're talking about
>creating three different colors, and a black. So I am forced to wonder
>what it is that makes it such a challenge to accurately emulate OEM
>color gamuts. You'd think that trial and error over time could allow
>ink formulators to 'dial in' each color over time, to near perfect
>accuracy.
>
>But I guess I'm looking at it from the same perspective as
>color-matching paint... which, it would seem, someone at my local
>hardware store can do pretty well, without a whole huge amount of
>experience or education - just an 'eye' for the effects that a change
>in a primary color can have on the final color balance.
>
>So - asking this not as a facetious jab, but a genuine lack of
>understanding of what's involved - what is it that makes it so hard to
>balance or emulate three little colors?
>
Matching the OEM ink in terms of color is probably not that difficult,
The hard part is to match color at a cheap price. After having
matched color the next problem is to match longevity. If someone had
a formula that matched color and longevity it is doubtful if it would
be price competitive with the OEM. The aftermarket ink is fine when
doing throw away type printing.
>BD
BD
> The aftermarket ink is fine when
> doing throw away type printing.
And I guess this is where I was misinterpreting the reports. I imagine
that it also depends on _what_ is being printed: for example, colored
documents or Powerpoint slides may be completely satisfactory if the
colors are slightly off from 'baseline'.
Still. I'm not altogether convinced that something is not 'wrong': I
spotted a thread on nifty-stuff concerning a 'green cast' in a
printout. I'll look over there.
Burt
"frank" <f...@nospamm.cmm> wrote in message
news:Hs2dnebZRpu74lPZ...@adelphia.com...
Frank - When I bought my i960 Canon printer I used the OEM carts until they
needed refilling. I then ran the same photo files with MIS inks. Both
tended to lean toward red/magenta and the OEM prints were somewhat
oversaturated. Although I own and enjoy using three Canon printers I have
read that their color profiles are not the best. What was the tendency with
the i9900? Also toward the red tones?
frank
Of course after the first few disappointing prints, I dl'ed i9900
specific icc's and tried printing in every graphic program I have (and
that's a lot of programs). I spent days trying to get a reasonably close
match from screen to print. Then after the oem ink started to run out I
went to after market suppliers. Finally settled on Hobbicolor which
works just fine for me.
Frank
zakezuke
I have yet to test my theory, but based on what everyone is saying, and
based on what you are saying, it SOUNDS like you got the wc8 version of
the ink rather than the the bci-6 compatable It's red in my eyes is
shifted slightly tward the yellow. There was another company who got
what was termed at the time "asian market" red and resold it as bci-6
compatable. I would need to see a scanned swab of both OEM and what
you got to really be sure. You may also wish to swap out just magenta
and peform the same tests.
I lack direct experence to be sure but your experence is inconsistant
with what i've read here.
BD
> you got to really be sure. You may also wish to swap out just magenta
> and peform the same tests.
I did swap the magentas back, and ran through about 3 full-page prints.
No real difference was evident, but the three pages may not have been
sufficient to transition the ink in the head. Not sure.
As to swabs - what exactly are we talking about here? Taking a cotton
swab and running a strip along a piece of paper, and then scanning that
for all to see? I can do that no problem, if you think it would be of
value...
BD
BD
> you got to really be sure. You may also wish to swap out just magenta
> and peform the same tests.
Here is a scanned swab of all 5 carts.
Differences are not *gross*, but are noticeable:
Yellow is noticeably lighter.
Magenta looks more red than 'pink',
Cyan is less rich.
I used the Kirkland glossy paper for the scan. Epson Perfection 1250
Pro scanner, Epson TWAIN 5.71a for the capture...
http://members.shaw.ca/robertrd/inkswab.jpg
I dunno.
I guess the fact that the raw swabs are noticeably different ain't
great...
Other opinions?
BD
zakezuke
BD wrote:
> > compatable. I would need to see a scanned swab of both OEM and what
> > you got to really be sure. You may also wish to swap out just magenta
> > and peform the same tests.
>
> Here is a scanned swab of all 5 carts.
>
> Differences are not *gross*, but are noticeable:
>
> Yellow is noticeably lighter.
> Magenta looks more red than 'pink',
> Cyan is less rich.
>
> I used the Kirkland glossy paper for the scan. Epson Perfection 1250
> Pro scanner, Epson TWAIN 5.71a for the capture...
>
> http://members.shaw.ca/robertrd/inkswab.jpg
>
> I dunno.
I'm seeing much in the way of green. Are you sure you're not using
kirkland's 4x6 paper? Well I just pulled a reference from
http://forums.dpreview.com/forums/read.asp?forum=1003&message=12198835
Lots of green there too.
Your OEM is about r190 and b87 (decimal) Discounting the extra green.
Your HOBI is about r171 and b76
My MIS is r160 b59. My Hobbi wc8 is r177 b15
http://i43.photobucket.com/albums/e377/urusei-yatsura/hobiVSmis.jpg
http://i43.photobucket.com/albums/e377/urusei-yatsura/hobiVSmis2.jpg
The fact that we are using different scaners and the fact that I don't
have any OEM handy to test does add some shadow of doubt, but this
should be enough information to contact hobbicolors and at least ask
about the possiblity of being shipped wc8 ink rather than bci-6
compatable. Looks to me like it's shifted away from the blue.
This is not conclusive but I feel a reasonable hypothesis.
> I guess the fact that the raw swabs are noticeably different ain't
> great...
>
> Other opinions?
http://forums.dpreview.com/forums/read.asp?forum=1003&message=12198835
frank
photograph) printed with both inks on the same paper using the same
settings?
Post it if you have any. Swabs and color strips can be misleading were
as photos seem to be much more telling.
THX
Frank
BD
> Post it if you have any. Swabs and color strips can be misleading were
> as photos seem to be much more telling.
> THX
> Frank
Indeed I do. I'll post this evening. This is the printout that got me
worked up in the first place. I had intended to post the photo samples,
but elected to do the generic test page instead.
It's a 'leaf-green' tone that is thrown off - almost to the point where
it looks brownish. I tried to correct it in Photoshop, but have had no
luck.
Yianni
You can't judge from the "button test" the colors. Because in other places
it's darker in other lighter. I do this test only for comparing colors with
eyes, and using about 10 tracks of ink and comparing this that is just next
to the first saturated ones.
--
Yianni
i...@mailbox9.gr (remove number nine to send me email)
"zakezuke" <zakez...@yahoo.com> wrote in message
news:1154409019....@b28g2000cwb.googlegroups.com...
BD
> Question: do you have any samples of the same photo (a real digital
> photograph) printed with both inks on the same paper using the same
> settings?
> Post it if you have any. Swabs and color strips can be misleading were
> as photos seem to be much more telling.
Well, here's the best I could do.
The scans don't give a really representative image, but the delta
between oem and hc is fairly consistent with what I'm seeing. HC's
greens are considerably lighter, and closer to 'brown' than green.
The scans seem to take away some of the saturation. The print with oem
ink is pretty decent, actually.
http://members.shaw.ca/robertrd/green_original.jpg
http://members.shaw.ca/robertrd/green_oem.jpg
http://members.shaw.ca/robertrd/green_hc.jpg
Very frustrating...
BD
zakezuke
-Original
>http://img97.imageshack.us/my.php?image=chitzenitza14yz.jpg
-MIS bci-6 compatable HP Glossy paper pro
>http://img97.imageshack.us/my.php?image=chitzenitza1bc600015fc.jpg
-Hobbicolors wc8 on a bci-6 printer HP glossy paper pro
>http://img215.imageshack.us/my.php?image=chitzenitza1bci700013nw.jpg
Scanner mp760
Printed using EZ photo print.
Given your swabs, the colorshift increased scans showing increased red
and green... while I have never met hobbicolors WC-6 ink, your results
look exactly what i'd expect from putting hobbicolors wc8 ink in a
bci-6 printer. If i'm right it may be possible to set your printer to
japanese mode and print using the ink you have and have it be
reasonably decent, and heck on par with a ip4200.
But feel free to print off my original and see for your self.
BD
> bci-6 printer. If i'm right it may be possible to set your printer to
> japanese mode and print using the ink you have and have it be
I'll research this, as I believe that part of the process whereby I
enabled CD printing on my 5000 also put it in 'Japanese' mode...
But am I hearing that this mode also alters the printer's color gamut
slightly? I may not be understanding you correctly...
BD
zakezuke
I believe the ink you are using is wc8 ink, not hobbicolors wc6 ink.
WC8 should be spot on with bci-7 which is what the Japanese pixmas
used, rather than cli-8 which is for the newer world market pixmas.
I could be wrong, I am evaluating something at great distance from me,
but everything you say fits my hypothesis.
The problem is I lack accurate data on whether one can with the
American printer use the American driver, and print using the Japanese
ink. I don't have a desktop pixma, only an AIO, which I "know" for a
fact requires the Japanese driver. But based on everything i've read,
the desktop pixmas don't behave in the same way.
BD
> WC8 should be spot on with bci-7 which is what the Japanese pixmas
> used, rather than cli-8 which is for the newer world market pixmas.
Mmm.
I may strike up a conversation with the Hobbicolors folks and discuss
it with them - I also may try to get one more kit, to make sure.
I guess it's also not a bad idea to try another formulation -
Formulabs, MIS, etc.
Another thing that might be worthwhile, is - if someone is willing to
indulge me - print a copy of my test image on their printer, using
*definitive* wc6 ink, and then mail it to me - COD or something - and
based on that I would have a comparison. If I ended up with 2 printouts
of the same image that looked different in spite of having been printed
with the same ink *presumably*, then I could contact the vendor and ask
them to be *sure* I had gotten the right stuff.
But meanwhile, I think I'll contact my Ebay seller, and strike up a
dialog about it.
measekite
BD wrote:
>Yianni wrote:
>
>
>>A mistake. After examining the two scans more carefully, it seems there are
>>differences to cyan too.
>>
>>
>
>It's all very frustrating.
>
>Is my situation 'unique' here? Or am I just demanding more from these
>inks than most people...? Seems to me that 'grey' should not come out
>looking 'green...' I hear so much good press about Hobbicolors, and yet
>on the first try I'm - really disappointed. ;-(
>
>
you do not know if the relabeler changed suppliers cause they will not
tell you nor will they tell you what you got cause they are afraid you
will go to their mfg/formulator and buy direct. That is what they told
me in a phone call. Unless you are a very high volume print user I
cannot understand why you want all of this grief. You have proven that
Canon provides the best results so maybe you should stick with it.