Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

new Tom's Hardware printer test: Canon vs HP

2 views
Skip to first unread message

Panos Stokas

unread,
Dec 10, 2005, 3:12:04 AM12/10/05
to
I've read this new THG review yesterday and I must say that it is the
worst THG review I've ever seen. For a website that rose out through
overclocking articles, I find it quite hypocritical for them to ignore
inkjet refilling.

Secondly: did THG test the printers for paper handling reliability? I
have printed at least 10 thousand pages on my Canon MP750 so far and I
didn't have a single misfeed, ever. I've printed out books using the
manual duplex feature without fear of a page misfeed that would ruin
the entire book. As an owner of 4 Hewlett Packard printers in the past,
I had experienced an average 1:200 misfeed ratio. I had successfully
dropped the ratio to 1:350 using my "grease on the rollers" method, but
the fact is that most efforts to duplex print resulted in frustration
due to page misfeeds.

Thirdly: is there any reference (considering that we are talking about
a hardware enthusiast and technically experienced site here) to the
*terrible* device drivers HP produces? Their javascript user interface,
their 750 MB installation requirements, the programs they run on
startup eating resources, and the fewer options?

Fourthly: THG dedicates 3 pages on quality tests, but they fail to
provide us with the original photo. My guess is: how could anyone say
which printer is best without being able to compare with the original?

Fifthly: THG compares printers in draft mode. That's laughable. Draft
mode in inkjet printers is just an advertisement trick. Considering the
quality degradation, if people would print on draft mode, people would
much better use a laser printer and enjoy much higher quality on the
same speeds without the inkjet noise.

Instead THG is focusing on features like LCD screens and card readers.
Just what the average THG reader needs...

Here is a simple comparison from my own experience:

* Canon has more reliable paper feed
* Canon offers two different ways to feed paper
* Canon has higher quality photos.
* HP has higher quality text.
* Canon is much faster on business documents where speed is important.
* Canon is much more quiet. Very important on business environments.
* Canon printers from Europe can print on CDs, you can modify them to
do so in the US. Very important for the home user.
* Canon has better drivers. I just can't stress how important that is
for the overall system health.
* Canon printers can be refilled and that saves you 95% of the ink
costs.

Canon printers are better, both for the business user and the home
user. Period.

zakezuke

unread,
Dec 10, 2005, 4:08:07 AM12/10/05
to
> I find it quite hypocritical for them to
> ignore inkjet refilling.

Why? When they test out motherboards they do their best to test in
both stock mode and overclocked mode. While I think it would be nice
of them to include some aftermarket ink tests... the truth of the
matter is this would increase the level of testing by a factor of at
least two to test one other option... and there are at least 5 major
options available assuming they are even aware of them.

They are being fair testing the products out of the box with the ink
out of the box.

> Thirdly: is there any reference
> (considering that we are talking about
> a hardware enthusiast and technically
> experienced site here) to the
> *terrible* device drivers HP produces?

Techncialy these are produced by Microsoft IIRC. And actually the HP
drivers tend to be pretty decent... it's the auto install routine that
i've noticed being screwy on some models.

> Fourthly: THG dedicates 3 pages on
> quality tests, but they fail to
> provide us with the original photo.

wow, good point.
http://www.tomshardware.com/2004/12/29/are_high/page6.html
You can find the photo here of the girl on the bike. Normally they are
good about this but I guess they forgot. Tell them about it, it's a
valid complaint, i'm sure they'll correct it.

> Fifthly: THG compares printers in draft > mode. That's laughable. Draft
> mode in inkjet printers is just an
> advertisement trick.

I use draft mode... totally. It does use less ink. The contrast isn't
as cool as regular or HQ mode but I do indeed do draft mode.


1. Panos Stokas
Dec 10, 12:12 am show options
Newsgroups: comp.periphs.printers
From: "Panos Stokas" <Panos.Sto...@gmail.com> - Find messages by this
author
Date: 10 Dec 2005 00:12:04 -0800
Local: Sat, Dec 10 2005 12:12 am
Subject: new Tom's Hardware printer test: Canon vs HP
Reply | Reply to Author | Forward | Print | Individual Message | Show
original | Report Abuse

I haven't had too many problems with mine... HP's paper feed is pretty
good from my experence.

> Canon offers two different ways to feed paper

Nice feature, not useful to some but a nice feature.

> Canon has higher quality photos.

That's opinion and that can be seen in their quality tests... ok not so
much so without the origional photograph. Actually from their last
review I would have picked the epson r800 as the best of the lot.
Canon is pretty good though.

> HP has higher quality text.

This could be... I actually prefer the canon for text, esp when doing 5
point kanji. If you can make out kanji at 5 point the printer is good.

> * Canon is much faster on business
> documents where speed is important.

What is this business document test? Tell us more about it.

> * Canon is much more quiet. Very
> important on business environments.

That's a good point, toms should include noise level.

> Canon printers from Europe can print
> on CDs, you can modify them to
> do so in the US. Very important for the > home user.

They do speak about CD printing actually. Your point on this is moot.
It's not important to all users but it is handy.

> * Canon has better drivers. I just can't
> stress how important that is
> for the overall system health.

Ummmm.... Actuallly I find canon's drivers to be kinda basic and not
all that spiffy. I mean at least with Epson you get a ton of options
including that fab microweave mode, HP pretty much the same deal. The
canon drivers have few options and the interface looks like it was
writen as a 16bit application. Further, HP drivers are at least writen
by microsoft for microsoft... and no one knows microsoft like
microsoft.

> Canon printers can be refilled and that
> saves you 95% of the ink costs.

When you take all factors into account, it's more like 80% to 90%.
Either you need refillable cartridges that cost $3 to $5 each or you
need canon oems to refill costing $10 to $15 each. Don't get me wrong
the figure is high, but you gotta take other things into account and
not just raw ink cost.

Panos Stokas

unread,
Dec 10, 2005, 5:33:18 AM12/10/05
to
> When they test out motherboards they do their best to
> test in both stock mode and overclocked mode

My point exactly. They do test overclock ability of motherboards but
they do not do the same for refill ability on inkjets.

> Techncialy these are produced by Microsoft IIRC

No. Microsoft is not to blame on this one. It's not the driver model
I am having a problem with it's...

> And actually the HP drivers tend to be pretty decent...

... the fact that they require you to install the entire software and
parts of their drivers are accessible only through a DHTML interface.
Also the fact that when you try to share HP drivers on a LAN you'll
have to wait until the HP driver has to spool the entire print job on
the server before it starts printing out pages.

Did you try to install one of the latest models? HP drivers are not the
same to what they have been a few years ago.

Do you really justify 400 MB for a printer device driver?

> > they fail to provide us with the original photo.

> http://www.tomshardware.com/2004/12/29/are_high/page6.html

Did you notice that these two images are 100% identical?

http://images.tomshardware.com/2004/12/29/are_high/ip8500_vive_detail_big.jpg

http://images.tomshardware.com/2004/12/29/are_high/8450_vive_detail_big.jpg

Another "mistake" I guess. Come to think, I used to like Tom's
Hardware.

> I use draft mode... totally. It does use less ink.

In a quality that simply doesn't justify inkjet technology with all
it's fallacies.

I would not disagree with your preferences, I just don't think draft
mode should be a factor in an inkjet test. But I understand this is
just my opinion.

> I haven't had too many problems with mine [paper feed]

Too many is just not good enough for serious use.

> Actually from their last review

... no serious conclusions could be drawn considering the problem
I've specified earlier.

> > * Canon is much faster on business
> > documents where speed is important.

> What is this business document test? Tell us more about it.

As I've said this is my personal experience. But I have printed the
same documents (I print plenty of business documents) on those printers
and the Canon prints in normal quality is roughly equal to HP at the
everyday (below normal but not draft) quality.

I doubt that you would find one test on the web that wouldn't give
Canon credits for speed in text & graphics documents.

> They do speak about CD printing actually.
> Your point on this is moot.

You mean Canon, right? HPs don't do direct CD printing.

> The canon drivers have few options and the interface
> looks like it was writen as a 16bit application.

What do you mean "few" options? They provide more options for page
setup than HP, which is exactly what a driver is meant for.

"looks like" ? Is this a technical debate or about aesthetics ?

> Further, HP drivers are at least writen
> by microsoft for microsoft...

What are you talking about? Where did you find the information that
HP's device drivers are written by Microsoft?

And how does that contribute to a point about driver software quality?

Knightcrawler

unread,
Dec 10, 2005, 6:48:24 AM12/10/05
to

Knightcrawler

unread,
Dec 10, 2005, 6:54:06 AM12/10/05
to
Lots of problems with the review:
http://www.tomshardware.com/2005/12/08/canon_hp_xmas_ink_jet_printer_fleet/

1. Canon has a model IP6600 which has:

1. LCD screen
2. Card Slots
3. 3000 nozzles
4. photo magenta and photo cyan
5. cd/dvd printing
6. 1 picoliter droplet

yet we are subjected to a review of the 4200/5200 models which the reviewer
states: "the Japanese vendor remained mired in its usual conservatism".

2. Ink Capacity

HP 8250

6 HP ink cartridges:
HP 02 Black, 16 ml,
HP 02 Cyan, 4.5 ml,
HP 02 Magenta, 4.5 ml,
HP 02 Yellow, 4.5 ml,
HP 02 Light Cyan, 4.5 ml,
HP 02 Light Magenta, 4.5ml

Canon IP4200/5200/6600:

13 ml per cart except pigment black for the 4200/5200 which is 26 ml. Why
don't reviews state how much ink we will be getting and how much those carts
will cost ?

Canon's have 3 times more ink and yet the carts cost about the same as HP
!!!

3. Third party inks

One way to dramtically reduce cost is to use third party ink. The cost
difference can be 20 times lower. I use Formulabs in all my Canons and the
quality is the same I was getting from Canon OEM ink. There is no problem
with color fading as the OEM's will always try to claim. I have pictures
that I did a year ago that are as bright and crisp as the day I printed
them. Proper storage and display is recommended for any picture you take.

The ink I purchased cost me $34 versus Canon OEM ink which would run me $600
for the same amount.

It's no secret that ink is where they make their money. Producing ink can be
as low as $4 per liter and yet they sell it for $1000 per liter or more. HP
makes it's lion share of profit on ink sales alone.


http://forumz.tomshardware.com/hardware/ftopicp-914073.html#914073

Sign up at Toms and add your replies their so he gets them.


Martin

unread,
Dec 10, 2005, 8:56:15 AM12/10/05
to
zakezuke wrote:

>>Fifthly: THG compares printers in draft > mode. That's laughable. Draft
>>mode in inkjet printers is just an
>>advertisement trick.
>
>
> I use draft mode... totally. It does use less ink. The contrast isn't
> as cool as regular or HQ mode but I do indeed do draft mode.

I can't really comment on the whole thing but I do agree that draft mode
on different printers is not a good benchmark.

Just looking at a C84 draft mode vs' a Canon MP750 and you can see a
MASSIVE difference in the ink used and the quality of the print. Why
they don't use "Normal" quality output is beyond me... That, after all,
is the benchmark most users will likely be using for regular inkjet
printing.

Anticipating a query re: photo printers, you'd also be sure to level the
benchmark at the "highest quality" level or at least test at that level
for additional comparison.

Arthur Entlich

unread,
Dec 10, 2005, 9:01:58 AM12/10/05
to
Having done some reviews for publications, in their defense, let me just
say that although it could be otherwise there, usually, reviews are
bought from an individual reviewer and some reviewers have different
priorities and see corresponding features and functionality differently.

I find it is best to look for a by-line when reading a review and if you
find that person tends to weight features differently than you do, to,
as you would with a movie reviewer, seek out a reviewer that has more
similar leanings to your own.

It may also make sense to writ the editor and express your concerns, as
it may alter who they hire to do reviews.

Art

zakezuke

unread,
Dec 10, 2005, 9:03:22 AM12/10/05
to
>> When they test out motherboards they do their best to
>> test in both stock mode and overclocked mode

>My point exactly. They do test overclock ability of motherboards but
>they do not do the same for refill ability on inkjets.

The thing is most reviewers don't take this into account. I agree it
would be nice but painfully hard esp since there are many choices for
ink, and many printers from which to test. And this is even assuming
the reviewer is familar with these options.

> Did you notice that these two images are 100% identical?

Nope, I can't get them to display... perhaps it's direct linking
tomshardware doesn't like.

> > I haven't had too many problems with mine [paper feed]
> Too many is just not good enough for serious use.

I have NEVER had an issue with letter sized paper in my psc 950. 4x6 I
can't say that with. Now my canon ip3000 on the other hand is pretty
good, but my prints are almost always slightly askew for some reason.
Perhaps I can do something to correct that but using my chopper with my
HP it's always spot on square.

> > They do speak about CD printing actually.
> > Your point on this is moot.
> You mean Canon, right? HPs don't do direct CD printing.

I read the same review you did, they spoke about CD printing on the
ip4200... so you can't say they didn't bring it up. You could say they
should bring up the fact that you can enable the feature... but keep
also in mind that that would be a seperate hack and the hack is pretty
dang young. Granted I joined the steve's forums thread a late
march/april... and the ip4000 was already kinda figured out but I can
take credit for some of the work done hacking the mp750/760/780. My
project was mostly compiling existing info I started in late may from
which knightcrawler sifted through my disorganized notes and made a
really spiffy site with easy to follow instruction blocks. I can't say
I wrote the book on the subject.... but my contribution was noteworthy
and I know a little bit about the subject. But you can't assume tom's
hardware knows about it... in fact you should tell them, they should
publish the work of my self and knightcrawler.

> What do you mean "few" options? They provide more options for page
> setup than HP, which is exactly what a driver is meant for.
>
> "looks like" ? Is this a technical debate or about aesthetics ?

I'm getting some odd ball issues with the canon drivers dealing with
Unicode, specificly Japanese... which is something that you wouldn't
expect from a Japanese company. They do offer a Japanese driver but
it's sorta full blown Fapanese... with no choice to use japanese but be
english... or a choice to meet it half way and display in romaji, at
least not from what i've seen.
It's hard to explain perfectly but some of the software from canon
results in memory leaks, the likes of which you only experence running
16bit applications in 2k/xp.

> What are you talking about? Where did you find the information that
> HP's device drivers are written by Microsoft?

There was an article a while back about a company who contacted HP
regarding adding some propriority features... navigating up the chain
only to discover that HP doesn't actually write the drivers and they
had to talk to microsoft... and contracted microsoft to do the work and
their custom drivers came with full HP logos.

> And how does that contribute to a point about driver software quality?

Well, one big issue I have with the canon drivers is a lack of real
hard core control over the use of the two blacks. It's pretty much a
use pigment on plain paper and dye on photo paper.... or oddly enough I
caught it using dye black on plain paper when I mixed in colored text
with black text... ran out of cyan and the black text was well, not
black anymore. On HP drivers at the very least it's at least object
oriented... not document oriented. As in text gets treated like text
and graphics get treated like graphics.

>> And actually the HP drivers tend to be pretty decent...

.> .. the fact that they require you to install the entire software and


> parts of their drivers are accessible only through a DHTML interface.
> Also the fact that when you try to share HP drivers on a LAN you'll
> have to wait until the HP driver has to spool the entire print job on
> the server before it starts printing out pages.

Ummm... i'll agree that the full install disc with all the extra crap
is rather bogus... and causes some serious tweeky weirdo stuff if it
screws up. Certainly a software issue coupled by a media issue... some
of those discs I got were bloody unreadable on any of my drives... and
I have some spiffy CD rom drives that read just about anything. But
you can't say you are required to install 500megs of crap... you can
just install the driver. It's not a problem. Not sure on the
spooling issue... I can't say I've noticed it.

Arthur Entlich

unread,
Dec 10, 2005, 9:41:25 AM12/10/05
to

Knightcrawler wrote:

>
>
> 2. Ink Capacity
>
> HP 8250
>
> 6 HP ink cartridges:
> HP 02 Black, 16 ml,
> HP 02 Cyan, 4.5 ml,
> HP 02 Magenta, 4.5 ml,
> HP 02 Yellow, 4.5 ml,
> HP 02 Light Cyan, 4.5 ml,
> HP 02 Light Magenta, 4.5ml
>
> Canon IP4200/5200/6600:
>
> 13 ml per cart except pigment black for the 4200/5200 which is 26 ml. Why
> don't reviews state how much ink we will be getting and how much those carts
> will cost ?
>
> Canon's have 3 times more ink and yet the carts cost about the same as HP
> !!!
>


In fact, there is a much more important issue. HP has introduced a
number of new features to their printers which restricts the amount of
ink wasted during cleaning cycles, etc. A millimeter of ink isn't
necessarily of the same value to another printer. Maybe the Canon uses
up 3 times as much ink as the HP for the same yield of documents


Art

measekite

unread,
Dec 10, 2005, 12:53:56 PM12/10/05
to

Knightcrawler wrote:

THEY WANT TO KNOW WHAT THEY ARE REVIEWING AND THE VENDOR LABELERS WILL
NOT TELL THEM.

>
>
>

measekite

unread,
Dec 10, 2005, 12:54:54 PM12/10/05
to

Panos Stokas wrote:

>>When they test out motherboards they do their best to
>>test in both stock mode and overclocked mode
>>
>>
>
>My point exactly. They do test overclock ability of motherboards but
>they do not do the same for refill ability on inkjets.
>
>

THEY KNOW WHO MAKES THE MOTHERBOARDS. THEY KNOW WHO MAKES THE CHIPS ON
THEM. THEY KNOW WHO MAKES THE CPU

measekite

unread,
Dec 10, 2005, 12:59:20 PM12/10/05
to

Panos Stokas wrote:

>I've read this new THG review yesterday and I must say that it is the
>worst THG review I've ever seen. For a website that rose out through
>overclocking articles, I find it quite hypocritical for them to ignore
>inkjet refilling.
>
>Secondly: did THG test the printers for paper handling reliability? I
>have printed at least 10 thousand pages on my Canon MP750 so far and I
>didn't have a single misfeed, ever. I've printed out books using the
>manual duplex feature
>

THEY HAVE AN AUTO DUPLEX FEATURE NOT A MANUAL DUPLEX FEATURE.

>without fear of a page misfeed that would ruin
>the entire book. As an owner of 4 Hewlett Packard printers in the past,
>I had experienced an average 1:200 misfeed ratio.
>

THAT IS NOT TRUE FOR ME

>I had successfully
>dropped the ratio to 1:350 using my "grease on the rollers" method, but
>the fact is that most efforts to duplex print resulted in frustration
>due to page misfeeds.
>
>Thirdly: is there any reference (considering that we are talking about
>a hardware enthusiast and technically experienced site here) to the
>*terrible* device drivers HP produces? Their javascript user interface,
>their 750 MB installation requirements, the programs they run on
>startup eating resources, and the fewer options?
>
>Fourthly: THG dedicates 3 pages on quality tests, but they fail to
>provide us with the original photo. My guess is: how could anyone say
>which printer is best without being able to compare with the original?
>
>Fifthly: THG compares printers in draft mode. That's laughable. Draft
>mode in inkjet printers is just an advertisement trick.
>

USE IT ALL OF THE TIME FOR BUSINESS DOCS

>Considering the
>quality degradation, if people would print on draft mode, people would
>much better use a laser printer and enjoy much higher quality on the
>same speeds without the inkjet noise.
>
>Instead THG is focusing on features like LCD screens and card readers.
>Just what the average THG reader needs...
>
>Here is a simple comparison from my own experience:
>
>* Canon has more reliable paper feed
>
>

ABOUT THE SAME

>* Canon offers two different ways to feed paper
>* Canon has higher quality photos.
>
>

YES

>* HP has higher quality text.
>
>

YES

>* Canon is much faster on business documents where speed is important.
>
>

YES

>* Canon is much more quiet. Very important on business environments.
>
>

SAME

>* Canon printers from Europe can print on CDs, you can modify them to
>do so in the US. Very important for the home user.
>* Canon has better drivers.
>

MAYBE

>I just can't stress how important that is
>for the overall system health.
>* Canon printers can be refilled and that saves you 95% of the ink
>costs.
>
>

NOT A FEATURE

>Canon printers are better, both for the business user and the home
>user. Period.
>
>

EXCEPT FOR AN OCCASSIONAL USER. THEN HP WITH INTEGRATED HEADS MAY BE
BETTER FOR THAT USER

zakezuke

unread,
Dec 10, 2005, 6:16:05 PM12/10/05
to
> Measkite said, "they know who makes the motherboards:

Do they? They report who's on the box which may or may not be who
actually manufactured it. This is more common among the cheaper
brands... and those more scummy OEM brownbox venders. And a
motherboard is much more than the company who assembled it... they are
filled with tons of parts made by verious people... and there have been
cases where a legit manufacturer got burned because they bought
mislabled transisters that failed in the field.

On this issue, with all due respect measkite, your point is moot.

zakezuke

unread,
Dec 10, 2005, 8:00:03 PM12/10/05
to
> Maybe the Canon uses up 3 times as much ink as the HP for the same yield of
> documents

This is a reasonable statement. the ip4000 using the 25ml bci-3e is
rated at 500pages. so about 20p/ml. The old HP #45a for example was
IIRC 42ml and rated for 833pages, so 19.8333p/ml. I don't honestly
know if the yield on the black is 200 pages or 450 pages. My memory
has faided as to the location of the ofical yields, and HP wasn't clear
as to their unit of measurement. Officedepot says 200 but they say
that for all the inks that have different volumes so i'm taking their
info with a grain of salt.

anyways "HP 02 Black, 16 ml" - knightcrawler's data
16ml @ 200 pages = 12.5p/ml
16ml @ 450 pages = 28.12p/ml.

I'd welcome better data to colaberate Authur's point... but I'm willing
to bet that HP's ink efficency has gone up by a factor of at least 50%,
where canon's efficency has stayed much the same since the bci-3 ink
series, and still quite similar in the cli-8 series.

aprestn5

unread,
Dec 11, 2005, 12:17:12 AM12/11/05
to
Panos Stokas wrote:
> * Canon has better drivers. I just can't stress how important that is
> for the overall system health.

I don't know that I'd agree with that. I've seen better drivers for my two
old Epsons (SC740 and SC860) and even an old HP 550C, than the one Canon
has for their iP4x00 printer, which (1) I had to go to their Japanese web
site for, and (2) is very basic - only allowed 600 dpi printing until I
found some tips on modifying the .ppd file.


Knightcrawler

unread,
Dec 11, 2005, 11:10:06 AM12/11/05
to
I find it a HUGE stretch to say the HP printers are 3 times as efficient as
Canon's. From what my cousin says he runs out of ink pretty quick when do a
run of pictures.

Also the 500 page rating has been tossed around a lot but it is not an
official rating.

The manual states 740 pages or 1500 pages using two different methods.

One would really need to take a test page filled with text and print on a
Canon and HP and find out in the real world what throughput they would
really have.

"zakezuke" <zakez...@yahoo.com> wrote in message
news:1134262803.6...@o13g2000cwo.googlegroups.com...

zakezuke

unread,
Dec 11, 2005, 7:56:45 PM12/11/05
to
> The manual states 740 pages or 1500 pages using two different methods.

The bci-3e was rated at 740 pages using a standard 1500 character per
page test at courior font 10 chacters per inch. This is in the service
manual which if you don't have it i'll be most happy to send you a
copy. This test was on the whole inadquate during the days of
dotmatrix printers let alone the use of proportional fronts. To give
you an idea at courier 10cpi we're talking an absolute max of 80
characters wide and 66 lines or or an absolute max of 5280. To be
fair, assuming 1 inch margins or 6 lines top, 6 lines bottom, and 10
characters left and right for a total of 60 x 64 we are still talking
3840 characters. The difference is a factor of 2.56 or 40%. Assuming
10cpi 1 inch margins (20 x 12) their estimate of 740 pages becomes
300pages or so. That's one jam packed piece of paper. Realisticly
speaking most pages have spaces between paragraphs. Let us assume 7
paragraphs with a single line between., and 7 extra lines for closure
(60x50). That's 3000 characters a factor of 2 or a difference of 50%
or 370 pages.

http://www.stevesforums.com/forums/view_topic.php?id=55083&forum_id=40&highlight=hp+02+black+yield
The 1500 page estimate is based on the ISO JIS-SCID #5 test. It's a
picture of a bike with an oversized wheel. You'll also notice the
bci-6 black is rated at 2000pages using this test when we all know that
on plain paper the pigmented black is used. It might be a decent
estimate of color use but not so much on the black.

The 500p estimate is based on the standard of 5% yield. This is
somewhere inbetween 1500 characters at 10cpi courier font and 3840
characters the same. It's a very reasonable estimate of how many pages
you would expect using a proportional front like Tahoma 12point,
average margins, and spaces between the paragraphs.

I appricate the fact that you are siting numbers which are 100%
accurate in the fact that these are the offical published numbers in
canon user manuals and service manuals. But check the data for your
self and my link to the image on steve's fourms. You'll see that the
tests they are using to achieve these numbers are bogus. No one prints
at courier 10cpi, nor is 1500 characters at this size a valid estimate
of an average page.

Boss Hog

unread,
Dec 11, 2005, 8:16:37 PM12/11/05
to
In article <1134349005.8...@g44g2000cwa.googlegroups.com>,
zakez...@yahoo.com says...

> This is in the service

> manual...

where do you folks get these from? PDF downloads, friendly local service
centres, employed in the industry?

zakezuke

unread,
Dec 11, 2005, 8:39:13 PM12/11/05
to
> where do you folks get these from? PDF downloads, friendly local service
> centres, employed in the industry?

Good question... in this case both knightcrawler and my self would have
gotten a pdf download from sites i'll site later in this responce.

I know knightcrawler keeps them on his site somewhere
http://pixma.webpal.info/

He and I would have gotten them from here
http://www.eserviceinfo.com/
http://www.fixyourownprinter.com/

Another resource for regular joes is to hunt around e-bay and there is
a gent who posts here from time to time selling manuals on CD-rom.

Message has been deleted

Arthur Entlich

unread,
Dec 11, 2005, 10:54:44 PM12/11/05
to
Since usually the manufacturers contract private companies to do their
yield ratings, or do them internally, There is , I believe still not an
absolute agreed upon formula, although I believe they are working on
something.

In general, they are supposed to use a 5% per color for rating yield,
but some companies use 5% of the "usually margined" printed area, rather
than the whole page, and there has been disagreement as to what a
"usually margined area' is. Is it 1" all around, 1.5" 2" ???

However, having said this, HP's newest line of printers is using a new
system to recirculate the inks and extract air, rather than dumping it
into the waste ink pads, so the ink gets used rather than in the "dump".

Considering how much ink is normally wasted on cleanings, this could
make a considerable difference. Also, if high color density inks are
used, with a very small dot, the ink can go a lot further.

Also, how the drivers distribute the ink amounts, especially, the low
colorant loaded inks, can make a huge difference. If a lot of low
colorant load inks get used when smaller dots of high colorant ink could
do, you will drain more ink. If the printer uses all colors to make
darker browns, greys and blacks, again more ink is used.

So, there is no simple answer. They need to come up with a standard
image that all manufacturers use, and the same level of quality is used.

The same goes for printing speed. HP, as just one example, quotes a 4 x
6" color photo print as taking 14 seconds from their 8250. But that's
for an image in draft mode, and that's starting with image number two,
the first image takes 22 seconds longer, due to driver and paper
loading. The draft image quality was considered a "lousy print" by
Popular Photography's reviewer. Using a better quality, a 4x6"
borderless print takes just over one minute (with a border takes about
35 minutes in medium mode), still very fast, and apparently a good image
quality, but hardly 14 seconds. I'm not singling out HP by the way, it
just happens to be the last review I read, as all the manufacturers do
the same thing.

So, caveat emptor is still the case with all of this until the industry
gets the hint that clients want honest and consistent information across
the board.

Art

Tony

unread,
Dec 12, 2005, 12:09:19 AM12/12/05
to

I have used www.manuals4you.com many times, excellent service, but you have to
pay which is fair enough. However the value for money is very good, you get
dozens of manuals on each CD and most manufacturers and models are available.
There is a Russian site that sells manuals also but I have no opinion about
their service or quality since I have never used them.
BTW fixyourownprinter.com no longer have free manuals available for bandwidth
reasons.
Tony

zakezuke

unread,
Dec 12, 2005, 2:43:48 AM12/12/05
to
> I just finished printing ~ 50 pages of pure black starting from almost full
> OEM BCI-3e on ip4000R. Single spaced Times at 10 pt plus some
> equations. Judging from the ink height, that used up about 25% of the
> capacity.

And the capacity of the sponge is about 20%. So assuming 25ml we have
5ml sponge 20ml tank. You believe you used 25% of the reservoir so you
believe you've used 5ml for 50 pages, or 10ml/page.

Based on the 5% yield 500p assuming also 25ml... that's 20p/ml

What i'm not clear on is you say pure black.. 10pt font numbers and
quations. Is this basicly full pages packed to the gills with info or
are there signifigent margins and spacing.

Message has been deleted

zakezuke

unread,
Dec 12, 2005, 3:56:47 AM12/12/05
to
> 10 pages/ml, I'm sure you meant.
> It was pretty heavy but, I'd say, to the point that more realistically
> reflects most people's idea of what "full text page" is.
>
> I guess no matter what it boils down to an old rule of thumb "a full
> rim on a cartridge" (give or take factor 2X).

Thanks for the correction.. i'm taking mussle relaxents and just
doubled the dose... while this would be a normal error for me i'm
erroring more normaly than normal.

Anyhow I'll have to top off my cartridges and do some hardcore
printing. On my road to recovery i'm learning Japanese and have lots
of things to print. I would do laser... but my laser is just heavy
enough that I should avoid moving it for at least another week. But my
unscientific evaluation of ink yields on the ip3000... the stuff I
typicaly print I tend to average one pack of paper per refill which
just so happens to be 500 sheets.

Arthur Entlich

unread,
Dec 12, 2005, 5:22:52 AM12/12/05
to
All of the below, and even ebay... you can buy many on line if you can't
find them at assorted web sites of other locations.

Art

Arthur Entlich

unread,
Dec 12, 2005, 5:41:18 AM12/12/05
to

zakezuke wrote:

>>I just finished printing ~ 50 pages of pure black starting from almost full
>>OEM BCI-3e on ip4000R. Single spaced Times at 10 pt plus some
>>equations. Judging from the ink height, that used up about 25% of the
>>capacity.
>
>
> And the capacity of the sponge is about 20%. So assuming 25ml we have
> 5ml sponge 20ml tank. You believe you used 25% of the reservoir so you
> believe you've used 5ml for 50 pages, or 10ml/page.

I think you lost a decimal there: .10ml/page, or one-tenth ml per page
or 1 ml per 10 pages. (I realize it's a typo, but I hought I would
correct it.)

Art

Arthur Entlich

unread,
Dec 12, 2005, 5:44:50 AM12/12/05
to
I'm not sure you should be operating "heavy" machines, let alone moving
them, while you are taking high doses of muscle relaxants ;-) Don't
they have warnings on those meds about that? ;-)

Art

Panos Stokas

unread,
Dec 12, 2005, 6:01:16 AM12/12/05
to
Maybe I didn't express my point quite well: I am not asking Tom's
Hardware to test alternative inks, but I do ask them to specify whether
a printer can be easily refilled or not.

Panos Stokas

unread,
Dec 12, 2005, 6:02:08 AM12/12/05
to
Very well. I shall post them my concerns... the main one being that
Tom's Hardware is changing from hardware enthusiasm to yet another
CNETsque consumer site.

zakezuke

unread,
Dec 12, 2005, 6:26:54 AM12/12/05
to
> I'm not sure you should be operating "heavy" machines, let alone moving
> them, while you are taking high doses of muscle relaxants ;-) Don't
> they have warnings on those meds about that? ;-)

Yes, they do... "Warning, operating this color inkjet while taking
mussle relaxents may result in prolonged periods of blank stares at the
pretty colors".

I'm not moving the laser for a while.

measekite

unread,
Dec 12, 2005, 1:35:10 PM12/12/05
to

Panos Stokas wrote:

THE MAJORITY PEOP[LE WHO BUY PRINTERS DO NOT GIVE A HOOT. THE MAJORITY
BY FAR OF PRINTER USER BUY OEM INK. SO WHY SHOULD THEY. A NON ISSUE

zakezuke

unread,
Dec 12, 2005, 5:48:25 PM12/12/05
to
> Measkite said... snipped per request

Given the fact that PCword at the very least wrote and artical on the
subject.. and this is your favorite mag... someone does indeed give a
*hoot*.

But the parent isn't talking about anysite, they are talking
tomshardware. a site of tweekers hackers and weirdos (no offence).
While I can not agree that it's practical to test two or three major
brands (Formulabs, Image Specalists, OCP), cartridge design and
refillability is something that I must concur should be a factor. You
your self said the people on here who refill are tweekers hackers and
weirdos... and tomshardware is or was for people who take soldering
irons, hammers, and hacksaws to their equipment. It's not for average
users but rather for people you've made it painfuly clear you have no
respect for.

Yours truly...

A man proud not to be respected by you.

Tony

unread,
Dec 12, 2005, 6:40:10 PM12/12/05
to

Bugger
I really must see what I can do with a hammer and hacksaw to one of my elderly
non-working PC's, perhaps I can turn it into a printer!
Thank goodness for the tinkerers, they invented the wheel amongst other things,
things that set us apart from the animals and Measekite.
Seriously though, for clarification. Measekite has recently stated things like
"95% of all users use OEM inks".
Firstly that is an outrageous lie, not supported by market research nor by my
personal experience.
Secondly if it were true then it follows that he can go away and talk to
another newsgroup about a subject dear to his heart since if only 5% use
non-OEM inks there is virtually no reason for his crusade.
Tony

zakezuke

unread,
Dec 12, 2005, 7:39:05 PM12/12/05
to
> I really must see what I can do with a hammer and hacksaw to one of my elderly
> non-working PC's, perhaps I can turn it into a printer!

That is a streach... but what isn't is taking some of those older
rackmounts and cutting a ATX sized hole in them. Or remounting a baby
atx power supply in a full sized AT style container. I my self have
modified 1/8 of inch steel dec cases to accept AT style boards.

> Seriously though, for clarification. Measekite has recently stated things like
> "95% of all users use OEM inks"

Even if that figure is accurate, which i'm willing to believe the
majority of printer users are using OEM... 5% is a significant market
share esp for such a young industry. Sure a minority but any figure
you can measure in terms of a percent for a market of 1 billion
units/year... that's no small potatoes. That's 50 million units sold
at about $5.00 (guessing) using products that in bulk cost under
$1.00/unit assuming recycling, perhaps $2.00 ish assuming not. so
we're talking a 250million dollar industry assuming measkite's number
is accurate... not huge but signifigent.

Numbers by http://www.freerecycling.com/... not the best resource but
quotable enough.

> Firstly that is an outrageous lie, not supported by market research nor by my
> personal experience

Probally is a figure he pulled from his bum... but even so if he's
going to say aftermarket is at least a 250million dollar industry I say
let him. A small piece of that pie is a legit business.

Tony

unread,
Dec 12, 2005, 11:27:54 PM12/12/05
to
"zakezuke" <zakez...@yahoo.com> wrote:
<snip>

>
>> Seriously though, for clarification. Measekite has recently stated things
>>like
>> "95% of all users use OEM inks"
>
>Even if that figure is accurate, which i'm willing to believe the
>majority of printer users are using OEM... 5% is a significant market
>share esp for such a young industry. Sure a minority but any figure
>you can measure in terms of a percent for a market of 1 billion
>units/year... that's no small potatoes. That's 50 million units sold
>at about $5.00 (guessing) using products that in bulk cost under
>$1.00/unit assuming recycling, perhaps $2.00 ish assuming not. so
>we're talking a 250million dollar industry assuming measkite's number
>is accurate... not huge but signifigent.
>
>Numbers by http://www.freerecycling.com/... not the best resource but
>quotable enough.
>
>> Firstly that is an outrageous lie, not supported by market research nor by my
>> personal experience
>
>Probally is a figure he pulled from his bum... but even so if he's
>going to say aftermarket is at least a 250million dollar industry I say
>let him. A small piece of that pie is a legit business.

See http://www.rechargermag.com/news.asp?id=200511505
Just one source, others vary from 15% to 35%
Tony

Tony

unread,
Dec 12, 2005, 11:49:58 PM12/12/05
to

See http://www.rechargermag.com/news.asp?id=200511505, one of several cources.
Tony

Tony

unread,
Dec 13, 2005, 2:07:58 PM12/13/05
to
Tony <TonytheT...@aim.com> wrote:
>"zakezuke" <zakez...@yahoo.com> wrote:
><snip>

>>
>>> Seriously though, for clarification. Measekite has recently stated things
>>>like
>>> "95% of all users use OEM inks"
>>
>>Even if that figure is accurate, which i'm willing to believe the
>>majority of printer users are using OEM... 5% is a significant market
>>share esp for such a young industry. Sure a minority but any figure
>>you can measure in terms of a percent for a market of 1 billion
>>units/year... that's no small potatoes. That's 50 million units sold
>>at about $5.00 (guessing) using products that in bulk cost under
>>$1.00/unit assuming recycling, perhaps $2.00 ish assuming not. so
>>we're talking a 250million dollar industry assuming measkite's number
>>is accurate... not huge but signifigent.
>>
>>Numbers by http://www.freerecycling.com/... not the best resource but
>>quotable enough.
>>
>>> Firstly that is an outrageous lie, not supported by market research nor by
>>>my
>>> personal experience
>>
>>Probally is a figure he pulled from his bum... but even so if he's
>>going to say aftermarket is at least a 250million dollar industry I say
>>let him. A small piece of that pie is a legit business.
>
>See http://www.rechargermag.com/news.asp?id=200511505
>Just one source, others vary from 15% to 35%
>Tony


Apologies for the reposts....technical problems.
Tony

Panos Stokas

unread,
Dec 13, 2005, 3:50:02 PM12/13/05
to
Please consider how many companies out there sell overclocking tools
(cooling boxes etc) and how many sell ink. The later outnumber the
former by far.

measekite

unread,
Dec 13, 2005, 3:53:04 PM12/13/05
to
WHO CARES ABOUT INK. YOU BUY A PRINTER AND YOU BUY THE REPLACEMENT
CARTS THE MFG RECOMMENDS. NOTHING IS MORE SIMPLE.

TOM'S HAS OUTGROWN THE OVERCLOCKING CROWD

zakezuke

unread,
Dec 13, 2005, 5:13:37 PM12/13/05
to
> Measkite said: Note snipped this time
> who cares about the ink?

Printers do. Seriously. And as you've noticed there are those in this
group who do care. You may call them churchies tweekers or weirdos....
but the fact that they exist proves you wrong on this point. Your
point is moot.

> tom's has outgrown the overclocking crowd

Hardly... you can say the soldering iron overclocking of boards is no
longer needed... that would be fair comment, but those of us paying
full price for hardware want the most bang for our buck... and tom's
goes out of their way to test equipment under extreme conditions. I
don't overclock my self but I respect any piece of hardware that will
operate above and beyond the labeled specifications.

m...@privacy.net

unread,
Dec 15, 2005, 8:24:08 AM12/15/05
to
In message <1134386814.6...@g49g2000cwa.googlegroups.com>,
zakezuke <zakez...@yahoo.com> writes

In that case they sounds like good muscle relaxants! Have you tried
putting a fractal program into a colour cycle?!

--
Timothy

zakezuke

unread,
Dec 15, 2005, 4:54:21 PM12/15/05
to
> In that case they sounds like good muscle relaxants! Have you tried
> putting a fractal program into a colour cycle?!

I did... but I couldn't take my eyes off the bottom of the coffee cup.

0 new messages