Epson Stylus II or HP 660Cse?
Mike Donovan
the HP 660Cse. We know that the HP has "colorsmart technology" and has
good reviews, but we demoed the Epson II and like it alot. ($450))
Any suggestions and comments would be welcomed. Please e-mail
OP...@Prodigy.com Thanks!
Cazares
Epson. I'm happy, especially with service and support. The output's also
better.
Please note: I also have an HP 4, which I use for normal text stuff.
It's faster (it also cost 3 times as much). But the Epson was the cheapest
per page for color copy and that's important. It's good enough to produce
camera ready copy for ads, flyers, business cards. I've been _extremely_
happy with being able to run heavy card stock through it, which I can't do
with the HP.
My major complaint is that Epson is slow. But that's because I'm used to
lasers.
Finally, I'm unhappy with HP's slowness and stickiness with regard to
updating drivers. Win95 has been having problems with printer support. HP
sits back and replies that it's Microsoft's problem. Epson went out and
formed a driver-writing team. I prefer that active effort to help users. HP
can go to Blame City if it wants - but that leaves me waiting around for
Microsoft. It feels pretty helpless.
Craig Gowdey
believe it will ocntinue to work for you? Any problems setting up, etc? Can you just print when
your ready, or is there special set up required to get things "right"?
In trying a demo model at CompUSA it fed a whole stack of paper through on 1 print pass. Not
particularly impressive. But I have seen the Tiger on glossy paper - WOW~!
I'd appreciate your feedback. thx
Craig Gowdey
just checkin'
Louis Jaffe
>My major complaint is that Epson is slow. But that's because I'm used to
>lasers.
Stay tuned.. there's a fix for the Epson's slow speed, at least on Macs.
Details forthcoming. LJ
Laurel Hickey
>I've observed multi sheet feeds with the Epson, have you seen that problem?
>just checkin'
I've been using lots of different papers in ES II and haven't had any
problem with multi sheet feeds. Zip. Nada. This includes long print runs of
10 to 30 hours straight.
Now if I could just convince it that PM6 and Powerpoint weren't enemy
applications ....
Laurel
lhi...@mindlink.bc.ca
Louis Jaffe
>I've observed multi sheet feeds with the Epson, have you seen that problem?
I've not had much problem with this. Seems to help to have a decent sized
stack of paper in the feed hopper. LJ
t_...@ix.netcom.com
of all, color matching is a very complex science. When using different
devices (i.e. monitor, printer, scanner), colors may not match due to each
device using a different 'color space.' For example, if you view 'pure' red
on 10 different monitors, you'd notice that the reds are different on each
monitor.
The same applies with printers. HP, EPSON, Canon, etc., all use various
ink types, so colors vary on different printers. As a result, you may
believe an EPSON will provide you with better color matches, while your
friend may find the exact EPSON to be crappy when it comes to printing a
duplicate image on his/her system, trades it in for an HP, and finds that
HP's are 'better' than the EPSON (in his/her case).
What it boils down to is that color matching methods (i.e. Pantone,
ColorSync 2.0, ICM [Win95 standard]) drivers need to be written still for
all devices, the best scenario being having color matching (CM) support
built-in to the OS. So, when it comes to CM support, Pantone currently is
the hands down winner, even though its an application-specific
implementation (i.e. Photoshop, Pagemaker etc., support Pantone), with no
operating system support. ICM and ColorSync on the other hand, are OS-based
APIs for color matching, but as of now, few software companies write to that
spec.
So what does all this babble mean? Well, easy to use, cheap color
matching is still far away. The most practical solution is Pantone-based
Color Matching Software (CMS), with the other OS-based CMS's, like ICM and
ColorSync, hopefully right around the corner. So next time when it comes to
color, look at the results of test printouts from computers at stores to see
how close they match. And remember, colors still may not match when you use
it with your home system. Despite the marketing hype of 'true color,' the
likes of HP's ColorSmart mostly emphasizes certain primary colors that are
used a lot (i.e. red, blue) to give the illusion that colors look closer to
what's on the monitor. Although that's not color matching, I guess that it's
better than nothing. But be forewarned, 'results may vary.'
I still like indicators like resolution and platform support as the
primary reasons to buy, vs. any promise by a salesperson that a printer will
print out the colors that you see on the screen. With color macthing for the
average home consumer, it's relies on plain luck, although more than likely,
most people will be satisfied with the colors they get from their printers.
But a few may be disappointed.
So next time you think about color matching, just remember, whether the
printer is an HP, EPSON, or whatever doesn't count. You still have to see it
with your own eyes to make sure.
Bob Taylor
: of all, color matching is a very complex science. When using different
: devices (i.e. monitor, printer, scanner), colors may not match due to each
: device using a different 'color space.' For example, if you view 'pure' red
: on 10 different monitors, you'd notice that the reds are different on each
: monitor.
: The same applies with printers. HP, EPSON, Canon, etc., all use various
: ink types, so colors vary on different printers. As a result, you may
: believe an EPSON will provide you with better color matches, while your
: friend may find the exact EPSON to be crappy when it comes to printing a
: duplicate image on his/her system, trades it in for an HP, and finds that
: HP's are 'better' than the EPSON (in his/her case).
You've hit on one of the key points here - color matching is not absolute.
With the exception of the production graphic arts crowd, most users don't
actually want the exactly "right" color, they want their print out to
look good - which may be quite different than the original. For example,
on most monitors, green is pretty much a pastel. If I'm printing a business
chart, I want a nice, deep, solid green, not a pastel. In images, the
sky needs to be blue and the grass needs to be green - even if the original
is not very blue or not very green.
: What it boils down to is that color matching methods (i.e. Pantone,
: ColorSync 2.0, ICM [Win95 standard]) drivers need to be written still for
: all devices, the best scenario being having color matching (CM) support
: built-in to the OS. So, when it comes to CM support, Pantone currently is
: the hands down winner, even though its an application-specific
: implementation (i.e. Photoshop, Pagemaker etc., support Pantone), with no
: operating system support. ICM and ColorSync on the other hand, are OS-based
: APIs for color matching, but as of now, few software companies write to that
: spec.
True, but even when all of this software gets squared away, we won't
"be there" yet. The reason is the different color gamuts - some colors
can't be displayed on some/other devices. Most monitors can display
colors most printers can't print, and most printers can print colors that
most monitors can't display. In addition, even when a printer/monitor
can display a color, they will display/print some colors better than
others (i.e., on most color printers, very light colors show pronounced
dithering artifacts). If we shoot for a "perfect match", we have to
figure out what to do with out-of-gamut colors (a non-trivial problem)
and we need to figure out how to make it look good. Often, real
users prefer something non-matched to something matched due to the
marking/display technologies involved, which isn't going to be solved
by software.
: So what does all this babble mean? Well, easy to use, cheap color
: matching is still far away. The most practical solution is Pantone-based
: Color Matching Software (CMS), with the other OS-based CMS's, like ICM and
: ColorSync, hopefully right around the corner. So next time when it comes to
: color, look at the results of test printouts from computers at stores to see
: how close they match. And remember, colors still may not match when you use
: it with your home system. Despite the marketing hype of 'true color,' the
: likes of HP's ColorSmart mostly emphasizes certain primary colors that are
: used a lot (i.e. red, blue) to give the illusion that colors look closer to
: what's on the monitor. Although that's not color matching, I guess that it's
: better than nothing. But be forewarned, 'results may vary.'
Actually, ColorSmart does a variety of things (many of which are proprietary).
It does, however, do full color/gamut matching, but the goal isn't necessarily
an exact match. Rather, it attempts to make good looking images - instead
of technically "correct" but poor looking images. It also works with ICM
on Windows and ColorSync on the Macintosh to handle "true" matching, but
tests I've seen have said that most users actually prefer the output
with ICM/ColorSync off. Graphics arts professionals need an exact match,
but most consumers neither need nor want this. This is why, in an
industry that supposedly has this all figured out, Fuji & Kodak films
give substantially different color - they're trying to optimize the way
their films look, but not necessarily get the exactly perfect color
balance.
: I still like indicators like resolution and platform support as the
: primary reasons to buy, vs. any promise by a salesperson that a printer will
: print out the colors that you see on the screen. With color macthing for the
: average home consumer, it's relies on plain luck, although more than likely,
: most people will be satisfied with the colors they get from their printers.
: But a few may be disappointed.
Although it's used widely, dot placement resolution (which is what most
companies advertise) is a pretty poor metric, but it's unfortunately the
one most people jump to. Most 24-pin dot matrix printers are theoretically
360DPI in color - but you'd have a hard time convincing me they provide
better color print quality than a 300DPI (in color) DeskJet 855C. The
resolution is a factor, but others such as dot size/shape (note that very
few of the "720 DPI" printers on the market have a dot size that would
be correct for 720 DPI), ink compostion, ink interaction with paper,
color matching technology, halftoning technology, etc...
A good analogy is a sports car. When buying a sports car, should you
buy the one with the engine with the highest RPMs? If so, my chain
saw is a screamer. What you really do is drive the car - since the
performance of the car is actually a fairly complex interaction between
engine performance (which is partly max RPM, but also depends on the
torque curve, turbo vs. normally aspirated, etc.), the transmission,
the tires, the suspension, aerodynamics, etc.. Picking any one on which
to base your decision would be a mistake - just as it would be in
picking a color printer.
: So next time you think about color matching, just remember, whether the
: printer is an HP, EPSON, or whatever doesn't count.
I guess I'd have to disagree with you on that one ;-)
: You still have to see it
: with your own eyes to make sure.
:
This I'd agree with - but most retailers make this pretty difficult. The
ideal would be printing the same image on each printer in similar print
settings (best vs. normal vs. draft, etc.) on the paper you expect to use
most of the time. For most people, this a copier paper or a bond. If
you're going to print primarily on special papers, use the appropriate
special paper from each manufacturer. In general, I think you'll find
that most look pretty good on their special papers, but all but a few
( :-)) drop pretty substantially on plain papers.
Hope this helps -
Bob Taylor
- not acting as an official representative of ...
HP Vancouver
t_...@ix.netcom.com
> Actually, ColorSmart does a variety of things (many of which are
proprietary).
> It does, however, do full color/gamut matching, but the goal isn't
necessarily
> an exact match. Rather, it attempts to make good looking images - instead
> of technically "correct" but poor looking images. It also works with ICM
> on Windows and ColorSync on the Macintosh to handle "true" matching, but
> tests I've seen have said that most users actually prefer the output
> with ICM/ColorSync off. Graphics arts professionals need an exact match,
> but most consumers neither need nor want this. This is why, in an
> industry that supposedly has this all figured out, Fuji & Kodak films
> give substantially different color - they're trying to optimize the way
> their films look, but not necessarily get the exactly perfect color
> balance.
I think that reaffirms the theme of my posting that color is still a matter
of preference. Unless it's a true color match to what's on the screen, it's
hard to expect a sizable majority of customers to 'like' a particular color
printout. Anotherwords, some people will worship ColorSmart; some won't.
Once a customer is satisfied with what he/she has created on-screen,
including an image's color selection, it should be the goal of the printer
to reproduce that as close as possible, not add color enhancements which
'certain' people prefer. Color matching should be the goal, giving consumers
the ability to find the colors they like best on-screen, instead of using
various (and conflicting) driver manipulations to attain a 'truer' color.
Ultimately, if a customer is satisfied with the appearance with the image
on-screen, the printer should be able to print it. So far that hasn't
materialized, but I'll wait just a while longer.
Bob Taylor
Glad I could be of help.
: > Actually, ColorSmart does a variety of things (many of which are
: proprietary).
: > It does, however, do full color/gamut matching, but the goal isn't
: necessarily
: > an exact match. Rather, it attempts to make good looking images - instead
: > of technically "correct" but poor looking images. It also works with ICM
: > on Windows and ColorSync on the Macintosh to handle "true" matching, but
: > tests I've seen have said that most users actually prefer the output
: > with ICM/ColorSync off. Graphics arts professionals need an exact match,
: > but most consumers neither need nor want this. This is why, in an
: > industry that supposedly has this all figured out, Fuji & Kodak films
: > give substantially different color - they're trying to optimize the way
: > their films look, but not necessarily get the exactly perfect color
: > balance.
: I think that reaffirms the theme of my posting that color is still a matter
: of preference. Unless it's a true color match to what's on the screen, it's
: hard to expect a sizable majority of customers to 'like' a particular color
: printout. Anotherwords, some people will worship ColorSmart; some won't.
Actually, the vast majority of people do - it's through extensive real
user testing that we've determined what people prefer - and it's rarely
an exact match to their screen. For a small percentage that are very
color knowledgeable and particular, they prefer an exact match. For
the majority of consumers, they want something that looks good, and
they typically don't even know how to adjust an image so it looks
good on their monitor. If it looks good on their monitor, they want
it to look good on the printout - if it looks bad on the monitor, they
still want it to look good on the printout. They rarely care that
it looks the "same", just that it looks good.
: Once a customer is satisfied with what he/she has created on-screen,
: including an image's color selection, it should be the goal of the printer
: to reproduce that as close as possible, not add color enhancements which
: 'certain' people prefer. Color matching should be the goal, giving consumers
: the ability to find the colors they like best on-screen, instead of using
: various (and conflicting) driver manipulations to attain a 'truer' color.
For the small segment of the market that goes to the trouble of creating
images that they really care exactly what the colors are, this is true.
For the majority of customers, they want "the looks good thing". Full
system-based color matching will help, but users will typically want
the thing that looks good rather than the thing that is "right". Think
about it from an audio standpoint - most people actually don't like
their stereos "flat" - they want the bass cranked up and the treble
singing the cymbals, because it sounds "good". A lot more people want
boomboxes than high end, perfectly flat audiophile systems. That
doesn't mean that either are the wrong answer - only that different
people prefer different results. While you may be an "audiophile",
most people aren't - and any company that tries to sell "audiophile"
equipment to the mainstream market will likely lose in the marketplace.
: Ultimately, if a customer is satisfied with the appearance with the image
: on-screen, the printer should be able to print it. So far that hasn't
: materialized, but I'll wait just a while longer.
If the majority of customers ever get sophisticated enough to manipulate
images so that they are always "good" on screen - and they are always
able to deal with colors that may not look "good" on a printer that look
"good" on the screen - then maybe this will be true. For quite a while,
though, people are going to care more about "good" than "right" - and
for the purposes of making customers satisfied, right is by definition
whatever the customer prefers. I recently had a discussion with a
salesman at a high-end video store who mentioned that, for the few
discriminating customers who cared, they had to readjust the vast
majority of TV's they sell. This is because TV manufacturers crank
the color balance/vibrancy of most TVs, since this is what comsumers
prefer. It isn't "correct", but it is the right answer for most
customers.
Bob Taylor
HP Vancouver