[prev in list] [next in list] [prev in thread] [next in thread] 

List:       koffice
Subject:    Re: karbon being DPI based (was karbon)
From:       James Richard Tyrer <tyrerj () acm ! org>
Date:       2003-10-26 6:57:55
[Download RAW message or body]

Thomas Zander wrote:
> On Fri, Oct 24, 2003 at 01:38:21AM -0700, James Richard Tyrer wrote:
> 
>> Thomas Zander wrote:
>> 
>>> On Wednesday 22 October 2003 09:06, James Richard Tyrer wrote:
>>> 
>>>> Thomas Zander wrote: Apparently not.  We just think that something
>>>>  new is a panacea.  It never is.
>>> 
>>> I can't place that 'apparently', I just told you that the problem 
>>> you have is solved in the RIP; what do you base your opposite 
>>> conclusion on?
>>> 
>> 
>> So, I conclude that you do not know what a Moir? pattern is.
> 
> Yes I do; I told you its solved in the RIP. You probably missed that 
> when you concluded that RIPS 'apparently' are not better then they were
>  10 years ago.

The question is not the RIP, it is the printers.  Laser printers are still
black and white.  NO shades of gray.

I'm still not sure you do; I just hope you don't since I spent quite
> some time trying to explain things to you;

No, you have wasted considerable time with presumptuous and arrogant
remarks.  You have offered no actual information to support your
presumptuous claims.

> the least thing I expect from you is that you read up on issues before 
> you comment on them.

I fully understand that anti-aliasing can help with Moré and other
resolution based artifacts.  But, what you don't appear to understand is that:

1.	AA will not entirely eliminate them.

2.	Laser printers have only 1 bit per pixel -- they don't do AA.
> 
>> I know of no RIP that solves that problem.
> 
> _all_ professional RIP software solves this problem. (if not older then,
>  say 3 years). Look at Barco, Xeicon, HP Harlequin software. HP sells 
> mid-range printers (max $3000) for companies that include the RIP 
> software either for each workstation; or for a network-based RIP. You 
> can probably find that kind of solutions for less.
> 
Since the problem is with the printer -- typical desktop 600 DPI laser
printer, there is no way that the RIP can solve the problem.
> 
>> But since GhostScript does not solve that problem, this is the 
>> important point for > 90% of the users.
> 
> 
> You are aware that moire is only a problem for one color printing, 
> right? And that rasterizing methods like dot diffusions, ordered dither,
>  rotating dots etc simply avoid this problem altogether?  Only 
> FloydSteinberg method of adaptive grayscale will have problems like you
>  mention.  And that is not commonly used.
> 
NONE of these will work with one bit per pixel -- or at least they will
fail at some point when the details are small enough.  It that case, the
smaller the detail, the worse the problem gets.
> 
>>> To repeat; the only component in the steam of usage that actually 
>>> can know what the output size is is the RIP, that part has to do the
>>>  conversion you fear is going wrong.
>>> 
>> 
>> My printer is sitting within view of me and I know that I am going to
>>  print on it.  Therefore, do I know the output size and the
>> resolution?
>> 
> 
> Do you? The avarage user you mentioned surely does not!

So what.  This user knows and that is all that matters to him.

But, I think that most users do know the resolution of their printers.  And 
I hope that users that use SVG to make small icons know what the final 
resolution of the smallest size will be.

> And if your printer breaks and you have to print in on your neighbors
> printer you will have to start drawing from scratch since your DPI
> settings are in your drawing. Scale your image and you loose again. Buy
> an image from an imagebank and you will (by your reasoning) be left with
> lousy images. Why in the real world ate those not a problem?: because
> the rasterizer solves it.
> 
Please try to understand that the raserizer is not a cure-all for all issues.

>>>> And, I think that your are not at all familiar with the current 
>>>> development version of The GIMP.
>>> 
>>> I am not.
> 
> ..
> 
>> I thought that anyone that professed to know as much as you would know
>>  that they are adding some vector support to The GIMP.
> 
> I can't work with the Gimp;  its a horrible application from a usability
>  perspective, it brings me nothing extra that a profession editor needs;
>  like having professional features like 16 bit color channels, CMYK etc.
>  In other words; nice for hobby; but not for anything outside of that.

Please get out your dictionary and look up arrogant.

>>>> GhostScript is the one that Linux normally uses.
>>> 
>>> I have no experience with that one
>> 
>> Then what are you talking about?
> 
> Postscript, even very cheap laser printers print postscript.
> 
?????

>>> ; although I do believe it is quite capable of the resterizing I 
>>> said it would need.
>>> 
>> 
>> Do you mean interpolation rather than rasterizing?
> 
> 
> Nope.  Rasterizing is not always interpolation since lpi and 
> dot-rotation as well as dot-gain settings are part of that process; with
>  only interpolation the result will look horrible.
> 
Printing a 100 DPI image at 600 DPI requires interpolation -- preferably
something better than just linear interpolation.  That is not rasterizing.
  Rasterizing is converting a vector image to a raster image.  Changing the 
resolution of a pixel image is NOT rasterizing.
> 
>>> I am more and more under the impression that you have little 
>>> experience in the world of Postscript and Rips;
>> 
>> That would be irrelevant because what I am talking about is printing 
>> to an office or desktop printer.  That is what most users do.
> 
> You are wrong that it is irrelevant since it is the sole output method 
> of any KDE application.  Any solution will _have_ to work in postscript.
> 
> 
Yes PostScript is the output method for most Linux applications.  So, most
users are going to either be using a printer with a built in PostScript
interpreter or GhostScript.  Neither of these alternatives is a
professional RIP or a professional high-resolution printer.
> 
> James, if you have no direct questions regarding this subject I will 
> consider this thread closed.

Perhaps it is just a translation problem.  But my perception is that you
are somewhat confused about what you discussed.  Your apparent skill at 
sophistic rhetoric gets in the way of a realistic discussion.  Perhaps you 
really believe all the questionable things you said.  In that case, I hope 
that the actual developers have a firmer grasp of the facts so that they do 
not make mistakes like Troll Tech made with the PostScript driver.

--
JRT

____________________________________
koffice mailing list
koffice@mail.kde.org
To unsubscribe please visit:
http://mail.kde.org/mailman/listinfo/koffice
[prev in list] [next in list] [prev in thread] [next in thread]