Re: [Inkscape-user] Inkscape Accurancy (was: Doublesided Print Alignment)
I guess for print resolutions (300 dpi and above) the accuracy when working with pixels is below 1/10 of a millimetre so working in mm should not be a pr
oblem. Or am I mistaken again?
Andi
SVG drawing accuracy is only limited by the number of significant digits you choose to use. It is not limited by the choice of units! Inkscape has a setting to control the number of significant digits (I think the default is eight). Look under "Inkscape Preferences --> SVG output".
SVG syntax allows lengths to be specified with one of the following units: px, em, ex, pt, pc, cm, mm, and in. The SVG spec defines 1in=90px. Thus one thousandth of an inch can be spcified as 0.09, .09px, 0.001in, or 0.0254000mm. You may find when you specify units other than pixels that Inkscape will silently convert to pixels; this is a good thing because not all programs honor the 1in=90px standard. Firefox will assume 1in=96px.
The following drawing yeilds three equal sized squres in Inkscape but in Firefox the first square is smaller. ########################################### <?xml version="1.0" standalone="no"?> <svg version="1.1" width="8.5in" height="11in" xmlns="http://www.w3.org/2000/svg" xmlns:svg="http://www.w3.org/2000/svg"> <g style="fill:none;stroke:green;stroke-width:5px"> <rect x="1in" y="1in" width="90" height="90"/> <rect x="1in" y="2.1in" width="1in" height="1in"/> <rect x="1in" y="3.2in" width="25.4mm" height="25.4mm"/> </g> </svg> ###########################################
Note: SVG syntax does not allow "<path>" elements to contain units. i.e. path elements can use pixels only.
-crjw
On Jan 8, 2011, at 9:23 PM, crjw wrote:
SVG syntax allows lengths to be specified with one of the following units: px, em, ex, pt, pc, cm, mm, and in. The SVG spec defines 1in=90px. Thus one thousandth of an inch can be spcified as 0.09, .09px, 0.001in, or 0.0254000mm. You may find when you specify units other than pixels that Inkscape will silently convert to pixels; this is a good thing because not all programs honor the 1in=90px standard. Firefox will assume 1in=96px.
Not quite.
The SVG spec defines the size of a px to be user agent dependent, but with a *default* of 90 DPI. Or 96 DPI. It depends on when the spec was referenced.
In any case, this value is only a suggested default for when the viewing software does not have any other information to go by.
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: zondag 9 januari 2011 6:36 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy (was: Doublesided PrintAlignment)
On Jan 8, 2011, at 9:23 PM, crjw wrote:
SVG syntax allows lengths to be specified with one of the
following units: px, em, ex, pt, pc, cm, mm, and in.
The SVG spec defines 1in=90px. Thus one thousandth of an inch can be spcified as 0.09,
.09px, 0.001in, or 0.0254000mm.
You may find when you specify units other than pixels that Inkscape will silently convert to pixels; this is a good thing
because not all programs honor the 1in=90px standard. Firefox will assume 1in=96px.
Not quite.
The SVG spec defines the size of a px to be user agent dependent, but with a *default* of 90 DPI. Or 96 DPI. It depends on when the spec was referenced.
In any case, this value is only a suggested default for when the viewing software does not have any other information to go by.
Is there a way to explicitly define the length of 'px' in SVG? If not, Tav, I think you are closest to the SVG developers, perhaps you can pass this issue to them?
cheers, Johan
On Jan 9, 2011, at 7:19 AM, J.B.C.Engelen@...2072... wrote:
Is there a way to explicitly define the length of 'px' in SVG? If not, Tav, I think you are closest to the SVG developers, perhaps you can pass this issue to them?
One common approach is to apply a viewBox to the top level svg object with some given physical size. Of course, things like transforms can start to confuse one's numbers. And then there is the problem that once a set size is applied, dynamic sizing is blocked on most viewing agents so that goes counter to what web designers and such need.
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: 09 January 2011 20:05 To: Inkscape User Community Cc: tavmjong@...206... Subject: Re: [Inkscape-user] Inkscape Accurancy (was: DoublesidedPrintAlignment)
On Jan 9, 2011, at 7:19 AM, J.B.C.Engelen@...2072... wrote:
Is there a way to explicitly define the length of 'px' in SVG? If not, Tav, I think you are closest to the SVG developers, perhaps
you
can pass this issue to them?
One common approach is to apply a viewBox to the top level svg object with some given physical size. Of course, things like transforms can start to confuse one's numbers. And then there is the problem that
once
a set size is applied, dynamic sizing is blocked on most viewing
agents
so that goes counter to what web designers and such need.
I think many users will be surprised at the uncertainty of size in Inkscape. (apparently the rulers and units in Inkscape are a lie!) If you draw a line with a certain length in Inkscape, you'd expect that there is a way to get exactly that line length (e.g. when printing or opening the file in another program) _if you want it to_. Of course one can always scale the image. On a website, one would just disregard the whole length metrics and do whatever the website designer wants it to be (effectively, the viewer should just do what it does now :).
Without a way to exactly specify lengths, Inkscape is unusable for technical drawing. Right now I do not even understand why SVG allows specifying lengths in inch _and_ px at the same time, without there being any definition of the ratio between 1 in and 1 px. The fix to SVG is extremely simple. What I want is the possibility of something like: <svg ... oneinch = "90px"> Viewers can just ignore this (like they silently do now), but at least now I can be certain that my 1 inch line will still be 1 inch when opening the file in Inkscape after a couple of years.
Cheers, Johan
On 11-01-09 03:10 PM, J.B.C.Engelen@...2072... wrote:
On a website, one would just disregard the whole length metrics and do whatever the website designer wants
(shudder) That gives me the creeps. Allowing a web designer to do whatever they want. Scary stuff.
The problem is with the monitors. They have to tell the OS what their resolution is (which some do) and their dpi (which none do). Until then, software must take its best guess. And anyone who thinks monitors should display images in real-life size is engaged in wishful thinking.
On Jan 9, 2011, at 12:48 PM, Shawn H Corey wrote:
(shudder) That gives me the creeps. Allowing a web designer to do whatever they want. Scary stuff.
The problem is with the monitors. They have to tell the OS what their resolution is (which some do) and their dpi (which none do). Until then, software must take its best guess. And anyone who thinks monitors should display images in real-life size is engaged in wishful thinking.
Most monitors I've used for years now report to the applications via the OS their DPI, including on OS X, MS Windows and Solaris & Linux via X11.
GDK provides us this handy method:
gdk_screen_get_resolution() http://library.gnome.org/devel/gdk/stable/GdkScreen.html#gdk-screen-get-reso...
Of course, one should note that 1) the function takes a screen object as a parameter... as more than one screen might be involved in a system. 2) a window might span more than one screen.
as a caution, the SVG spec does state: http://www.w3.org/TR/SVG/coords.html#Units ...
Note that use of px units or any other absolute unit identifiers can cause inconsistent visual results on different viewing environments since the size of "1px" may map to a different number of user units on different systems; thus, absolute units identifiers are only recommended for the ‘width’ and the ‘height’ on and situations where the content contains no transformations and it is desirable to specify values relative to the device pixel grid or to a particular real world unit size.
Now, the initial coordinate system in SVG is what matters: http://www.w3.org/TR/SVG/coords.html#InitialCoordinateSystem ...
If the SVG implementation is part of a user agent which supports styling XML documents using CSS2 compatible px units, then the SVG user agent should get its initial value for the size of a px unit in real world units to match the value used for other XML styling operations; otherwise, if the user agent can determine the size of a px unit from its environment, it should use that value; otherwise, it should choose an appropriate size for one px unit. In all cases, the size of a px must be in conformance with the rules described in CSS2 ([CSS2], section 4.3.2).
Quite complex, and allows for much variation.
The CSS spec includes some fun diagrams, including variable-sized pixels for display and print: http://www.w3.org/TR/CSS2/images/pixel1.png http://www.w3.org/TR/CSS2/images/pixel2.png
And that in CSS, a "reference pixel" is defined using "the visual angle". Fun, huh? Oh, and the latest CSS specs refer to 96 DPI, whereas originally they referenced 90 DPI.
On 11-01-09 08:16 PM, Jon Cruz wrote:
Now, the initial coordinate system in SVG is what matters: http://www.w3.org/TR/SVG/coords.html#InitialCoordinateSystem
Well, it's clear that they don't know what they're talking about. In the example given, they mixed pixels with points and expect it to look the same no matter the output device.
Pixels are a small but indeterminate size. There is no reason to assume that they are 90 or 96 or any other number per inch. Mixing pixels with real-world measurements is just asking for trouble.
Of source, this is the organization that renamed typefaces to font-families, so I guess stupidity is to be expected.
On 11-01-09 12:23 AM, crjw wrote:
You may find when you specify units other than pixels that Inkscape will silently convert to pixels; this is a good thing because not all programs honor
No, this is a bad thing. It means the renderer can change the size of the image. The only thing I want pixels for is icons. For anything else, I want the image to remain the size I given it, regardless of the software.
On Jan 9, 2011, at 6:08 AM, Shawn H Corey wrote:
No, this is a bad thing. It means the renderer can change the size of the image. The only thing I want pixels for is icons. For anything else, I want the image to remain the size I given it, regardless of the software.
Ah, but there's the rub. There are many definitions of "remain the size I given it".
Among others there is the question of intent. For example, does one always intend an image to span half the width of the display when I hold it in my hand, but only 1/30th of the display when I bring it up in my living room? Very often the answer to that one is "no".
And does one want to measure a distance in strict linear means on a physical object, or does one want the same retinal image size? (Apple has bumped this one up in the consciousness of the consumer, but W3C and CSS already addressed it)
On 11-01-09 01:57 PM, Jon Cruz wrote:
Ah, but there's the rub. There are many definitions of "remain the size I given it".
Among others there is the question of intent. For example, does one always intend an image to span half the width of the display when I hold it in my hand, but only 1/30th of the display when I bring it up in my living room? Very often the answer to that one is "no".
And does one want to measure a distance in strict linear means on a physical object, or does one want the same retinal image size? (Apple has bumped this one up in the consciousness of the consumer, but W3C and CSS already addressed it)
When I say I want an image that is 6 inches by 4 inches, I mean that and nothing else. The problems is with the monitors; they are measured in pixels. Printers, however, print to paper of standards sizes.
When you measure distance, you measure the physical distance. The size of the image on the retina is irrelevant.
On Sun, Jan 09, 2011 at 02:15:00PM -0500, Shawn H Corey wrote:
On 11-01-09 01:57 PM, Jon Cruz wrote:
Ah, but there's the rub. There are many definitions of "remain the size I given it".
Among others there is the question of intent. For example, does one always intend an image to span half the width of the display when I hold it in my hand, but only 1/30th of the display when I bring it up in my living room? Very often the answer to that one is "no".
And does one want to measure a distance in strict linear means on a physical object, or does one want the same retinal image size? (Apple has bumped this one up in the consciousness of the consumer, but W3C and CSS already addressed it)
When I say I want an image that is 6 inches by 4 inches, I mean that and nothing else. The problems is with the monitors; they are measured in pixels. Printers, however, print to paper of standards sizes.
When you measure distance, you measure the physical distance. The size of the image on the retina is irrelevant.
People with some kinds of poor eyesight routinely like *everything* to be twice as big. Evidently, you're not one of them.
-- hendrik
On 11-01-09 02:19 PM, Hendrik Boom wrote:
People with some kinds of poor eyesight routinely like*everything* to be twice as big. Evidently, you're not one of them.
That's called zoom. It does change the size, just the magnification.
On Sun, 9 Jan 2011 14:19:11 -0500 Hendrik Boom <hendrik@...2611...> wrote:
On Sun, Jan 09, 2011 at 02:15:00PM -0500, Shawn H Corey wrote:
On 11-01-09 01:57 PM, Jon Cruz wrote:
Ah, but there's the rub. There are many definitions of "remain the size I given it".
Among others there is the question of intent. For example, does one always intend an image to span half the width of the display when I hold it in my hand, but only 1/30th of the display when I bring it up in my living room? Very often the answer to that one is "no".
And does one want to measure a distance in strict linear means on a physical object, or does one want the same retinal image size? (Apple has bumped this one up in the consciousness of the consumer, but W3C and CSS already addressed it)
When I say I want an image that is 6 inches by 4 inches, I mean that and nothing else. The problems is with the monitors; they are measured in pixels. Printers, however, print to paper of standards sizes.
When you measure distance, you measure the physical distance. The size of the image on the retina is irrelevant.
People with some kinds of poor eyesight routinely like *everything* to be twice as big. Evidently, you're not one of them.
-- hendrik
Not posted for a while but this one really got my attention!
Then let them indeed rescale everything i.e. the entire desktop *display*. I have poor eyesight and if I want to see detail I'll zoom in, not attempt to change the underlying structure of a document.
I agree with Shawn in inch is an inch. I would be extremely annoyed if my PCB designs were arbitarily re-sacled :(
On 09.01.2011, at 20:40, Abrolag <abrolag@...16...> wrote:
On Sun, 9 Jan 2011 14:19:11 -0500 Hendrik Boom <hendrik@...2611...> wrote:
On Sun, Jan 09, 2011 at 02:15:00PM -0500, Shawn H Corey wrote:
On 11-01-09 01:57 PM, Jon Cruz wrote:
Ah, but there's the rub. There are many definitions of "remain the size I given it".
Among others there is the question of intent. For example, does one always intend an image to span half the width of the display when I hold it in my hand, but only 1/30th of the display when I bring it up in my living room? Very often the answer to that one is "no".
And does one want to measure a distance in strict linear means on a physical object, or does one want the same retinal image size? (Apple has bumped this one up in the consciousness of the consumer, but W3C and CSS already addressed it)
When I say I want an image that is 6 inches by 4 inches, I mean that and nothing else. The problems is with the monitors; they are measured in pixels. Printers, however, print to paper of standards sizes.
When you measure distance, you measure the physical distance. The size of the image on the retina is irrelevant.
People with some kinds of poor eyesight routinely like *everything* to be twice as big. Evidently, you're not one of them.
-- hendrik
Not posted for a while but this one really got my attention!
Then let them indeed rescale everything i.e. the entire desktop *display*. I have poor eyesight and if I want to see detail I'll zoom in, not attempt to change the underlying structure of a document.
I agree with Shawn in inch is an inch. I would be extremely annoyed if my PCB designs were arbitarily re-sacled :(
I guess your 1in x 1in PCB should be only shown at the width of 1in on your screen then...
I guess what I'm trying to say is that the whole discussion about absolute measurements can only be made if it also involves an output device.
Otherwise the only important thing is that Inkscape can work precisely in any given measurement system and that one inch is always 90 pixels so that relative size keeps the same. That's good enough for everything, mapping that to an output device is a whole different story...
Andi
-- Will J Godfrey http://www.musically.me.uk Say you have a poem and I have a tune. Exchange them and we can both have a poem, a tune, and a song.
Gaining the trust of online customers is vital for the success of any company that requires sensitive data to be transmitted over the Web. Learn how to best implement a security strategy that keeps consumers' information secure and instills the confidence they need to proceed with transactions. http://p.sf.net/sfu/oracle-sfdevnl _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user
-----Original Message----- From: Andreas Neustifter [mailto:andreas.neustifter@...155...] Sent: zondag 9 januari 2011 22:43 To: Inkscape User Community Cc: inkscape-user@lists.sourceforge.net Subject: Re: [Inkscape-user] Inkscape Accurancy
[...] Otherwise the only important thing is that Inkscape can work precisely in any given measurement system and that one inch is always 90 pixels so that relative size keeps the same. That's good enough for everything, mapping that to an output device is a whole different story...
The problem is that apparently, it says nowhere that 1 in = 90 px always.
-johan
On Jan 9, 2011, at 3:45 PM, J.B.C.Engelen@...2072... wrote:
-----Original Message----- From: Andreas Neustifter [mailto:andreas.neustifter@...155...] Sent: zondag 9 januari 2011 22:43 To: Inkscape User Community Cc: inkscape-user@lists.sourceforge.net Subject: Re: [Inkscape-user] Inkscape Accurancy
[...] Otherwise the only important thing is that Inkscape can work precisely in any given measurement system and that one inch is always 90 pixels so that relative size keeps the same. That's good enough for everything, mapping that to an output device is a whole different story...
The problem is that apparently, it says nowhere that 1 in = 90 px always.
No, that is up to dynamic display systems and/or users to decide.
And to control such, that is where the viewBox attribute comes in.
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: maandag 10 januari 2011 2:18 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy
On Jan 9, 2011, at 3:45 PM, J.B.C.Engelen@...2072... wrote:
-----Original Message----- From: Andreas Neustifter [mailto:andreas.neustifter@...155...] Sent: zondag 9 januari 2011 22:43 To: Inkscape User Community Cc: inkscape-user@lists.sourceforge.net Subject: Re: [Inkscape-user] Inkscape Accurancy
[...] Otherwise the only important thing is that Inkscape can work precisely in any given measurement system and that one
inch is always
90 pixels so that relative size keeps the same. That's good enough for everything, mapping that to an
output device
is a whole different story...
The problem is that apparently, it says nowhere that 1 in = 90 px always.
No, that is up to dynamic display systems and/or users to decide.
Of course, but right now the user does not have the option of choosing the exact dimensions the SVG author intended. What I meant with "1 in = 90 px always" is not how the image should be shown but how the image should be interpreted.
And to control such, that is where the viewBox attribute comes in.
So <svg width="2in" height="2in" viewBox="0 0 180 180"> would relate lengths in inches to px for the whole document right? Perhaps we can do that then, and use the page's size for the width, height and viewbox. (using Inkscape's choice of 1 in = 90 px)
-Johan
Hi all!
On 10.01.2011, at 12:11, J.B.C.Engelen@...2072... wrote:
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: maandag 10 januari 2011 2:18 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy
And to control such, that is where the viewBox attribute comes in.
So <svg width="2in" height="2in" viewBox="0 0 180 180"> would relate lengths in inches to px for the whole document right? Perhaps we can do that then, and use the page's size for the width, height and viewbox. (using Inkscape's choice of 1 in = 90 px)
Just for the record: I like this idea, but make it optional.
And sorry for making anyone angry with my comments, but judging by the comments this seems to be a big issue for some.
Andi
On Jan 10, 2011, at 3:11 AM, J.B.C.Engelen@...2072... wrote:
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: maandag 10 januari 2011 2:18 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy
No, that is up to dynamic display systems and/or users to decide.
Of course, but right now the user does not have the option of choosing the exact dimensions the SVG author intended. What I meant with "1 in = 90 px always" is not how the image should be shown but how the image should be interpreted.
Yes, there is. A viewBox attribute will do just that. If a user just wants the scalable graphics to scale, then viewBox can be left off. If not, it can be added.
Also... CSS changed from a fallback of 90 DPI to 96 DPI.
And to control such, that is where the viewBox attribute comes in.
So <svg width="2in" height="2in" viewBox="0 0 180 180"> would relate lengths in inches to px for the whole document right? Perhaps we can do that then, and use the page's size for the width, height and viewbox. (using Inkscape's choice of 1 in = 90 px)
Yes... with the caveat that 90 DPI is an outdated default that will most likely A) be replaced by the newer 96 and B) will be drawn from the DPI reported from GTK instead of hardcoded (as is the recommendation for behavior)
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: Tuesday, January 11, 2011 09:50 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy
On Jan 10, 2011, at 3:11 AM, J.B.C.Engelen@...2072... wrote:
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: maandag 10 januari 2011 2:18 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy
No, that is up to dynamic display systems and/or users to decide.
Of course, but right now the user does not have the option
of choosing
the exact dimensions the SVG author intended. What I meant with "1 in = 90 px always" is not how the
image should be
shown but how the image should be interpreted.
Yes, there is. A viewBox attribute will do just that. If a user just wants the scalable graphics to scale, then viewBox can be left off. If not, it can be added.
Also... CSS changed from a fallback of 90 DPI to 96 DPI.
And to control such, that is where the viewBox attribute comes in.
So <svg width="2in" height="2in" viewBox="0 0 180 180">
would relate
lengths in inches to px for the whole document right? Perhaps we can do that then, and use the page's size for the width, height and viewbox. (using Inkscape's choice of 1 in = 90 px)
Yes... with the caveat that 90 DPI is an outdated default that will most likely A) be replaced by the newer 96 and B) will be drawn from the DPI reported from GTK instead of hardcoded (as is the recommendation for behavior)
This is very sad news. Right now I am very afraid that someone will change 90 to 96 in Inkscape such that all older files will be opened with wrong dimensions (in inches, cm, ...).
-Johan
On 11 Jan 2011, at 08:49, Jon Cruz <jon@...204...> wrote:
On Jan 10, 2011, at 3:11 AM, J.B.C.Engelen@...2072... wrote:
-----Original Message----- From: Jon Cruz [mailto:jon@...204...] Sent: maandag 10 januari 2011 2:18 To: Inkscape User Community Subject: Re: [Inkscape-user] Inkscape Accurancy
No, that is up to dynamic display systems and/or users to decide.
Of course, but right now the user does not have the option of choosing the exact dimensions the SVG author intended. What I meant with "1 in = 90 px always" is not how the image should be shown but how the image should be interpreted.
Yes, there is. A viewBox attribute will do just that. If a user just wants the scalable graphics to scale, then viewBox can be left off. If not, it can be added.
Surely if you want it scalable you define it in px, which isn't a fixed unit. If you define it in inches it should be an exact dimension.
Also... CSS changed from a fallback of 90 DPI to 96 DPI.
This default needs to not be updated in inkscape until we work out a way of not screwing up legacy files. To hell with complying with an updated CSS default until we comply with using units other than px.
And to control such, that is where the viewBox attribute comes in.
So <svg width="2in" height="2in" viewBox="0 0 180 180"> would relate lengths in inches to px for the whole document right? Perhaps we can do that then, and use the page's size for the width, height and viewbox. (using Inkscape's choice of 1 in = 90 px)
Yes... with the caveat that 90 DPI is an outdated default that will most likely A) be replaced by the newer 96 and B) will be drawn from the DPI reported from GTK instead of hardcoded (as is the recommendation for behavior)
Niether a or b should be done until it doesn't screw up old files.
Incidentally the spec says we can use inches mm etc, so why don't we allow that? Surely if that got fixed this issue gos away, seems like a pretty major spec non compliance on our part...
On 11-01-13 03:33 PM, John Cliff wrote:
This default needs to not be updated in inkscape until we work out a way of not screwing up legacy files. To hell with complying with an updated CSS default until we comply with using units other than px.
It seems to me that the only way is to lock in at 90 pixels per inch.
On Thu, Jan 13, 2011 at 03:44:03PM -0500, Shawn H Corey wrote:
On 11-01-13 03:33 PM, John Cliff wrote:
This default needs to not be updated in inkscape until we work out a way of not screwing up legacy files. To hell with complying with an updated CSS default until we comply with using units other than px.
It seems to me that the only way is to lock in at 90 pixels per inch.
Or:
Provide an explicit operation to convert measurements to pixels or from pixels, at the user's request. Or convert the SVG file to one that overrides context to explicitly request 90 pixels per inch.
Unless the file contains an explicit marker that it whould be rendered with CSS default conversions, use 90 pixels per inch within Inkscape. Put such a marker in all new inkscape files, including edited old files, specifying pisels per inch explicitly or explicitly leaving it up to context, as the user requests.
-- hendrik
-----Original Message----- From: Andreas Neustifter [mailto:andreas.neustifter@...155...] Sent: zondag 9 januari 2011 22:43 To: Inkscape User Community Cc: inkscape-user@lists.sourceforge.net Subject: Re: [Inkscape-user] Inkscape Accurancy
On 09.01.2011, at 20:40, Abrolag <abrolag@...16...> wrote:
On Sun, 9 Jan 2011 14:19:11 -0500 Hendrik Boom <hendrik@...2611...> wrote:
On Sun, Jan 09, 2011 at 02:15:00PM -0500, Shawn H Corey wrote:
On 11-01-09 01:57 PM, Jon Cruz wrote:
Ah, but there's the rub. There are many definitions of
"remain the size I given it".
Among others there is the question of intent. For
example, does one always intend an image to span half the width of the display when I hold it in my hand, but only 1/30th of the display when I bring it up in my living room? Very often the answer to that one is "no".
And does one want to measure a distance in strict linear
means on a
physical object, or does one want the same retinal image size? (Apple has bumped this one up in the consciousness of
the consumer,
but W3C and CSS already addressed it)
When I say I want an image that is 6 inches by 4 inches,
I mean that
and nothing else. The problems is with the monitors; they are measured in pixels. Printers, however, print to paper of
standards sizes.
When you measure distance, you measure the physical
distance. The
size of the image on the retina is irrelevant.
People with some kinds of poor eyesight routinely like
*everything*
to be twice as big. Evidently, you're not one of them.
-- hendrik
Not posted for a while but this one really got my attention!
Then let them indeed rescale everything i.e. the entire desktop *display*. I have poor eyesight and if I want to see detail
I'll zoom
in, not attempt to change the underlying structure of a document.
I agree with Shawn in inch is an inch. I would be extremely
annoyed if
my PCB designs were arbitarily re-sacled :(
I guess your 1in x 1in PCB should be only shown at the width of 1in on your screen then...
??
I guess what I'm trying to say is that the whole discussion about absolute measurements can only be made if it also involves an output device.
Not so. If I make a drawing of a square of 1x1 cm, I want the SVG document to say it is 1x1 cm. Doesn't matter how it is displayed on screen. Then I can mail the document to someone else, he opens it, and sees: ah, the square is 1x1 cm. Think about technical drawing, CAD stuff.
-Johan
On 11-01-09 06:51 PM, J.B.C.Engelen@...2072... wrote:
Not so. If I make a drawing of a square of 1x1 cm, I want the SVG document to say it is 1x1 cm. Doesn't matter how it is displayed on screen. Then I can mail the document to someone else, he opens it, and sees: ah, the square is 1x1 cm.
No so. Like you, he would only see what his monitor displays. But when he prints it or sends it to an automated tool, it comes out 1x1 cm.
-----Original Message----- From: Shawn H Corey [mailto:shawnhcorey@...155...] Sent: maandag 10 januari 2011 1:15 To: inkscape-user@lists.sourceforge.net Subject: Re: [Inkscape-user] Inkscape Accurancy
On 11-01-09 06:51 PM, J.B.C.Engelen@...2072... wrote:
Not so. If I make a drawing of a square of 1x1 cm, I want the SVG document to say it is 1x1 cm. Doesn't matter how it is displayed on screen. Then I can mail the document to someone else, he
opens it, and
sees: ah, the square is 1x1 cm.
No so. Like you, he would only see what his monitor displays. But when he prints it or sends it to an automated tool, it comes out 1x1 cm.
(Sorry, I am getting angry. It would help if people would try to understand what we are saying, instead of repeating their earlier statements)
When he sends it to an automated tool, the tool software says: "hey this is SVG. And SVG does not specify what 1 px length is. So how large should this thing be?"
Apparently, I have still not explained our worries well enough. What we want is a document that specifies a line to have a certain length. The document should say: Line = 2 inch. This information is *not* to be used by the display system, or whatever output system *unless* it is desired to do so by the user. It seems that paths can only be specified in units of "px", which means we have to exactly relate the "px" unit to an absolute unit like inches.
Again, our concerns have nothing, whatsoever, to do with how the line is displayed. It has to do with how the dimensions are saved to SVG, and whether those dimensions is precisely defined or not.
If I draw a line of "2 inch" in Inkscape, save the SVG, and then open the SVG in another program. That program should also say that the line is 2 inch long. *Independent* of the DPI choice that program makes, or whatever.
The use case scenario is for a technical drawing, say a printed circuit board layout. It is not possible to specify the line to be 2 inch long in Inkscape, it is converted to "px" instead. So, because of the 90dpi choice of Inkscape, the line is 180px long in the SVG document. But the document should then specify 1 inch = 90 px, perhaps only in an Inkscape namespaced parameter if still nobody understands this important issue.
Yes, one solution is to always draw a scale bar on the drawing. Such that you can rescale the image accordingly when opening it in a different program. But this is ofcourse quite cumbersome.
-Johan
On Sun, Jan 09, 2011 at 09:08:05AM -0500, Shawn H Corey wrote:
On 11-01-09 12:23 AM, crjw wrote:
You may find when you specify units other than pixels that Inkscape will silently convert to pixels; this is a good thing because not all programs honor
Maybe someone should look at the actual SVG file as text so see if it contains inches or pixels, or a mixture of both. And whether if specifies a conversion factor.
Of course internally, inkscape has to convert to pixels because it is displaying onto a screen. But it shouldn't save that way unless the file explicitly specifies the size of a pixel. ()of course if the user specifies pixels, it should just keep the pixel measurement.)
No, this is a bad thing. It means the renderer can change the size of the image. The only thing I want pixels for is icons. For anything else, I want the image to remain the size I given it, regardless of the software.
There's a problem if both pixels and inches are explicitly used. <then the conversion can affect the image structure.
-- hendrik
participants (8)
-
unknown＠example.com -
Abrolag -
Andreas Neustifter -
crjw -
Hendrik Boom -
John Cliff -
Jon Cruz -
Shawn H Corey