[svg-developers] Re: real world unit vs screen resolution or display device
--- In svg-developers@yahoogroups.com, "kanmac1" <[EMAIL PROTECTED]> wrote: > we've got a major problem with SVGs. It may have an absolutely > distroying impact on our application. We use Adobe SVG Viewer 3.0 in > the IE6.0. > Our charts have fixed width and height given in real world units (like > mm or px, - whatever) and they must be displayed exactly so on the > screen - independent of the screen resolution or the display device > parameters. Look at this. > > > > > > I expected the ASV to show me always!!! a rectangle which is 10cm wide > and 10 cm high and which starts exactly 5cm (on x-axis and y-axis) > from the (0,0) point of the SVG coordinate system. These sound like fairly odd requirements; I generally want my charts to take up the same *percentage* of the screen area, no matter how large or small the display is -- from a 12" laptop to a 20" desktop. I've determined the information I want to show, and I want to the user to see all of that information as it was intended. SVG excels at this, by using percentage width and heights or an appropriate viewBox. However, if indeed you are graphing something where you are displaying actual units of length, and need to have perfectly calibrated real-world units on the monitor, this is not possible to do automatically. Not for the ASV plugin, not for IE or any other browser, not even for a Java or native C app on most OS platforms. If you want true calibration, you'll need to provide the user with a tool to indicate (by clicking and dragging next to a ruler on their monitor?) what "10cm" really represents on their display. You can then calculate a px to mm conversion factor, and store that on the server for each user. Think of the palm digitizer calibration tool as an example. Thanks, --kirby - To unsubscribe send a message to: [EMAIL PROTECTED] -or- visit http://groups.yahoo.com/group/svg-developers and click "edit my membership" Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/svg-developers/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
[svg-developers] Re: real world unit vs screen resolution or display device
Barend, thanx for Your answer but I'm still surprized. Maybe I'm stupid or limited but look: If I use Visual Basic (I know, it's awful) I've got a Screen object there. Screen object gives me an attribute TwipsPerPixel for both vertical and horizontal directions. And it is exactly what I thought. It gives me a real ratio to go from device units to real world units. And I'm pretty surre for the C (C++) there must be API to get these ratio. Of course it is applicable for MS Windows (what I'm interested in actually). But nonetheless thank you for the explanation that Adobe SVG doesn't support it. It's good to know. Another question is: why is this real world unit posibility included in the standard if it is not supported at all. I thing it is natural that if I write 100mm, I see 100mm and not anything else. By the way. If I say what am I suposed to get? 100 pixels wide rect? How the Adobe calculates the size in pixels of the rect? regards Maciek --- In svg-developers@yahoogroups.com, Barend Köbben <[EMAIL PROTECTED]> wrote: > I am sorry to see that you got surprised by it, but please note that it's NOT an SVG issue, it's true for any screen-drawing... > > If you state that "mm (...) must be displayed exactly so on the > screen - independent of the screen resolution or the display device", you are asking an impossible thing. > > It might calculate how many pixels theoreticaaly would go into a mm if the screen is supposed to have a 72 dpi resolution, but how is the application to know how big a pixel REALLY is on your screen...? It might be able (depending on the OS and so on) to find out to what resolution your graphics card is set and maybe even how large the screen is "supposed" to be, but as soon as somebody fiddles for example with the screen controls and changes the horizontal or vertical sync, or connects another type of monitor, the size of you rectangle would change! > > > > Barend Köbben > International Institute for Geo-information Sciences and Earth Observation (ITC) > PO Box 6, 7500AA Enschede (The Netherlands) > ph: +31 (0)53 4874253; fax: +31 (0)53 4874335 > _ > - To unsubscribe send a message to: [EMAIL PROTECTED] -or- visit http://groups.yahoo.com/group/svg-developers and click "edit my membership" Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/svg-developers/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
[svg-developers] Re: real world unit vs screen resolution or display device
Guys, thank You all for the answers. These are very very bad news for us, but that's a life. Thanx a lot once again cheers Maciek --- In svg-developers@yahoogroups.com, "Doug Schepers" <[EMAIL PROTECTED]> wrote: > Hi, Maciek- > > As Barrend correctly points out, your assumption is not quite correct, and > is possibly dangerous. Kirby's suggestion is exactly what I would have > suggested: give the user a settings dialog to allow them to calibrate the > measurements for themselves; if they really need the precision, they may > have to break out a physical ruler and measure the calibration bar manually. > I would also make sure that the data is stored for one screen only, and not > the user; if they log on with a different computer, the settings have almost > certainly changed; heck, even the same screen can change. But you can only > do so much. If they are depending upon precision, it is up to the user to > make sure that the screen is correctly calibrated. In fact, I might suggest > that if possible, the user could even change the screen to match the output. > > As for why these units are included, I would suggest 2 reasons: > 1) For properly calibrated screens, they are correct; > 2) For print measurements, they are often correct (depending on the viewBox, > CTM, zoom level, printer driver, etc.) > > Regards- > Doug > > doug . schepers @ vectoreal.com > www.vectoreal.com ...for scalable solutions. > > > Barend Köbben wrote: > | > | > | Hi Maciek > | > | > (...) TwipsPerPixel for both vertical and horizontal directions. > | > And it is exactly what I thought. > | > | Hmm, I don't think so. It probably is on most systems quite > | close to "what you thought". But even VB on an MS system > | cannot truly garantuee this in any way. If I want i can > | distort the image on my monitor by twiddling the knobs so > | that it only takes up half the actual screen. Your TWIPS VBA > | system would not be able to know this... > | > | And I think that eg. using the same code on a PocketPC > | platform (PDA), you might get quite big differences. > | > | > Another question is: why is this real world unit posibility > | included > | > in the standard if it is not supported at all. > | > | That is a very valid question. I cannot see inside the ASV > | code Adobe uses, but I think it does it 'sort of' by > | estimating what 100 mm might in pixels be. Just as the twips > | measurement in VBA only 'guestimates'... > | > | We're having this same discussion often with rprogrammers who > | want to put a scle fraction on screen maps, saying the map is > | eg. 1:10,000. As a cartographer I always advise them to use > | 'relative' sclaes only, the best one just showing a sclae bar > | that is scaled along with your map and basically says, 'look > | this is what I consider to be xx km on the map': > | > | 0 20km > | |-| > | > | (only correct when using a monotype font ;-) > | - To unsubscribe send a message to: [EMAIL PROTECTED] -or- visit http://groups.yahoo.com/group/svg-developers and click "edit my membership" Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/svg-developers/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
RE: [svg-developers] Re: real world unit vs screen resolution or display device
Hi Maciek > (...) TwipsPerPixel for both vertical and horizontal directions. > And it is exactly what I thought. Hmm, I don't think so. It probably is on most systems quite close to "what you thought". But even VB on an MS system cannot truly garantuee this in any way. If I want i can distort the image on my monitor by twiddling the knobs so that it only takes up half the actual screen. Your TWIPS VBA system would not be able to know this... And I think that eg. using the same code on a PocketPC platform (PDA), you might get quite big differences. > Another question is: why is this real world unit posibility included > in the standard if it is not supported at all. That is a very valid question. I cannot see inside the ASV code Adobe uses, but I think it does it 'sort of' by estimating what 100 mm might in pixels be. Just as the twips measurement in VBA only 'guestimates'... We're having this same discussion often with rprogrammers who want to put a scle fraction on screen maps, saying the map is eg. 1:10,000. As a cartographer I always advise them to use 'relative' sclaes only, the best one just showing a sclae bar that is scaled along with your map and basically says, 'look this is what I consider to be xx km on the map': 0 20km |-| (only correct when using a monotype font ;-) Barend Köbben International Institute for Geo-information Sciences and Earth Observation (ITC) PO Box 6, 7500AA Enschede (The Netherlands) ph: +31-(0)534874253; fax: +31-(0)534874335 _ [Non-text portions of this message have been removed] - To unsubscribe send a message to: [EMAIL PROTECTED] -or- visit http://groups.yahoo.com/group/svg-developers and click "edit my membership" Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/svg-developers/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
RE: [svg-developers] Re: real world unit vs screen resolution or display device
Hi, Maciek- As Barrend correctly points out, your assumption is not quite correct, and is possibly dangerous. Kirby's suggestion is exactly what I would have suggested: give the user a settings dialog to allow them to calibrate the measurements for themselves; if they really need the precision, they may have to break out a physical ruler and measure the calibration bar manually. I would also make sure that the data is stored for one screen only, and not the user; if they log on with a different computer, the settings have almost certainly changed; heck, even the same screen can change. But you can only do so much. If they are depending upon precision, it is up to the user to make sure that the screen is correctly calibrated. In fact, I might suggest that if possible, the user could even change the screen to match the output. As for why these units are included, I would suggest 2 reasons: 1) For properly calibrated screens, they are correct; 2) For print measurements, they are often correct (depending on the viewBox, CTM, zoom level, printer driver, etc.) Regards- Doug doug . schepers @ vectoreal.com www.vectoreal.com ...for scalable solutions. Barend Köbben wrote: | | | Hi Maciek | | > (...) TwipsPerPixel for both vertical and horizontal directions. | > And it is exactly what I thought. | | Hmm, I don't think so. It probably is on most systems quite | close to "what you thought". But even VB on an MS system | cannot truly garantuee this in any way. If I want i can | distort the image on my monitor by twiddling the knobs so | that it only takes up half the actual screen. Your TWIPS VBA | system would not be able to know this... | | And I think that eg. using the same code on a PocketPC | platform (PDA), you might get quite big differences. | | > Another question is: why is this real world unit posibility | included | > in the standard if it is not supported at all. | | That is a very valid question. I cannot see inside the ASV | code Adobe uses, but I think it does it 'sort of' by | estimating what 100 mm might in pixels be. Just as the twips | measurement in VBA only 'guestimates'... | | We're having this same discussion often with rprogrammers who | want to put a scle fraction on screen maps, saying the map is | eg. 1:10,000. As a cartographer I always advise them to use | 'relative' sclaes only, the best one just showing a sclae bar | that is scaled along with your map and basically says, 'look | this is what I consider to be xx km on the map': | | 0 20km | |-| | | (only correct when using a monotype font ;-) | - To unsubscribe send a message to: [EMAIL PROTECTED] -or- visit http://groups.yahoo.com/group/svg-developers and click "edit my membership" Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/svg-developers/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
