Ok, what about attaching a scanner that you have already calibrated?
Should be easy to up the contrast and then scan for edges.
Another option would be to grab a pair of cheap electronic calipers and
hacking them onto a USB device to read the output values.
Alternatively: Hack a USB mouse and rig the H and V rollers to a table.
Then attach a 'pointer' to these rollers and read the H and V values as
the item is 'drawn' and translate them into the appropriate size
(knowing
the dimensions of the table and calibrating accordingly). Hmmm, maybe
a graphics tablet would do, but it'd get damaged easily.
It would avoid the monitor getting damaged...!
I reckon the scanner would be the more accurate option and you can then
store that for each client.
I wouldn't count on the pixels being square by the way.
Cheers,
Luis.
On 30 Jan 2007, at 18:52, Walton Sumner wrote:
I have a survey instrument which can require interface elements to
have
exact physical lengths, especially a "visual analog scale" that is
traditionally 10 cm. Not knowing what the screen settings will be
on client
machines, I have for the time being put a button on the first
screen of the
survey instrument that allows a survey administrator to drag it's
right edge
until it is 2 cm long. Then anything else in the stack that needs an
absolute size can be rescaled accordingly. You can put some
constraints on
the assertions, like expecting no more than 100 pixels/2cm and no
less than
50 (one real value is 83 on this iMac). So far I've assumed that
pixels are
squares, hope that's safe.
As long as the survey instrument resides on a given machine, this is a
one-time, set-and-forget event. The rev program reads instructions
(including VAS lengths) in text files to implement new surveys, so the
client holding the pixels/cm data should not be replaced very often.
It is not at all elegant, but I expect that the obvious variations
on this
theme are pretty reliable (eg, "How long is this line?", "Click 3
cm from
the end of this line"), and perhaps less trouble for the user than
asking
for several unfamiliar measurements. Still, if there is a
transparent way,
I'd like to use it!
To use the info that Luis found, it seems like you would have to
assert a
pagerect in a print command, then get the printscale used in
printing the
page, then calculate the absolute size of the card on screen. Has
anyone
tried that?
Seems like the simplest thing to do should be to read the diagonal
of used
screen space from the hardware, but I do not see a rev command to
do that.
Maybe there is a shell command to read the diagonal on some
systems, but I
have not found it.
Walton Sumner
Found this under the 'print' entry in the Rev Dictionary:
'The pageRect is the rectangle into which the card is printed, and
consists of four integers separated by commas: the left, top,
right, and
bottom edges of the printed card, in points. (There are 72 points
to an
inch.) The card is scaled to fit the specified pageRect. If you don't
specify a pageRect, the card's size depends on the printScale
property.'
Might help in calibrating the size.
Cheers,
Luis.
Luis wrote:
Hmmm, ok. What you could then do is open a dialogue box asking for
dimensions (17 inch, 19 inch, etc), resolution, type (CRT, LCD or
Plasma) and dpi. If the dpi is unknown by the user you could try to
default to the most common dpi for that 'type' of monitor.
Cheers,
Luis.
Mark Schonewille wrote:
Hi Luis,
Yup, my main monitor is of the brand "unknown".
Mark
--
Economy-x-Talk
Consultancy and Software Engineering
http://economy-x-talk.com
http://www.salery.biz
Get your store on-line within minutes with Salery Web Store
software.
Download at http://www.salery.biz
Op 19-jan-2007, om 17:00 heeft Luis het volgende geschreven:
Hiya,
If you can determine the monitor type you could probe a
database of
configurations/resolution/dpi settings etc and then calculate the
image size (using its coordinates) based on this information.
I haven't looked into this but I reckon OSX and WXP have these
in the
system somewhere (so they can 'plug and play').
Cheers,
Luis.
_______________________________________________
use-revolution mailing list
[email protected]
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
_______________________________________________
use-revolution mailing list
[email protected]
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
