By kulajw - 3/24/2017
Hi,
I wonder what is real difference between setting size or position of stimuli in pixel vs points in Inquisit.
First I claryfy my current understanding of thi issue.
According to Wikipedia Pixel: "it is the smallest controllable element of a picture represented on the screen" [Pixel]
"The number of pixels in an image is sometimes called the resolution". So number of pixels in one inch/cm/mm depends on screen resolution, the higer resolution the more pixels is placed in one inch/cm/mm.
Consequently stimuli of 100 pixels height might have different physical height (e.g. as measured in mm) depends on screen resolution - it might look different (in terms of height) on different computers.
According to Wikipedia Point is "the smallest unit of measure (...), 1⁄72 of an international inch (about 0.353 mm)" [Point]
Consequently stimuli of 100 points height should have the same physical height measured in mm (e.g 100 points x 0.353 mm = 35.3 mm) regardless of screen resolution. Stimulus height defined in points will look exactly the same on all screens independently of screen resolution.
My questions are:
Is this pixels and points notation work in this way in Inquisit? ( I can't imagine how one points can be independent from screen resolution)
How Inquisit can figure out how to properly calibrate one points as always e.g 0.353 mm?
What in a hypothetical situation when a point height (assume 0.353 mm) is smaller than one pixel height under actual screen resolution?
Thank in advance for all comments.
|
By Dave - 3/26/2017
+xHi, I wonder what is real difference between setting size or position of stimuli in pixel vs points in Inquisit. First I claryfy my current understanding of thi issue. According to Wikipedia Pixel: "it is the smallest controllable element of a picture represented on the screen" [ Pixel] "The number of pixels in an image is sometimes called the resolution". So number of pixels in one inch/cm/mm depends on screen resolution, the higer resolution the more pixels is placed in one inch/cm/mm. Consequently stimuli of 100 pixels height might have different physical height (e.g. as measured in mm) depends on screen resolution - it might look different (in terms of height) on different computers. According to Wikipedia Point is "the smallest unit of measure (...), 1⁄ 72 of an international inch (about 0.353 mm)" [ Point] Consequently stimuli of 100 points height should have the same physical height measured in mm (e.g 100 points x 0.353 mm = 35.3 mm) regardless of screen resolution. Stimulus height defined in points will look exactly the same on all screens independently of screen resolution. My questions are: Is this pixels and points notation work in this way in Inquisit? ( I can't imagine how one points can be independent from screen resolution) How Inquisit can figure out how to properly calibrate one points as always e.g 0.353 mm? What in a hypothetical situation when a point height (assume 0.353 mm) is smaller than one pixel height under actual screen resolution? Thank in advance for all comments.
> Is this pixels and points notation work in this way in Inquisit? ( I can't imagine how one points can be independent from screen resolution)
Yes and no.
> How Inquisit can figure out how to properly calibrate one points as always e.g 0.353 mm?
It can't. The operating system provides functions which give *approximate* results for the conversion from pixels to any fixed physical measurement unit (be it mm, cmm, inches, etc.). The results can be quite "off" in practice, however.
> What in a hypothetical situation when a point height (assume 0.353 mm) is smaller than one pixel height under actual screen resolution?
Then you will get an approximate result. No display in the world can display anything smaller than one pixel.
If your paradigm requires that stimuli have certain physical sizes and/or distances, some sort of calibration routine is required. See e.g. https://www.millisecond.com/forums/Topic14608.aspx
|
By seandr - 3/27/2017
Hi Kulajw,
If your experiment doesn't require precise physical sizes, then I'd recommend the following
1) Avoid both pixels and points and use percentages instead for all size and positions
2) Specify defaults->canvasaspectratio
Both of these steps will insure your visual presentations scale appropriately on displays of different sizes, shapes, and resolutions. This is especially important if you are running tests over the web.
-Sean
|
By kulajw - 3/27/2017
Thank you guys !
So now it is clear for me:
if I require precise physical sizes in my paradigm I need to implemen calibration routine.
If i only need to have similar scaling acros difren screens I can use percentages and canvasratio.
But still I am not sure, waht is a difrence beetwen defining stimuli size in pixels vs definig in points ? (is there any?)
Thank you for response.
kulajw
|
By Dave - 3/27/2017
+xThank you guys ! So now it is clear for me: if I require precise physical sizes in my paradigm I need to implemen calibration routine. If i only need to have similar scaling acros difren screens I can use percentages and canvasratio. But still I am not sure, waht is a difrence beetwen defining stimuli size in pixels vs definig in points ? (is there any?) Thank you for response. kulajw
> But still I am not sure, waht is a difrence beetwen defining stimuli size in pixels vs definig in points ? (is there any?)
A point is a defined physical measurement unit, just like a centimeter is. Pixels, however, do not have a fixed relationship to any physical measurement unit, i.e. while you can convert millimeters to centimeters or inches, etc., there is no universally true conversion from pixels to millimeters, etc. Meaning: The physical size of one pixel on display A can be wildly different from the physical size of one pixel on display B. This also means that in some cases / on some high-resolution displays one pixel will in fact be smaller than one point.
|
By seandr - 3/27/2017
The relationship between a point and a pixel is as follows:
PT = DPT * PX / 72
PT: Points
PX: Pixels
DPI: Dots (pixels) per inch of the monitor, printer, etc.
Windows knows the resolution of the monitor but not its physical size, so the DPI is always assumed to be 96. For cross platform compatibility, we also assume 96 DPI on Mac when calculating point size. Note that your actual monitor may pack more pixels into an inch than that.
-Sean
|
By kulajw - 3/29/2017
+xThe relationship between a point and a pixel is as follows: PT = DPT * PX / 72 PT: Points PX: Pixels DPI: Dots (pixels) per inch of the monitor, printer, etc. Windows knows the resolution of the monitor but not its physical size, so the DPI is always assumed to be 96. For cross platform compatibility, we also assume 96 DPI on Mac when calculating point size. Note that your actual monitor may pack more pixels into an inch than that. -Sean Thank you for explanation, now I am well educated in this topic ;)
So it seems to me that points are quit awkward to use.
Best
|
|