Musemath Home Page  

Monitor Resolution

March 31, 2010

This is now obsolete, the circle distortion on non-native resolution LCD monitors is no longer a problem. The resolution independence from Apple apparently all had to do with bitmaps zooming in on mobile devices. I had noticed the bitmap emphasis for various software, but never understood the move away from vector zooming. I do now.

I expect to delete this page shortly.

 

May 30, 2008

Keeping circles round at various resolutions (for Geometry Demonstrations). I have been looking at this problem for some time now - some visual demonstrations (flash and applets) show distortion (especially a problem with circles) because of higher resolution monitors with varying aspect ratios, wide screens, older video cards, poorly designed video cards and the like. I have not yet found a good solution - hopefully one will turn up. Here's a revised summary of what has been done so far.

March 11 , 2008

I would like to think now that now that Macintosh has apparently stopped distorting circles, (Resolution Independence, Leopard 10.5 See: http://blog.wired.com/gadgets/2006/12/apple_patent_pe.html
http://www.bltt.org/accessibility/resolution_independence.htm) Microsoft will be doing whatever it takes to do the same, fairly quickly; no guarantee, they have their own Expression Suite they may want to emphasize, but there are 3rd party programs advertising Resolution Independence which may be gobbled up, MS has some ideas of it's own, i.e. http://research.microsoft.com/~cloop/LoopBlinn05.pdf
and I suppose they could surprise me and lease the Mac Technology.

There will still be older software and computers, possibly for some time, and if I find some time, I will follow up: The plan: 1. A range of aspect ratios. 2. Actionscript to smooth out circles based on some average of each range (presumably a conditional function with inequalities covering the probable aspect ratios). 3. Apply to applicable parts of "Circular Functions 5" animation as that seems to be the one most likely to address the issues seen so far.

Again, there is a fair chance this will take awhile, I consider working on the math to be more important at this time.

 


 

Different systems - Detection of Possible Variables on Movie Dimensions - What I am thinking of is to resize the animation or parts of it based on the viewers settings. These are the readings from your computer as derived by the Flash Capabilities Object.


Number of pixels high (screen height in pixels)
Number of pixels wide (screen width in pixels)
Pixels are made up of even smaller components, the "color dots" on the monitor screen. This is the ratio of color dots in the height and width of each pixel - Flash 8 ALWAYS returns 1.
Here a dot means a pixel. This is different than in printing. BUT, this Flash 8 version ALWAYS reads 72.

 

You should be able to change them on the XP Operating System as follows:

Warning - On my system - XP - all my desktop icons get reorganized.

Desktop>RightmouseClick>Properties>Settings>Screen Resolution slider (try a change, then)> Advanced> DPI Setting Scrollbox (try a change - then - restart) > Refresh this page. Compare the readouts above.

 

 

With a bit of coding you can get what the Operating System says is the actual resolution:

 

 


These are from some of my first research on the problem:

After a bit of research, the problem is beginning to come into focus.
Check out the following:

http://alarmingdevelopment.org/?p=66

"Brother, can you spare a pixel?

there is one dirty little secret no one is talking about: Flash is resolution-dependent.

Flash, like HTML, still lives in the world of 72 pixels per inch. My screens are all at least 110 ppi, 50% denser.

Update: I have verified, in Apollo, that it is straightforward to scale the entire UI. And flash.system.Capabilities.screenDPI purports to tell you the dpi of the screen, so you could automatically scale. But it always returns 72, so you can’t...

And:

http://webkit.org/blog/55/high-dpi-web-sites/

"Now this may not be a huge problem yet, but as displays cram more and more pixels into the same amount of space, if a Web browser (or any other application for that matter) naively continues to say that one pixel according to the app’s concept of pixels is the same as one pixel on the screen, then eventually you have text and images so small that they’re impossible to view easily."

And:

http://channel9.msdn.com/ShowPost.aspx?PostID=136843

Windows Presentation Foundation videos (These are Microsoft Gurus touting their own product)

"Plus I think WPF is going to work better with the high-dpi displays that will come out in the future. Flash is vectors, but there's a lot of references to pixels, and Flash apps may not scale as gracefully on high-dpi displays unless the developer jumps through a lot of hoops to check the dpi & resolutions. Scaling and resizing apps under WPF (and having them still work) seems like it's easier than Flash."

 

 

March 10 , 2008

The following animation shows a conditional change of dimensions on the variable "e", the aspect ratio. At least some of it seems to be working.

resolution_mar10_4.html


CAVEAT- In order to test this it is necessary to change resolution repeatedly, this inconveniently repositions my PC desktop icons and I presume it will do the same for yours, please be prepared if working from a PC. I ended up using: Desktop>Rclick>ArrangeIconsBy>AutoArrange; now at least they stay in the same place even if it's the wrong place.

 

March 8 , 2008

This animation shows a conditional change of dimensions on a few common screen resolutions (Note that I had to make a few changes in the original animation as well as add in the code noted): scene_modified_2.html

 

 

***

NOTE: after going through the above and going to the Macintosh lab at school apparently they upgraded to OS 10.5 Leopard and now have "Resolution Independence" a patented technology. (A closer look indicated it was a late version 10.4__ Tiger OS.)

See: http://blog.wired.com/gadgets/2006/12/apple_patent_pe.html
http://www.bltt.org/accessibility/resolution_independence.htm

At any rate, All my circles looked round (pretty much, nothing objectionable) regardless of what resolution or browser I used so the new code doesn't do anything bad and is probably ok and should work for older macs and current PC's and, while I'm at it, some of the animation doesn't look as good at the higher resolution and could use a bit of upgrade so I might as well see what can be done. Microsoft will presumably be doing something equivalent reasonably soon but older computers and OS's of all kinds will probably be around for awhile given the current economic conditions.

 

http://en.wikipedia.org/wiki/Computer_display_standard - shows some of the various resolutions and pixel aspect ratios; historical, contemplated or in actual use. Their apparently unending variety (necessitated by technology innovation; pda's,laptops, cellphones, etc.) indicates I think, why any resizing put into code should be based on aspect ratio (pixel or otherwise). I should think a formula which resizes the .swf content (i.e. mc's) based on the difference between authoring and user aspect ratios (probably a range of ratios) might be a solution, though redesign to take this into account appears necessary to keep some mc's from running off the page.

I do not yet know if this is possible. To be worked on as time allows.

Table of computer display standards
Display resolution (pixels) Aspect ratio Decimal
720×350 (text) 72:35 2.057
640×200 (128k)
320×200 (64k)
160×200 (32k)
16:5
16:10
4:5
3.2
1.6
0.8
720×348 (250.5k) 60:29 2.069
640×350 (224k) 64:35 1.829
640×480 (307k) 4:3 1.333
320×200 (64k)
640×480 (307k)
16:10
4:3
1.6
1.333
1024×768 (786k) 4:3 1.333
640×480 (307k)
640×350 (224k)
320×200 (64k)
720×400 (text)
4:3
64:35
16:10
9:5
1.333
1.829
1.6
1.6
800×600 (480k) 4:3 1.333
1024×768 (786k)
640×480 (307k)
4:3
4:3
1.333
1.333
1152×864 (786k)
640×480 (307k)
4:3
4:3
1.333
1.333
320×240 (75k) 4:3 1.333
480×270 (126k) 16:9 1.778
240×160 (38k) 3:2 1.5
160×120 (19k) 4:3 1.333
1280×720 (922k)
1280×800 (1024k)
1440×900 (1296k)
16:9 or 16:10 1.778
_or_
1.6
1280×1024 (1310k) 5:4 1.25
1440×900 (1296k) 16:10 1.6
1680×1050 (1764k) 16:10 1.6
1600×1200 (1920k) 4:3 1.333
1920×1200 (2304k) 16:10 1.6
2048×1080 (2212k) 1.9 1.896
2048×1536 (3146k) 4:3 1.333
2560×1600 (4096k) 16:10 1.6
2560×2048 (5243k) 5:4 1.25
3200×2048 (6554k) 25:16 1.563
3200×2400 (7680k) 4:3 1.333
3840×2400 (9216k) 16:10 1.6
4096×1716 (7029k) 2.39 2.39
4096×3072 (12583k) 4:3 1.333
5120×3200 (16384k) 16:10 1.6
5120×4096 (20972k) 5:4 1.25
6400×4096 (26214k) 25:16 1.563
6400×4800 (30720k) 4:3 1.333
7680×4800 (36864k) 16:10 1.6

 

March 5 , 2008

What I'm thinking of right now is a conditional statement with or without a function in which the capabilities.resolution (x/y) is evaluated (if that is the right word) and if it's not the same aspect ratio that was used to create the movie, then the mc of the circle inside the movie is resized so it looks round.

The idea would be to resize based on aspect ratio (presumably ranges of different ratios), that way HD monitors could go up to7680x4800 or whatever but the resizing would still work.

The circles don't have to be absolutely perfect, they just have to look more like circles than ovals, the quicker the viewer jumps to the conclusion that a circle is being demonstrated the better. The viewer can also be educated to the values of some tolerance by slight imperfections - but slightly less imperfect than I'm seeing now.

I have no idea whether this is possible.

 

March 4 , 2008

Stage.scaleMode is tried below (unsuccessfully), the code is placed in the first frame of the movie - it dictates how the movie is sized relative to the player. But it's not really what I want - exactFit will reshape the mc in the animation so it can be dragged into the shape of a circle (If the winow is not maximized) but ideally, a first time viewer would go through the animations intuitively, quickly, without having to figure out unessessary proceedural steps, whether they're justified or not, etc. There would be as few distractions from content as possible.The other options; showAll. noBorder, and noScale did not do what I want either..

 

resolution_mar4_1_exactFit.html

 

Another idea: but just resizing the mc, not replacing the animation itself :

http://www.dynamicdrive.com/forums/showthread.php?t=27933

BLiZZaRD

"well, you could try something like the following:


var e=(screen.width/screen.height);
if (e=1.66){
window.location.replace("http://www.example.com/page2.htm");
else{
window.location.replace("http://www.example.com/");
}

and just add an if for each major aspect ratio (I think there are 6 or 7) umm.. 1.33, 1.37, 1.66, 1.78, 1.85 and 2.35 are the ones I know off the top of my head."

 

March 3 , 2008

The following can change the dimensions of the movie clip based on the user's screen resolution using Flash's Systems Capabilities Object. At least it changes on my own computer when I remember to refresh the animation. Everything is a bit exagerated here as a first trial.

1. Click on the link below

2. CAVEAT- In order to test these it is necessary to change resolution repeatedly, this inconveniently repositions my PC desktop icons and I presume it will do the same for yours.

Change your monitor resolution to 1024 or 800 or whatever is written in the script on the animation. (PC, XP - Desktop > Rclick> Properties> SettingsTab> ScreenResolutionSlider) (Mac had a monitor icon> SystemPropertiesPreferences? and more options)

3. Refresh the animation (View>Refresh) ( I didn't do this yesterday and thought the script was no good at all, so I don't know yet how these will work on other computers or the various browsers).

4. Verify if the image has changed.

5. I found the code at http://www.actionscript.org/forums/showthread.php3?t=130355by Noct, dated 03/07/2007

 

resolution_feb29_2_widthBlank.html

 

September 23, 2007

I have no answers yet, for the most part the animations look ok - on some monitors distortion caused by higher resolutions, different aspect ratios, wide screens with older video cards etc.cause the circles to look like ovals. The animations were created on a stage of 550:400 px, 1.375 aspect ratio - I will probably just put some kind of notes somewhere and let it go for now.

 

May 5, 2007

1. New animation flash_resolution.html This shows a bit of the issue with high resolution but does not show the distortion problem, that will take a bit more work.

2. High Display DPI (PPI) is and apparently will be a continuing problem - distortion and small print. Obviously, I am not alone. I can find no ready solutions. My site is still faring better than many others but resolution is expected to continue to increase and the problem has to be addressed if not immediately solved.

The immediate problem for me is that on one school campus the 15 inch LCD Monitor has a native resoltion of 1280 x 1024 but is set to 800 x 600 so the text is large enough for the students to see what they're typing. Unfortunately, this results in very poor quality image and for me, more importantly, the distortion of the Flash animation particularly noticeable on circles and squares.

Apparently, with an LCD monitor, changes in resolution should be in integral multiples (same idea as in string overtones - 1/2, 1/3, etc.) that is, you could change a 1600 x 1200 LCD to 800 x 600 and still have good quality, but a 1280 x 1024 would look best at 640 x 512 and there is no such setting in the OS.

I can reset the monitor to 1280 x 1024 and the Flash looks fine but then I need the XP magnifier (Start>AllPrograms>Accessories>Accessibility>Magnifier) and/or the Internet Explorer 7 browser's Zoom feature for most everything else. This is inconvenient to say the least.

I understand the Flash problem has to do with its fixed 72 DPI (PPI) in the source code which is still extant I believe on the new Adobe CSR Version just released though they seem to be paying some attention to the distortion problems, at least so far as raster images are concerned.

The problem may get much worse with the very high resolution displays due out in a year or so (200 DPI).

 

Back to top