I have written a Win32 C++ program.
I would like to duplicate the visual content of the left window every frame, so it appears at another location on the screen. (One rectangle on the left side of the screen, and an identical rectangle on the right side of the screen.)
This is for stereo rendering, where the left rectangle is seen in the user's left eye, and the right rectangle in the right eye.
The two rectangles must be "frame synced", meaning that they must show the same frame at all times. Or to put it another way, if the right eye is behind the left by "one draw call", then the viewer becomes ill.
NOTE: I have the source code to both windows, because I created them. (Both windows reside within the same exe at the moment.)
What I have tried:
BitBlt(), by drawing a rectangular portion of the desktop HDC into the right-side window. I am calling BitBlt() in the WM_PAINT message of the right-side window: This works, BUT the right-side window is 1-2 frames behind the left.
Potential Solutions:
In the left-window's WM_PAINT message, draw both windows with the same contents. Is this possible somehow?
Use one, single giant window to draw both rectangles. This would mean all drawing occurs in the same WM_PAINT call. However, I do not know how to copy part of a window to another part. I wish I could just tell Win32 to 'draw again', but this time start 1000 pixels to the right...
Do "something" with DirectX. (No idea what.)
Any help would be greatly appreciated!
I don’t know if you still need an answer, but I don’t think you can do it better than using the DWM Thumbnails API [1]. You basically do an association of a region on your window <-> a region of a foreign window, and then each time the windows are composed and presented on the screen, the specified rectangle of the foreign window will be painted on the specified rectangle of your window. DWM will do scaling if necessary and you can also specify a transparency level. This is the most efficient way to do it, as the compositor does all the heavy lifting, so it is at the lowest level you could go without writing kernel mode code (custom drivers etc). You can’t draw on top, and won’t “see” the foreign window data on your window if you BitBlt it for example, as the image is not actually drawn on your window, but overlaid on top when the final image of your window is composed and presented to the user by the window manager. But for most purposes, it works really well.
[1] https://learn.microsoft.com/en-us/windows/win32/dwm/thumbnail-ovw
Related
How to extract the image from this https://www.google.com/maps/#45.8118462,15.9725486,3a,75y/data=!3m7!1e2!3m5!1sAF1QipOH6lgU7bug2ndyW-9-Uq0kgKqcKDtnGei2N5Qo!2e10!6shttps:%2F%2Flh5.googleusercontent.com%2Fp%2FAF1QipOH6lgU7bug2ndyW-9-Uq0kgKqcKDtnGei2N5Qo%3Dw150-h150-k-no-p!7i3024!8i4032
(If the link disappears let me describe how to reproduce the question. Find any shop on Google Maps that has the "shop title image" appearing in the shop details on the left side when you click on that shop. Click on that image to expand it across the whole viewport.)
I found the <canvas> element that I guess contains the image. I tried to do .getContext('2d') on that canvas element, but I keep getting null.
If you are getting null when doing getContext("2d") it's because an other type of context was created already, in this case, a "webgl" one.
To convert that canvas to a new image, you'd normally call canvas.toBlob() (whatever the context type).
And if you need to crop that canvas content, you'd draw it on an other canvas.
But since they did not prevent the WebGL context to throw away its drawing buffer (by passing preserveDrawingBuffer in the getContext call), you'll only get a transparent image back from it.
Anyway none of these methods will retrieve the original image, but they will create a new image entirely (probably of lesser quality, and bigger in size). To retrieve the original image, check the network tab of your dev tools, or if you need to do it programmatically, inject a script that will spoof all fetch, XHR and HTMLImageElement objects in order to log their resource URL. But that becomes dirty.
I'm only using one scene object for my entire 2d sidescrolling platformer with two layers: There is a hudLayer for the controller, and the gameLayer for everything in the game.
When I control my character to walk into a door (sprite with a physicsBody that has a callback that takes me to the next level), the next level loads by these instructions:
remove all event listeners
stop all actions for both layers (the gameLayer typically has been doing a Follow action that follows the player, and is limited to the size of the backgroundSpriteNode)
remove all children from the gameLayer
set gameLayer position to Vec2(0,0)
load level 2 (a complete copy of what level 1 should look like... load the background sprite to gameLayer, playerSprite, add event listeners)
add event listeners back to the hudLayer
The only problem is that for whatever reason, the level only partially loads the way it should... For instance, coins appear in the spots they should, but platforms don't seem to. Neither that or the player, or other characters. Not sure really why these objects are specifically not loading in the correct positions, but maybe someone who has experienced this sort of problem before can help?
Most nodes (such as platforms) are appearing below the bottom of the screen when they should be fully shown at the bottom of the screen. Characters are appearing down there too when they should be appearing above the platforms.
P.S. After testing a bit more, I've come to realize that SOMETIMES some of the platforms are loading in the correct positions, but sometimes they aren't. Not sure why. A fix for this may be to load things one at a time with a delay between them. But I'd rather not have to put delays and just figure out what the deal is.
Figured it out: for whatever reason, I needed to add a delay in there. I think for whatever reason, by calling removeAllChildren(), something got messed up with the positions. It was my suspicion since SOME of the time, parts of the next level would load, but only some parts, and only sometimes.
Here's the code I added to run an action on the gameLayer once all children were removed, etc...
runAction(Sequence::createWithTwoActions(DelayTime::create(0.1), CallFunc::create( [&] ( ) {
loadLevel2(this);
})));
I have created a dot matrix slider on a canvas element (much like the sort you get in the stock exchange). Currently I have each letter laid out as an individual matrix and then, through a succession of loops I have these letter converted into one large matrix.
I then go through and draw this matrix column by column up to a maximum amount of columns. The matrix then gets redrawn every X milliseconds, offsetting the viewable area of the matrix by one each iteration until it eventually loops. The large matrix doesn't get redrawn every time.
My issue is that the animation doesn't seem smooth at lower redraw intervals - it jumps about. I've checked the frame rate and it seems fine but occasionally it jumps and I can't work out why!
The function with the settings is right at the bottom of the JSFiddle.
dotMatrix({
animationDelay: 1000,
canvasSelector: '#dot-matrix',
dotRadius: 2,
loop: true
});
Here are some steps you could do:
Prerender each char to an off-screen canvas in a solid color on transparent background. This canvas will be used as a sprite-sheet later. Drawing a rounded rectangle is a relative expensive operation especially when you need x number of it per char.
Set up a gradient for the colors and store that too in an off-screen canvas (by now we can see memory-caching is coming to the rescue..).
Every time you update use requestAnimationFrame instead of setInterval. The latter is not able to synchronize to monitor update. Scroll delay can be calculated using the number of frames elapsed as well as the time since last update (see rAF doc for details).
For each update clear, then "blit" the letter from the sprite-sheet canvas to main canvas. When all are blitted change composite mode to source-atop and blit the gradient canvas on top, and reset composite mode to source-over (don't use save/restore - they are expensive).
In the last step you could also use composite mode copy instead of clearing canvas and using source-over, as that will remove any previous existing pixels for the area you draw to.
Hope this gives you some inputs to improve the performance.
I currently have an html page that produces a line graph inside a canvas element. I have also implemented a feature so that when you click on a point in the graph, a table below the canvas displays the information for that graph point, and clears if you click somewhere that isn't a graph point on the canvas. However, there is currently no indicator for what graph point you have selected.
I would like to make it so that the dot you click on either gets bigger, or highlighted, or something. Unfortunately, I don't think there's any way to do that on the one canvas without needing to redraw the entire graph afterwards. If I drew a bigger dot on top of the original, I would then need to erase it and replace it with the smaller one, as well as redrawing the segment of the line graph that was covered by the larger dot.
I was looking into using a second canvas on top of the first with identical dimensions, using a z-index to control which was on top. However, I don't know how this would affect the click event. If two canvasses are on top of each other, is it possible for the click event to register the lower canvas? Does it only register the higher one? I guess I could just change the click event to be for whichever canvas is on top, but keep all the code for drawing on the canvas the same.
Any advice for how to solve this problem?
Solved a few of my own questions.
Firstly, the click event will only recognize whichever canvas is on top. However, I did like I said, and changed my click event to be for the top canvas, while leaving all instructions on the inside to be for the top canvas. Thank goodness I didn't use the 'this' keyword, or the change would have been much more annoying.
To make the graph points highlight, I first added a second canvas on top of my first one, placing them on top of each other by making their positions "absolute" and giving them a z-index of 0 and 1 respectively. Then, inside the if statement where the original click event recognized that a point had been clicked on, I told the top canvas to draw a larger dot on the same coordinates as the first dot (which I had saved in an array). First, I had it clear the top canvas though, so any other highlighted dots would no longer be highlighted. If the canvas was clicked on somewhere other than a dot, nothing was highlighted.
So, I'm pretty much a beginner in Flash and Actionscript (using AS3, as I said in the title), and I'm trying to make a basic escape the room game. I haven't gotten far, and right now that's because every time I test my game (or publish preview it) the graphics get this annoying outline. Here it is when tested: http://i305.photobucket.com/albums/nn228/chokingondrama/flash.png
Every outline corresponds to some object present in the game, most of which have an alpha component of 0 since they're on different sides of the room. This didn't happen before, but once I added the code that allowed the player to change their view with the arrow (each viewpoint/wall is a different frame) these appeared.
It's a little different when published to HTML, basically it just gives each image a white background: http://i305.photobucket.com/albums/nn228/chokingondrama/html.png
Also, it would be nice if somebody could give me advice on how to make sure importing to flash won't result in lower quality.
Thanks in advance. If needed, I'll post any part of the code.
Some tips:
Don't set alpha to 0, instead use the visible property, setting movieclip.visible = false will make it a lot more efficient.
As for the importing and quality, after you import to stage or library, bring up the library (ctrl + l), and right click on the file you imported, go to properties. If it's an image, set compression to lossless, and allow smoothing.
For audio, go to file-> publish settings, and change audio stream and audio event (whichever you might use) to 128kbps.
As for your main question, I need more info, if you want you can post your source. It might be because of how you are placing your graphics on the stage.
For each of your MovieClips in question:
Try disabling button mode and see if the rectangles go away.
movieClipName.buttonMode = false;
If that doesn't help, or you really want button mode, try setting
movieClipName.tabEnabled = false;
There's a chance that since you added keyboard interaction each of your MovieClips are now expecting to be selected by the user when they press the tab key, much like any normal web form.
tabEnabled in the docs
You could also try
movieClipName.focusRect = false;
focusRect in the docs