I'm currently working on 2 Android projects and seems like I have a problem with understanding how to handle the density of resources. I have a HTC One M8, which has ~440 dpi, therefore I'm using XXHDPI. I always use XXHDPI resources and wrap content for the elements. On one of the projects, all the resources are png formats, again using XXHDPI and wrap content and when running the app, it looks great.
On the other hand, the resources from the other project are .9.pngs. Whenever I use XXHDPI and wrap content, in the preview, the design of the elements is altered, the pictures are really shrank.
However, if I use a HDPI, the size of the picture will look good, but the pixel's density will be crap. As far as I know, I should never hardcore the width and the height of the picture, but get the picture with the specific density and just use wrap content.
Am I missing something or there is a problem with the resources?
Related
Its very clear for me what sizes should I use for the different versions of ic_launcher.png inside drawable-hdpi, drawable-ldpi, drawable-mdpi, etc.
What I don't understand is what sizes should I use for a picture (a png) that is simply used in one Activity as an ImageView.
I have this image in high resolution
In a normal pocket device this image should be like 90% width
In a tablet or larger devices, maybe 90% is too big
So, the question is:
Should I create several versions of this an_image-file.png for each folder in drawable? or just use a single version of it and adjust its width via XML? I'm asking because I'm worried about using the best practice in this subject.
There is no clear answer. Only way you can solve this is by trial and error. Sizes are defined for icons only for other image views you will have to try it on 5" smartphone 7" nexus 2 tab and 10" tablet.
At least for these three resolution you will have to create images, for other devices XML layouts will automatically adjust.
Everytime, when I supply resource to drawable folder, here is what I did
drawable-xhdpi (96x96 px)
drawable-hdpi (72x72 px)
drawable-mdpi (48x48 px)
drawable-ldpi (36x36 px)
Most of the time, I merely use GIMP to perform dummy scale down from the largest size image from drawable-xhdpi. I do not perform any further pixel editing on the scaled down image.
Recently, I realize, if I only supply 1 highest resolution image, Android system will internally perform image size scale down.
drawable-xhdpi (96x96 px)
drawable-hdpi (empty)
drawable-mdpi (empty)
drawable-ldpi (empty)
I tested on 2 devices. It works for me. I was wondering is this a good technique to avoid cumbersome work to supply so many different size images? Any side effect on this technique?
I was wondering is this a good technique to avoid cumbersome work to
supply so many different size images? Any side effect on this
technique?
I imagine there are (at least) two:
This may not always work with 9-patches, especially if the stretchable area is defined by one or more single pixels. For example, if you would provide such a 9patch as xhdpi drawable, then on an mdpi device that single pixel will effectively be 'halved' and thus either disappear completely or blended with the surrounding transparent pixels. The latter is generally true for any upscale/downscale operation, so it's quite likely that your 9patch will not appear as intended.
Larger images will simply take up more space in memory. Especially on low(er)-end devices, with a limited amount of internal memory and a reasonably strict heap space limit, you're likely to run out of memory fast when loading images way larger than required for displaying.
Regarding your comment: it's fairly easy to come up with a sample 9patch that will not visually look as intended across all different screen densities if you supply it only as xhdpi resource. Consider the following 9patch:
Enlarged snapshot for the sake of visibility:
Obviously, the idea is that when this 9patch gets stretched, the result is a 1-pixel thick horizontal blue line. Now, drop this 9patch in just the xhdpi folder and compare the results on xhdpi vs. mdpi:
Clearly, the mdpi device has scaled down the 9patch from the xhdpi folder and the result doesn't look like intended.
Anyways, my point is that in a lot of cases not supplying 9patches for every density bucket may end up looking fine. Just be aware of the fact that there are definitely scenarios out there for which it may not give the desired result. Also, take into consideration the memory argument.
The technique you are using will work on fewer number of devices..
Because in market there are a lot of phones with different resolutions and different sizes.
In some low resolution devices whose screen size in smaller,the UI will not look proper...
Now take an example that you are using a background image for splash screen whose resolution is 720*1280(xhdpi) and you have put it in Xhdpi drawable folder than when you view this drawable in low end device like Samsung euopa or HTC wildfire the image would look squeezed type.
there would be a lot other situation where you would be required to use different set of images....
If the technique mentioned by you would work in all phones then why would had Android developed different folder for different set of images.... :P
For quite a long time, I have been declaring resource folders for 4 different densities:
drawable-ldpi
drawable-mdpi
drawable-hdpi
drawable-xhdpi
In the layout's XML, I have been using fixed widths (while still density-independent) such as 128dp for those graphics.
However, when more and more large-screen phones, and especially tablets, were introduced, that approach did not work anymore. Although you provide density-independent resources this way, the layout will not look good on large screens.
This is why I think I need to add Dimension resources that depend on the screen size, for use in the XML layouts, e.g.:
values
values-w600dp
values-w720dp
values-w1024dp
But does that mean that I should drop supporting those 4 density containers? Or do I need to provide 16 resource folders, i.e. one for every combination of density and size?
I can't find any good help in the Android documentation as to this topic.
drawables and layouts are different. To answer your question, should you stop support those densities. Yes, but you should still support xdpi and hdpi. Romain Guy recently said that modern devices like the Nexus 7 (at a tvpi) can scale the assets properly enough that mdpi isn't really needed. And nobody uses ldpi anymore. Last I looked is was less than 2% of the market.
About layouts. A Nexus 7 (1280x800 tvdpi) would use something from the values-w1024dp but still get assets from the drawable-hdpi folder. Those two aren't mutually exclusive. Something like a S3 would also pull from the values-w1024dp but use drawable-xdpi. You only need to provide an alternative layout if your use-case calls for it.
So do you need 16 different things? No. You do need xdpi & hdpi (if not mdpi). You may want to include alternative layouts for different sizes. You can be as specific as you want or as generic. Unless you're doing a hybrid app for both phone & tablet (7 & 10 in) you probably don't need a lot of xxxx-sizexxx folders.
In the layout's XML, I have been using fixed widths (while still density-independent) such as 128dp for those graphics.
This is probably a source of your issues. Your layouts should be as fluid as possible using wrap_content and match_parent. Fixed sizes should be reserved for padding around the sides and image where you know the size ahead of time. If you do this, your layout should look decent at any size from a small 320 x 200 to a GTV size.
1) Regarding dimensions in your layouts (values/dimens.xml):
values values-w600dp values-w720dp values-w1024dp
But does that mean that I should drop supporting those 4 density
containers? Or do I need to provide 16 resource folders, i.e. one for
every combination of density and size?
No, you don't need to provide different dimensions per dpi bucket (hdpi/xhdpi), because the dimensions are already being scaled up or down based on the device (if you're using dp instead of px). So for dimensions, you only need to provide values for devices with different sizes (hence the name, values-smallest possible width-600-dp). Because you don't want 16dp padding on a phone AND 16dp on a 10" tablet as well. You'd want 64dp instead. And no, it doesn't matter what density the device has. It still needs to have the same padding on the respective device width. So for dimens, you only need to think about the device's actual physical dimensions.
2) Regarding drawables scaling for different resolutions (drawables/xdpi):
The system scales them appropriately for the device. You don't need to worry about this. Also, you don't need to add any other buckets here. Just use mdpi/hdpi/xhdpi and maybe xxhdpi because many new devices are going to use the new density in the future.
Conclusion: there are 2 different UI building components that vary according to 2 different rules: drawables based on screen density and dimensions based on screen size. Do not mistake one for the other and think you need tens of buckets in the values folder, because that's not only wrong, it's just mind boggling.
Last two questions stayed unanswered, I hope "third one's the charm" works :)
I want application that is HDPI use those drawables in folder "drawable-xhdpi" and LDPI devices those in "drawable-mdpi". So I don't have to duplicate same images. Is this possible?
Yes, Android scales drawables, selecting the drawable that will produce the best result. But not all scaling combinations work that well. Downscaling to MDPI can produce noticeably poor results. XHDPI to HDPI generally produces pretty good results. But given that the overwhelming majority of devices these days are HDPI, it's probably worthwhile to have HDPI resources. HDPI to XHDPI scaling is not terrible. Some 1280x720 devices uses XHDPI resource. These generally look ok. Google TV uses XHDPI as well. On a huge screen, scaling errors are visible.
LDPI devices are -- for all practical purposes -- non-existent now.
True that generating scaled resources is an enormous PITA. But, in my opinion, you really need to do MDPI and HDPI. And once you've got that unpleasant workflow down, it doesn't require a lot of extra work to generate XHDPI. For what it's worth, if your artwork is in vector format, the vast majority of resources don't need tweaking for specific resolutions. There doesn't seem to be any compelling need to pixel-align edges of square objects, for example. Usually it's only interior features in complex icons and artwork that benefit from some amount of pixel-pushing.
It's pretty clear that XHDPI is going to be come more and more common. Trust me, you don't want to go back and re-do all your artwork in XHDPI after the fact. My advice: if you intend to still be shipping your app a year from now, suck it up, and do the nasty as you go, while it's still relatively easy.
In my experience, if you place an image in just some of the drawable folders, but not all, the OS will choose the "best" image for the device that you are using, even if it is not exactly the perfect fit. For example, if you put an image called arrow.png ONLY in the drawable-hdpi folder, all devices will use that image for the arrow drawable, and scale it down or up appropriately, stretching or shrinking the image.
That being said, you should be able to accomplish what you want by simply putting your image in the right folders and allowing the devices to choose the correct one. For example, if you were trying to accomplish your task with only one image, arrow.png:
drawable-xhdpi/ -> arrow.png
drawable-hdpi/ -> empty
drawable-mdpi/ -> arrow.png
drawable-ldpi/ -> empty
drawable/ -> empty
If you use the app on an ldpi device, the device will use the mdpi image, as you wanted, because it is closest to the correct resolution. On the hdpi phone, it will use the xhdpi image, because it is again the closest to the correct resolution.
Yes. Android automatically downscales/upscales resources that it cannot find. If you were to only put resources in the XHDPI folder it would just work, Android would take care of resizing them to work in all the other densities.
We should add this code in MainActivity class and copy image to dhpi folder with ic_launcher, ic_launcher2, ic_launcher3 and ic_1 is image name, we can change size: width and height
int array_image[]={R.drawable.ic_launcher,
R.drawable.ic_launcher2,
R.drawable.ic_launcher3,
R.drawable.ic_1
}
I have a game which displays a full screen image in the background. At the moment I have one image size (1280x800). This works well on large resolutions but on smaller screens the shrinking somewhat degrades the image. You can see jagged edges and it is noticeably worse than what you could achieve using photoshop software.
I have different image sizes, but I am unsure how to utilize them. I know there are different dpi folders, but you can have resolutions of 480x320 and 1280x768 with the same dpi so I don't think these can be of use here.
I believe you can have different layout files for different screen sizes, but the image is not drawn using xml (and in fact would not be possible for my game).
I can only think that I must create a different file name for each size. Then when choosing which image to use I could take the screen dimensions and select the correct one? I am struggling to see how I can make an image look good on both 240x320 and 1280x800 resolutions.
All of the resource qualifiers in the framework can be applied to drawables, not just the dpi designators. In other words, you could create folders like this to segment your images as well:
drawable-ldpi
drawable-xlarge
drawable-normal
drawable-sw480dp
drawable-sw720dp
Even examples like these work...
drawable-v10
drawable-land
And so on...
You can create as many or as few different qualified directories for your image assets as you think necessary to preserve the quality. The Supporting Multiple Screens article in the SDK docs helps describe most of the qualifiers that best fit scaling image assets.
HTH!