I would recommend using a profiler like Scout to see what the difference is.
Depending on your test case, Flex defaults to 24fps and does run a lot of code. -Alex On 4/30/16, 4:46 PM, "jude" <flexcapaci...@gmail.com> wrote: >In my comparisons the animation in a Flash movieclip is usually much more >fluid than animation in Flex yet I can't figure out why. Does Flex snap to >pixel values while animating? If so is this something that's part of the >Animate classes we can fix? The last time I looked into it we were using >Number not int but that might have been Flex 3. > >I found this short video ><http://www.paulirish.com/2012/why-moving-elements-with-translate-is-bette >r-than-posabs-topleft/> >on how using translate in HTML is better than animating the top / left >values. In the video the object that is animated using the top and left is >updated by stepping from pixel to pixel on the CPU while the object that >uses translate uses subpixel images on the GPU. If it's a case of simply >using GPU mode instead of auto, which IIUC auto almost always chooses CPU, >then are there any downsides to switching to GPU? Or is there something we >can change in the Flash Player to animate on subpixels? > >The alternative is that Flex invalidation architecture is causing some >issues but I have no proof. I'm guessing there might be syncing issues the >way the refresh rate on a monitor that is different than the framerate of >an animation causes issues. But the difference in Starling animations (all >GPU?) and movieclips from Flash (CPU) both seem much more smooth than Flex >animation. > >Another alternative is that Flex is trying to do too much. If you have a >container that has 50 elements in it and then you use animate the height >from 0 to 100 or the other way around, each element is actually getting >invalidated. Then, the question would be is there a way to turn off >invalidation while resizing and just do something like resize of a bitmap >snapshot. > >FYI Sorry if we had this discussion before but I can't find how to search >the mailing lists.
