- From: Miguel Casas-Sanchez via GitHub <sysbot+gh@w3.org>
- Date: Fri, 09 Mar 2018 14:36:09 +0000
- To: public-media-capture-logs@w3.org
@developer239 I think the key here is that the source is a 'desktop', which makes up for a special type of VideoFrames passing along the `MediaStreamTrack`, namely ones backed by Textures, i.e. they stay inside the GPU driver, so when the `Context2D` creates a `Blob`, it's probably using faster GPU resources, whereas `grabFrame()` will use Skia to pull the data out of the Texture, which is slower. Also consider that the `blob` you get might be lazily evaluated, so at the time you get to `//do some magic` you might not quite have the data, but if you were to, e.g. save the `blob`, then you'll hit a delay. That'd be cool to double check. Otherwise this seems to be more a Chrome issue than a Spec issue. So, closing it. Thanks for your feedback! -- GitHub Notification of comment by yellowdoge Please view or discuss this issue at https://github.com/w3c/mediacapture-image/issues/197#issuecomment-371829267 using your GitHub account
Received on Friday, 9 March 2018 14:36:11 UTC