This interview is quite shallow, it doesn't really reveal anything, does it? It's general considerations and strong emphasis of the non-importance of SP1, all adding up to a flat feeling having watched the interview. Charles does give Mike Nash some reason
to sweat with a frank question. That was the highlight, heh. On the other hand, maybe there isn't really anything to tell, except to expect more compatibility and performance, which is great in and of itself.
I assume there is a reason why you have to explicitly say you want your code parallelized. It's still an imperative world with islands of declarativeness. But I assume that in the future the compiler or CLR will auto-parallelize some code, where it can determine
the safety of it.
But the team has its priorities right. It first creates the foundation for parallelism. Then later it can start to think about (or someone else can) the cases that are safe to parallelize - and where it makes sense to do so.
Maybe, with all the threads running on the machine, it wouldn't make sense to parallize some tasks, even though it would be possible to do so.
Just a thought...
One thing is for sure. The C#/CLR team has a great understanding of how to maximize impact and value whilst keeping disruption minimal.
One thing I didn't pick up from the video - is the Parallel Task Lib built on top of the existing Thread Pool stuff?
It's a nice idea to be able to refactor an application from a clients point of view into multiple tiers and platforms.
Is there some set of shared concepts for programming against both Silverlight and the browser? I'm not sure it makes that much sense, but that would make it seamless for code to run in either environment without needing to decide which. On the other hand, for
it to make sense, it would have to offer some benefit in terms of flexible controls and effects that are not easily done in the pure browser environment alone.
One possible example of that may be the JS/DOM 3D engine. On the other hand, I can't believe anything non-trivial can be animated - generated perhaps, although it could heavily influence scrolling and reflow speed.
It's cool though.
There are also several 3D engines for Flash. I've toyed with one myself, but it used to be that they were quite slow and not really usable for anything non trivial. The faster runtime in Flash 9 may be improving that but probably the software rendering is still
the bottleneck. For static rendering, it can be quite useful, though. Flash also has the ability to render to bitmaps, so you can do raytracing if you want to. That is, you *could*, but at a framerate not worth mentioning.
Can hardware accellerated 3D be embedded in the browser. If not, then why not? What are the challenges?
There is also the potential to have shared concepts for Silverlight and Flash. Obviously, they're very similar and have similar goals. But the question becomes: is it worth the effort. If you're already "going plugin", you might as well "go Silverlight" and
forget Flash altogether.
It was a good video. It could use a more going-deep like detailing of architechture and whiteboarding though. Some more information about why it doesn't use the Vista Audio Service / Engine. Something about their experiences with MMCSS.