HN user

Haldir

17 karma
Posts0
Comments12
View on HN
No posts found.

The techempower Benchmarks are related to ASP.NET core, that one you can use on Desktop too.

The coreclr is only maybe 10-15% faster than Desktop, but not a magnitude or more compared to old ASP.net.

I don't think that the clr is slower than the jvm, but quite a few basic frameworks are compared to their Java counterparts

Topic is misleading. It might be true on .Net as the Version is 3.0.3, which might be the latest released Version for .Net (there is a prerelease of 4.8) but compared to Java it is ancient.

Let's look at some real live x265 benchmarks, because you skipped the part where they did it on a beefy dual socket server. Let's take: http://x265.ru/en/x265-hd-benchmark, that benchmark is 1080P, let's assume you're correct with your simplicity and it might be closer to reality.

The fastest listed system there does a blazing 33fps (well that's a broadwell, the 2015 dual socket server is probably atleast the xeon equivalent of it).

And we haven't even looked at the image quality/compression artifacts.

I disagree with your assessment.

Rivvr is apparently a spinoff of Sixa, the appropriate techcrunch article https://techcrunch.com/2016/12/09/sixa-secures-3-5m-as-it-la... repeats the 11ms latency. Maybe it's the same technology? The comments to the techchrunch article do not sound convincing either.

I wonder how that should work. Less than 11ms per Frame compression time, let's say 9ms (as we need to Account for the way back too for latency) for a 2160 x 1200 stream + audio. And you can't see any compression and we're talking Close up here. Not possible with h264/h265.