- From: Ben Kelly <notifications@github.com>
- Date: Wed, 08 Apr 2015 19:00:07 -0700
- To: whatwg/streams <streams@noreply.github.com>
Received on Thursday, 9 April 2015 02:00:36 UTC
@domenic correctly pointed out on IRC that I need to prime the jit before my first measurement. Doing this gives more realistic results for the 10 chunk case: ``` 10 chunks in pipe: sync-read: 0.5 us/chunk vs. promise-read: 39.3 us/chunk ``` So ~0.4ms to read 10 chunks buffered in a pipe. Still much slower than the sync case, but not so slow in absolute terms to be considered ridiculously broken. --- Reply to this email directly or view it on GitHub: https://github.com/whatwg/streams/issues/320#issuecomment-91089642
Received on Thursday, 9 April 2015 02:00:36 UTC