W3C home > Mailing lists > Public > public-web-perf@w3.org > August 2011

Re: [Performance Timeline] Need higher resolution timers

From: James Robinson <jamesr@google.com>
Date: Tue, 23 Aug 2011 14:28:26 -0700
Message-ID: <CAD73mdLOX0HGRZPU0NHmRLiaDW-kBoJaxq-wZF7=oti23STfkQ@mail.gmail.com>
To: Jatinder Mann <jmann@microsoft.com>
Cc: James Simonsen <simonjam@chromium.org>, public-web-perf <public-web-perf@w3.org>, Zhiheng Wang <zhihengw@google.com>
On Tue, Aug 23, 2011 at 2:25 PM, Jatinder Mann <jmann@microsoft.com> wrote:

>  I agree that without sub-millisecond resolution, you cannot get the
> precise framerate based on the timestamps of two separate frames. You can
> still calculate the framerate by counting the number of frames per second.
> ****
>
> ** **
>
> I think this will be a good topic for discussion for tomorrow’s conference
> call. If we do decide to allow for microsecond resolution, we should do so
> in the constraints of respecting the Unix epoch, so developers can easily
> compare timestamps with Date.now(), and be inclined to not change Navigation
> Timing unless it is for a very compelling reason.
>

That's not possible. Date.now() changes when the system clock is adjusted,
the monotonic clock (by definition) does not.  Whenever the system clock is
adjusted the two values become incomparable.  I think it's far better to be
up front about the fact that they timestamps are not comparable by using a
separate timebase so that developers are less surprised when clock
adjustments happen in the wild.

> ****
>
> ** **
>
> I will add this topic to the agenda.
>

Let's discuss technical issues on the mailing list, we are more likely to
get broader feedback and reach a technically sound consensus than on a phone
call.

- James

> ****
>
> ** **
>
> Jatinder****
>
Received on Tuesday, 23 August 2011 21:28:54 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:04:31 UTC