[x3d-public] event tracing
brutzman at nps.edu
Thu Nov 3 09:45:08 PDT 2016
Very interesting. Wondering, do you think this might lead to violations of the loop-breaking rule for events?
Essentially, the rule says that all event values generated during a given event cascade (i.e. between frame draws) are checked. If two events are both produced by the same source into the same ROUTE, then duplication (likely a loop) has occurred. Such recurrences are dropped to avoid infinite loops.
I've always assumed that this means the duplicate events have the same timestamp, thinking for typical implementations that is probably what occurs.
However, if the check is performed by each individual ROUTE, which can tell if a redraw has occurred since last receiving an eventOut value, perhaps strict comparison of timestamp values isn't necessary?
184.108.40.206 Execution model
[...] "Nodes that contain output events shall produce at most one event per field per timestamp. If a field is connected to another field via a ROUTE, an implementation shall send only one event per ROUTE per timestamp. This also applies to scripts where the rules for determining the appropriate action for sending output events are defined in 29 Scripting component."
"Event cascades may contain loops where an event E is routed to a node that generates an event that eventually results in E being generated again. See 220.127.116.11 Execution model, for the loop breaking rule that limits each event to one event per timestamp. This rule shall also be used to break loops created by cyclic dependencies between different sensor nodes."
29.2.7 Asynchronous scripts
"Some languages supported by X3D browsers may allow Script nodes to spontaneously generate events, allowing users to create Script nodes that function like new X3DSensorNode nodes. In these cases, the Script is generating the initial events that causes the event cascade, and the scripting language and/or the browser shall determine an appropriate timestamp for that initial event. Such events are then sorted into the event stream and processed like any other event, following all of the same rules including those for looping."
On 11/3/2016 6:32 AM, Andreas Plesch wrote:
> On Mon, Oct 31, 2016 at 6:54 PM, Don Brutzman <brutzman at nps.edu <mailto:brutzman at nps.edu>> wrote:
> Tres cool, thank you for this information Andreas.
> Trace performance of your teapot widgets is absolutely tremendous, even while printing trace values on the screen (historically a bog on performance). No discernible change in perceived responsiveness, with/without firebug tracing, even when i interactively try to provoke sluggishness (running good old Windows 7 firefox).
> I slightly updated the event logging to use performance.now rather than event.timeStamp which is consistent across browsers and to also report the difference between the x3d timestamp and the time of logging which should happen just after dispatching of the DOM event and before drawing to the webgl canvas. The x3d timestamp is created at the beginning of processing of x3d events for each cascade, I believe. This difference varies considerably, about between 2 ms and 20 ms, even for repeats of the same events. Chrome shows a little less difference on average. So I am not sure how accurate these timings are in relation to when things really happen and what external factors influence these.
all the best, Don
Don Brutzman Naval Postgraduate School, Code USW/Br brutzman at nps.edu
Watkins 270, MOVES Institute, Monterey CA 93943-5000 USA +1.831.656.2149
X3D graphics, virtual worlds, navy robotics http://faculty.nps.edu/brutzman
More information about the x3d-public