[Opensim-dev] Sim and Phys Frames per Second (FPS) Issues
Mike Chase
mike.chase at alternatemetaverse.com
Mon Mar 2 21:00:12 UTC 2015
11 does seem really low. I suppose someone has profiled this in the
past. Any idea what the bottleneck is.
Mike
On 3/2/15 3:54 PM, Dahlia Trimble wrote:
> I believe 11 sim FPS is the target value. It would probably never go
> above this, but a number consistently lower than 11 fps would indicate
> performance problems. If the "lie" is simply a linear scaling, then it
> would have no impact in the ability to use the number as a comparitive
> statistic. since all such numbers would be similarly scaled.
>
> I also believe some of these numbers eventually make it to the viewers
> and are used to adjust moving entity velocity when the sim is running
> slowly. I'm not sure of the exact path but I think "time dilation" is
> involved. Altering these values may induced undesirable movement
> effects when the simulator is running slower than normal.
>
> Raising the target above 11 would also likely induce issues as other
> parts of the code assume 11 is the target.
>
> On Mon, Mar 2, 2015 at 11:16 AM, Sean M <mondesire.sean at gmail.com
> <mailto:mondesire.sean at gmail.com>> wrote:
>
> There are 3 frame rates: /simulation/, /physics/, and /client/.
> Please, someone correct me if I am wrong: the /simulation fps/ is
> how many simulation loops can be processed in a second, where each
> loop processes physics, scripting, region updates, etc; /physics/
> /fps/ is just how many physics loops get processed per second,
> where each loop calculates collisions, gravity changes, and other
> movement updates for all appropriate objects in a region Both
> physics and simulation (phys/sim) fps are server-side. The fps the
> /client/ is concerned with is graphics rendering on the viewer
> (firestorm, singularity, etc) and is literally how many visual
> frames/shots are displayed per second to the client. From a client
> point of view, 11fps is low because that is a low amount of image
> updates the client is visually seeing on their viewer. I am
> concerned with the simulation and physics FPSs being reported with
> highly inflated numbers. I want to use sim and phy FPSs as
> measures of how my region is performing but concerned that the
> correctionFactor of 5 is invalidly skewing these metrics, making
> both fps useless.
>
> Michael, or anyone else, do you know why OpeSim ticks at 11 fps?
> Why 11? Is my understanding of what constitutes a "frame" correct?
>
> On Mon, Mar 2, 2015 at 1:37 PM, R.Gunther <rigun at rigutech.nl
> <mailto:rigun at rigutech.nl>> wrote:
>
> 11fps ? is that not very low. 22fps gives more rome or 33fps.
> But i admit that i read fps maby wrong, and have nothing today
> with the framerate or smoothness on the screen..
>
>
>
> On 2015-03-02 19:28, Michael Emory Cerquoni wrote:
>> The reason physics and scripting are locked at 11fps is
>> because this is what the OpenSimulator heartbeat ticks at,
>> the reason it is multiplied is to satisfy the viewer
>> statistics, I am not sure its possible to have it report the
>> legitimate numbers without some wierd side effects, but I
>> could be wrong, you would have to experiment, I suspect
>> though that changing this could lead to a lot of badness.
>>
>> On Mon, Mar 2, 2015 at 1:23 PM, Sean M
>> <mondesire.sean at gmail.com <mailto:mondesire.sean at gmail.com>>
>> wrote:
>>
>> Greetings,
>>
>> We at the MOSES project have noticed Simulation and
>> Physics frames per second (FPS) have a few issues that we
>> are trying to resolve. The issues are
>> producing suspicious performance statistics for the
>> analysis of the current version of OpenSim that we are
>> running.
>>
>> First, there is a correction
>> factor (m_reportedFpsCorrectionFactor) that the raw
>> SimFPS is multiplied by. The comment in the following
>> line is a bit curious because it indicates that the FPS
>> is artificially inflated to "lie" about the actual FPS
>> being so low:
>>
>> OpenSim/Region/Framework/Scenes/SimStatsReporter.cs: Line 317
>> // We're going to lie about the FPS because we've been
>> lying since 2008. The actual FPS is currently
>> // locked at a maximum of 11. Maybe at some point this
>> can change so that we're not lying.
>> int reportedFPS = (int)(m_fps *
>> m_reportedFpsCorrectionFactor);
>>
>> Also, lines 174 and 227 mention the use of this
>> correction factor.
>>
>> Second, this multiplier also comes into play in the Scene
>> where there is a MinFrameTime, which seems to be the
>> minimum reported amount of time to process a frame:
>> OpenSim/Region/Framework/Scenes/Scene.cs:Line 723
>>
>> Both of these variables, the correction factor and
>> MinFrameTime, are concerning from a statistics view point
>> as they are generating skewed and massaged numbers;
>> therefore, I have a few questions:
>>
>> 1) Is it commonly known that Sim and Phy FPSs are
>> inflated to maintain the "lie"? And if so, will it be
>> corrected to be an accurate reporting of processed frames
>> per second?
>>
>> 2) What exactly are the definitions for OpenSim's
>> Simulation (Sim) FPS, Physics (Phy) FPS and a frame (I
>> have found conflicting and vague definitions on the wiki)?
>>
>> 3) What are the known performance consequences of setting
>> the m_reportedFpsCorrectionFactor to 1 and MinFrameTime to 0?
>>
>> Thanks,
>> Sean M.
>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at opensimulator.org
>> <mailto:Opensim-dev at opensimulator.org>
>> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-dev
>>
>>
>>
>>
>> --
>> Michael Emory Cerquoni
>>
>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at opensimulator.org <mailto:Opensim-dev at opensimulator.org>
>> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-dev
>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at opensimulator.org
> <mailto:Opensim-dev at opensimulator.org>
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-dev
>
>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at opensimulator.org <mailto:Opensim-dev at opensimulator.org>
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-dev
>
>
>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at opensimulator.org
> http://opensimulator.org/cgi-bin/mailman/listinfo/opensim-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://opensimulator.org/pipermail/opensim-dev/attachments/20150302/bbfb152f/attachment-0001.html>
More information about the Opensim-dev
mailing list