[mapguide-users] Mapguide 2.0 Beta1 vs 1.2 performance

Traian Stanev traian.stanev at autodesk.com
Thu Nov 22 11:41:50 EST 2007 

Is your map polygon feature heavy by any chance? In general having many small polygons will make less difference than having a few very large polygons, due to the huge difference in pixel fill rate between the two renderers. Our GD renderer uses a custom polygon rasterizer routine that is extremely fast (but output is not antialiased). There are ways to optimize AGG fill rate, but I have not yet had time to do that.

Another thing that could cause such a difference is if you have huge amounts of labels in there.

But either way there is no excuse for GD taking 10 seconds now when it only took 2 seconds in 1.2 and that must be fixed. Is it possible for you to send a screenshot -- it may be helpful in determining what's going wrong.


Traian




> -----Original Message-----
> From: mapguide-users-bounces at lists.osgeo.org [mailto:mapguide-users-
> bounces at lists.osgeo.org] On Behalf Of Bruno Scott
> Sent: Thursday, November 22, 2007 9:07 AM
> To: mapguide-users at lists.osgeo.org
> Subject: RE: [mapguide-users] Mapguide 2.0 Beta1 vs 1.2 performance
>
>
> here is my test today with GG and AGG
>
> mg2.0+AGG-fusion 17 sec
> mg2.0+AGG+fusion  30 sec
> mg2.0+GD-fusion 10 sec
> mg2.0+GD+fusion    20 sec
>
> GD seems faster than AGG
>
> A little precision, i have a huge screen resolution 1900x1200
> I did my test full screen
>
> I have also noticed something, the image generated by fusion is a lot
> larger
> than the one without fusion.
> We see it when we pan, the image is about 20% bigger than what we see
> on the
> screen.
> 20% bigger in 4 direction make the image about twice big.
> And with a resolution of 1900x1200 this could explain the gap in
> performance
> with or without fusion.
>
>
> Bruno
>
>
> Bruce Dechant wrote:
> >
> > The performance difference can be attributed to the fact that this is
> a
> > Beta version being compared to a Release version.
> >
> > Some things to consider:
> > - 1.2 used the GD renderer, whereas, 2.0 is using the new AGG
> renderer
> > (looks much better)
> > - there are several places in the code that have not been optimized
> yet
> > (algorithms, memory usage, thread usage, etc...)
> > - Fusion is a brand new feature of the 2.0 release and has not been
> > optimized either
> > - this is a Beta and doesn't necessarily reflect the final release
> code.
> > :)
> >
> > Thanks,
> > Bruce
> >
> > -----Original Message-----
> > From: mapguide-users-bounces at lists.osgeo.org
> > [mailto:mapguide-users-bounces at lists.osgeo.org] On Behalf Of Paul
> Spencer
> > Sent: November 21, 2007 10:04 AM
> > To: MapGuide Users Mail List
> > Subject: Re: [mapguide-users] Mapguide 2.0 Beta1 vs 1.2 performance
> >
> > It uses the mapdefinition that is set in the application definition,
> > if its a tiled map then it should be tiled in the client, otherwise
> it
> > should be a regular old map draw.
> >
> > Has anyone else got feedback on this?  I saw fusion running on a
> > laptop yesterday, not tiled, full screen, and it was milliseconds
> > (<500ms) to pan/zoom (didn't look at select).
> >
> > Paul
> >
> > On 21-Nov-07, at 3:22 AM, Zac Spitzer wrote:
> >
> >> sounds like fusion ain't hitting the tile cache???
> >>
> >> On Nov 21, 2007 6:44 PM, Bruno Scott <bscott at geomapgis.com> wrote:
> >>>
> >>> I have 2 identical vm-ware with windows 2003 running on the same
> >>> machine
> >>> I start one of the vm-ware, I do the testing, I stop it and then
> >>> start the
> >>> other to do the other bench Bench is done on the sheboygan sample
> >>> at 1:5000
> >>> scale
> >>>
> >>>        MG 2.0 with fusion              Mg 2.0 without
> >>> fusion           Mg 1.2
> >>>
> >>> pan     29 sec                  12 sec                          2
> sec
> >>>
> >>> select  28 sec                  12 sec                          2
> sec
> >>>
> >>>
> >>> I know it's only a beta but why is that slow?
> >>>
> >>> Bruno Scott
> >>>
> >>> --
> >>> View this message in context:
> >>> http://www.nabble.com/Mapguide-2.0-Beta1-vs-1.2-performance-
> tf4848550s16610.html#a13872479
> >>> Sent from the MapGuide Users mailing list archive at Nabble.com.
> >>>
> >>> _______________________________________________
> >>> mapguide-users mailing list
> >>> mapguide-users at lists.osgeo.org
> >>> http://lists.osgeo.org/mailman/listinfo/mapguide-users
> >>>
> >>
> >>
> >>
> >> --
> >> Zac Spitzer
> >> http://zacster.blogspot.com/
> >> +61 405 847 168 (aussie moible)
> >> _______________________________________________
> >> mapguide-users mailing list
> >> mapguide-users at lists.osgeo.org
> >> http://lists.osgeo.org/mailman/listinfo/mapguide-users
> >
> > _______________________________________________
> > mapguide-users mailing list
> > mapguide-users at lists.osgeo.org
> > http://lists.osgeo.org/mailman/listinfo/mapguide-users
> > _______________________________________________
> > mapguide-users mailing list
> > mapguide-users at lists.osgeo.org
> > http://lists.osgeo.org/mailman/listinfo/mapguide-users
> >
> >
>
> --
> View this message in context: http://www.nabble.com/Mapguide-2.0-Beta1-
> vs-1.2-performance-tf4848550s16610.html#a13897170
> Sent from the MapGuide Users mailing list archive at Nabble.com.
>
> _______________________________________________
> mapguide-users mailing list
> mapguide-users at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/mapguide-users

More information about the mapguide-users mailing list