Hello dear programmers,

I'm developing an application with Mappoint using VC++.
What I'm doing is getting a position from the GPS and put a smalle circle on the map. For testing I use a "static" GPS (in our office) that doesn't move, we just use it for testing. It all works well the GPS sends every second a new position (wich is finally the same position).

But the problem is it seems that every position I get, mappoint needs more processingtime. So after a few minutes, mappoint has to process about 300 pushpins, and than the processorfan really has to cool the cpu, because Mappoint sucks almost all of the cpu.

What I did is making a new pushpin everytime I get a new position.
And if I want to follow a vehicle on the map with its trace (these are the old pushpins on the map) than Windows will crash after 5 minutes.

So how can I solve this situation?

Again, I use this everytime I get a new position:

m_Loc = m_Map.GetLocation(Lat, Lon, 1);
m_Pin = m_Map.AddPushpin(m_Loc.GetLocation(), Txt);
m_Pin.SetSymbol(21);
m_Loc.GoTo();

I hope somebody can help me, so thank you very much

Hi,

I'm sorry I forget to mention that RAM-memory doesn't leak, so there is no memory leakage, only the cpu usage get's 100% by Mappoint only. I mean mappoint as an ActiveX ofcourse

Hi,

300 pushpin is not so mutch because the map does not to handle all 300, only 1 comes new every second. Maybe the problem is because you have honderds pushpins on same location with the test ?

Thank you for your reaction Wilfried,

I don't think it has to do with the same position on the map. An the reason I think so is, I have another project wich I use it for experimenting things with mappoint.

And the funny thing is, that I don't have that problem, though I have 500 pushpins, the cpu load remains the same. Only the RAM increases a little bit (you can see a very small change).


So I've to figure out what's the cause of increasing the cpu-load. It's a little bit difficult to find out, because I used cut&paste, so I don't see differences in code....

I hope someone can give a tip/tric to solve this. If I find the cause, I'll report it to you immediatly.

Hello programers,

I found out what the evil bug is (actually it's more a bad coding of mine ).
If you look at the code below:

void SomeFunction()
{
m_Loc = m_Map.GetLocation(dLat, dLon, 1);
m_Pin = m_Map.AddPushpin(m_Loc.GetLocation(), Txt);
m_Loc.GoTo();
}


Everytime I get a new latitude. longitude, I call this function to draw on the map. But by a wrong coding, Txt has the same string. Than something goes wrong, I don't really understand what's going on in Mappoint.
But somehow if I add a pushpin on the same location with the same NAME (that's Txt stands for) than uses more processingtime.

Can somebody explain me why that is? Because I didn't really dive in the mappoint API, I just looked for the needed functions to call, to make my application work.


I hop somebody can tell me why Mappoint behaves like that.

Do i understand you right? You add pushpins with AddPushpin(aLocation, aName) and this adding gets slower for every pushpin you add?
If this is the case, I encountered the same problem during performance tests and found an easy but strange solution:
don't use
AddPushpin(aLocation, aName);

try this one:
pin = AddPushpin(aLocation, "");
pin->Name = aName;


don't ask my why this is faster, but it's really much faster, the more pushpins you already have the greater the speedup is.

Example:
Adding 256 pushpins to a map with an empty string as name took 4.185 seconds
Adding 256 pushpins to a map with "moo" als name took 39.841 seconds
Adding 256 pushpins to a map with an empty string as name followed by pin->Name = "moo" took 6.177 seconds

It's even worse with 1024 pins:
No name: 13.269 seconds
"moo" as name: 38 minutes and 32.676 seconds!
No name followed by pin->name = "moo": 23.352 seconds

by the way, the length of the name makes only a small difference, if it's 1 character long or 50 doesn't really matter.

edit: and using different names for each pushpin doesn't matter too if i remember right (and my pushpins had 1° distance between each other)

Hi,

Thanks for feedback. Ok so that is the reason I did never encountered this problem because in each application all pushpins have exclusive name.

Probably it has something to do with database management. I assume there is an internal index on the pushpin's name, so a internal search to it bounces on the same over and over again. Probably it increment the internal index until EOF or name change.

I can not come to a better explanation / reason... But the nice thing is that problem is found

Well,

I hope other programmers have value information about this and I want to thank all of you.