>>>Hi Peter
>>>
>>>I have found that the sudden drop in memory usage only happens if the vfp app goes from a non-minimized state to a minimized state. VFP does not even fire its internal memory free-up routine if the app is not visible! In other words I tried
>>>
>>>With _SCREEN
>>>
>>> .visible=.F.
>>> .WindowState=2
>>> .WindowState=1
>>> .Visible=.T.
>>>
>>>EndWith
>>>
>>>but vfp did nothing iro memory since it apparently realizes that it is not visible so therefore does not need to do the windowstate changes and therefore no memory clean up. Otherwise the above would be great LOL.
>>>
>>>Jos
>>
>>Jos,
>>
>>I've been trying to manipulate (fool) the thing regarding the flicker, but to no avail.
>
>You have tried the following, haven't you?
>

>THISFORM.LOCKSCREEN = .T.
>..control changes here, with necessary refreshes
>THISFORM.LOCSCREEN = .F.
>
>

Well, including that for _VFP, _VFP.AutoYield, NoDefault, _VFP.Left=-1600 and more of this stupid stuff you can think of in half an hour of time.

I just tried to prove memory leaks by means of our own app, but I just can't do it. But again, AFAIK from off W2K these problems are over.

>
>
>
>
>
>>I'd still be interested in the result on alt-tab.
>>Beware of not having the wrong base-angle here : "... and therefor no memory cleanup". Yeah, if VFP DOES that. But what if the OS doing it (IOW releasing the leaking memory then).
>>
>>I tried to think of what VFP could release because of minimizing its window, but can't come up with anything, except for added up screen pixels or so.
>>
>>Jos, is it hard to provide some code that will show the decreasing memory (i.e. emulate this behaviour) ? I'd really like to go along with this since I'm always in for more "speed". Again, this is (IMO) connected to it.
>>
>>Peter