[Qgis-user] jumping scale

[Richard Duivenvoorde]

Carson Farmer wrote:
>>> This behaviour is normal if the project works with an unprojected CRS 
>>> such as the default WGS-84, because the scale differs in each point 
>>> of the canvas. The canvas would has to be re-zoomed while panning to 
>>> keep the fixed scale. But isn't.
>>>
>>> However the problem doesn't concern projected CRSes, so just set your 
>>> project CRS to any (say the Dutch 28992), enable on-the-fly and the 
>>> scale should remain fixed.
>>
>> I tried to set up a project in epsg:28992 and load a shape in 28992, 
>> but still have 'jumping scales' (though it indeed looks like in the 
>> North-South direction the jumps are much larger then in West-East 
>> direction).
>>
>> Should I file a bug for this?
> I still don't think you should file a bug, as epsg: 28992 is an oblique 
> stereographic projection. This means it is a conformal projection (i.e. 
> it preserves angles). It does not preserve distances or areas, and so 
> the scale will vary. You are best to use some sort of equidistant 
> projection.

Hi All,

Borys is right. If besides setting the projection to epsg:28992 I ALSO 
set the units to meters it works as I want to (that is: being able to 
zoom to a more or less fixed zoomlevel and use that zoomlevel for 
'fieldwork-printouts').

Don't know how this relates to the terms Carson mentioned.
I still think things go wrong somewhere, even if there is a change in 
style for geograpic projections, then still the jumps I see are too 
large I think.
And setting both project and data projection to epsg:28992 (in which 
indeed every number is a meter), it's still strange that apparently qgis 
does the math in degrees for the scale-indicator? And helping qgis by 
setting it to meters it's all right?

But I'm not too much into geodesy too know when exactly the scales 
SHOULD jump.

Thanks for your help all!

Regards,

Richard Duivenvoorde