Off by 1 pixel display problem

GoogleCodeExporter commented 8 years ago

What steps will reproduce the problem?
1. Pan the map display around, and note the edges of the tiles as they are
brought over from the network.
2. Occasionally a tile's right edge will have a black line (1 pixel wide)
3. This only happens on the device (iPod 2.0 and iPhone 2.1 firmware). It
is not producible on simulator.

What is the expected output? What do you see instead?
No tiles should have a black edge.

What version of the product are you using? On what operating system?
Reproducible as of r76

Original issue reported on code.google.com by pspda...@gmail.com on 3 Nov 2008 at 5:31

GoogleCodeExporter commented 8 years ago

Image of defect

Original comment by pspda...@gmail.com on 3 Nov 2008 at 5:37

Attachments:

IMG_0009.png

GoogleCodeExporter commented 8 years ago

Problem confirmed :(

Original comment by jose...@gmail.com on 3 Nov 2008 at 6:56

Changed state: Accepted

GoogleCodeExporter commented 8 years ago

Any idea on the root cause? Seems to me like there's some floating point 
accuracy error.

Original comment by pspda...@gmail.com on 6 Nov 2008 at 7:23

GoogleCodeExporter commented 8 years ago

Doubt its a floatingpoint error. If it was, it wouldn't be exactly 1 pixel. 
It'll be 
an addition ordering error - somethings are adding in the wrong order when its 
both 
panning and zooming, or something like that.

Original comment by jose...@gmail.com on 6 Nov 2008 at 7:50

GoogleCodeExporter commented 8 years ago

I've also seen this on the simulator, I don't think it's device specific.

Original comment by kennygrant on 9 Nov 2008 at 3:50

GoogleCodeExporter commented 8 years ago

Its not; and its related to zoom ordering or something.

I added some debug code to list the bounding boxes of all the tiles onscreen 
(x1,y1,x2,y2). When the screen first loads, here's where the tiles are 
positioned:
2008-11-10 18:20:38.368 MapView[28301:20b] Image at -103.000015, 144.500015 
153.000015 400.500061
2008-11-10 18:20:38.369 MapView[28301:20b] Image at -103.000015, 656.500061 
153.000015 912.500122
2008-11-10 18:20:38.373 MapView[28301:20b] Image at 153.000015, -111.500015 
409.000061 144.500015
2008-11-10 18:20:38.374 MapView[28301:20b] Image at 153.000015, 144.500015 
409.000061 
400.500061
2008-11-10 18:20:38.374 MapView[28301:20b] Image at -103.000015, 400.500061 
153.000015 656.500122
2008-11-10 18:20:38.375 MapView[28301:20b] Image at -103.000015, -111.500015 
153.000015 144.500015
2008-11-10 18:20:38.375 MapView[28301:20b] Image at 153.000015, 400.500061 
409.000061 
656.500122
2008-11-10 18:20:38.376 MapView[28301:20b] Image at 153.000015, 656.500061 
409.000061 
912.500122

... And after panning and zooming a bit, the numbers are like this:
2008-11-10 18:19:07.920 MapView[28292:20b] Image at 21.897797, 583.302063 
319.240997 
880.645264
2008-11-10 18:19:07.921 MapView[28292:20b] Image at -275.568390, -11.414162 
21.774872 
285.929108
2008-11-10 18:19:07.925 MapView[28292:20b] Image at 21.912323, -11.455841 
319.255585 
285.887421
2008-11-10 18:19:07.927 MapView[28292:20b] Image at 319.240967, 285.958801 
616.584167 
583.302002
2008-11-10 18:19:07.928 MapView[28292:20b] Image at 319.255585, -11.455841 
616.598877 
285.887421
2008-11-10 18:19:07.928 MapView[28292:20b] Image at -275.568390, 285.929108 
21.774872 
583.272339
2008-11-10 18:19:07.929 MapView[28292:20b] Image at 319.240967, 583.302063 
616.584167 
880.645264
2008-11-10 18:19:07.930 MapView[28292:20b] Image at 21.897797, 285.958801 
319.240997 
583.302002
2008-11-10 18:19:07.930 MapView[28292:20b] Image at -275.568390, 583.272339 
21.774872 
880.615601

Notice that in the second bunch of numbers they don't quite line up properly; 
eg tile 
1's right hand side is on pixel 319.240997 and tile 4's left hand edge is at 
319.255585. Thats _much_ more than floating point roundoff error. I think when 
the 
tiles are zoomed about a point the zoom is incorrectly applied before the pan, 
or 
something like that. Its subtle, but as you've pointed out its definitely a bug.

Original comment by jose...@gmail.com on 10 Nov 2008 at 7:24

GoogleCodeExporter commented 8 years ago

This seems to be fixed along with the fixes to small jumps.

Original comment by jose...@gmail.com on 20 Nov 2008 at 2:25

Changed state: Fixed

GoogleCodeExporter commented 8 years ago

Although the frequency has lessened, this problem can still be reproduced. Can 
this
bug be reopened and looked at?

Original comment by pspda...@gmail.com on 29 Nov 2008 at 8:22

GoogleCodeExporter commented 8 years ago

Original comment by jose...@gmail.com on 30 Nov 2008 at 7:35

Changed state: Accepted

GoogleCodeExporter commented 8 years ago

I have the problem also on device environment.

Original comment by acs...@gmail.com on 27 Feb 2009 at 1:32

yy20111011659 / route-me

Off by 1 pixel display problem #17