Issue Inferencing along the Green Axis...
-
-
In the images @Wo3Dan showed, the intersection of the Green axis vanishing point with the horizon seems to be visible. But if you ponder a moment, you must realize that this point is infinitely far away and can never actually be seen! The ratio of pixel to distance (foreshortening) thus becomes infinite as you move up the view away from the origin! I think this makes it impossible for the inference engine to determine where you are intending the cursor to be in the model, and it gives up.
-
Hi Steve, Have you tried the model I attached?
I only made these observations, ...... marking them all with *****. This does however not explain why SketchUp can't find the axis. (and I don't understand your explanation as to why)
You are right that the vanishing point can't bee seen due to infinit distance. But looking at the 'blind' area on screen this seems to be quite large compared to the entire viewport.
I wonder if the program couldn't be made to narrow down the blind area to something significantly smaller.The model origin's location on screen has no influence.
-
Interesting observations regarding the centre screen inferencing 'blind spot'. This may explain why inferencing is, on occasion, so incredibly frustrating — when most of the time it works like magic. Seems the work-around for now is to draw
outside of that centre circle. -
I opened your file Wo3Dan and tried it out but didn't see any issues...
-
Yes, Gerrit, I tried your model. As saved, inferencing on the green axis indeed fails. But if I orbit so that the axis isn't so straight back in the view, green axis inferencing works fine. Having the green axis oriented too close to the "into the screen direction" in perspective seems to induce the problem, and I was suggesting that maybe this direction causes the inference engine to struggle trying to infer how far back in the "into the screen direction" the cursor is marking in empty space. I should emphasize that my explanation was only a theory - without access to the SketchUp code I'm just guessing based on symptoms.
-
@utiler said:
I opened your file Wo3Dan and tried it out but didn't see any issues...
So did I, to see if my uploaded file was any better, .... fixed something that I wasn't aware of.
But it is still the same, the indicated area inside the yellow "circle" doesn't trigger inference to any of the three axes. Is there no center blind spot at all for you? is this a screen / graphics card / driver depending issue? -
Ok, done it again.
This time it didn't work on the RED or BLUE axis.... Note I was within a component moving other components around. I also tried to orbit around to various locations to trick it a bit but I just couldn't find those two axises...
The only way is to add guidelines which then works.
-
-
And just tried a new model with nothing in it....
Can't find the red axis.
-
@utiler said:
And just tried a new model with nothing in it....
Can't find the red axis.Just tested a model similar to the one in your file, feb 13.
With the camera position you presented I could find R, G and B axes in FOV less than 10 deg upto 120 degrees. No problem here.
The blind spot that seems to only appear on my laptop decreases with a larger FOV.
Weird.
Advertisement