[Bf-funboard] Curves creating and Transform

[Matt Ebb]

On 29 Dec 2004, at 1:02 PM, Martin Poirier wrote:

> Which brings a nice question.
> What to do when the user tries to tilt a 2D curve?
> Nothing (cancel the tilt operation) or toggle on the
> 3Dness of the curve?
>
> I'd personnally be more in favor of solution 2. While
> a bit magical, it keeps in line with the "standard"
> Blender line of thought that the user knows what he
> wants to do.

Agreed

> Indeed, that's why I suggested a special toggle to
> choose between proportionnal handle editing or not.
> This would be very similar to what Illustrator does.
> Simple hotkey (if we can find one) and button in the
> 2D view header (akin to the selection modes for
> meshes, the PET, ...).

And agreed again.

> http://mke3.net/blender/interface/interaction/illu_whitearrow.mp4
>
> Kinda off topic, but I think this is a great example
> why I think the Blender workflow of transform through
> hotkeys and simple keyboard/non precise clicks
> activated constraining is superior to click
> drags/transform widgets (not even mentionning the fact
> that holding down a mouse button during a motion that
> could demand precision is ridiculous).
>
> At approximately the half of the video, look how you
> try to move the handle by click dragging, misses the
> hotspot (by not much) and end up not moving anything
> AND loosing your selection.

Heh, well the real problem is Illustrator's dumb method of selection - 
you need to click almost exactly on the point, which isn't easy on a 
high-ish res screen. If it had a system like Blender, where it guessed 
from the closest available points, that problem wouldn't happen - 
nothing to do with dragging or not.

> Others can feel free to disagree, of course.

Indeed ;) I'm with Tom here - there's a more inaccuracy in pressing a 
mouse button then there is in releasing it - you have to impart a lot 
more force to press and click it, than just to release the pressure. 
That force can cause the pointing device to slip - the most obvious 
example being using a tablet.

I don't know if you've ever tried it, but attempting to use a tablet 
with Blender fits the description of 'ridiculous' perfectly :) With a 
tablet, it's almost impossible to hit an exact point by pressing down 
from the pen being in a raised position. To do this, you need nerves of 
steel (not for shaky fingered coffee drinkers) and also a very precise 
aim, so that the pen travels down perfectly perpendicular to the tablet 
surface. Usually, when you click, it comes in on an
angle (how many people here write or draw with a pen perfectly 
vertical? Not many, I bet) meaning that the point you see on the screen 
when you have the pen in the air and decide "I want to click here" is 
often different to the point where the click actually occurs. It's far 
more accurate to just release the pressure, when the pen is already 
where you want it.

It's a real shame, I love my tablet and really wish I could use it with 
Blender but it's just such a waste of time.

> That brings up another problem though. How do you
> calculate the handle for a newly created CV (extrusion
> or Ctrl-click)?
> Though what could be done is apply the auto-handle
> calculations to calculate the original handles and
> then have the resulting mode be aligned.

Yep.

> Agreed on both points.
> For the later, it really helps jumpstart a discussion
> if the original poster brings arguments/debatable
> points and not just "lets do this" kind of things.

Absolutely. Though I do expect a higher standard from the non-newbies 
here that should know better.

Cheers,

Matt
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2355 bytes
Desc: not available
Url : http://projects.blender.org/pipermail/bf-funboard/attachments/20041229/fd740b21/smime.bin