Hi Marco thanks for the bug report!

> - UI-size skips instantly to 30

Unfortunately I can't seem to repeat this one over here, I'll keep trying on a few different systems.

I may need to add some kind of extra debugging output to the next beta to try and track this one down.


> - Long object names in the Browser (might be normal behaviour..)

Yeah I wasn't quite sure whether to leave that or not because it is kind of good with long names to see as much of the name as you can.

But it's pretty messy having things draw over top of each other like that.

What about a kind of hybrid behavior - if there wasn't any selection dot there at all then I could let the text run into that selection dot area, but if there is a dot displaying then the text would get truncated before the dot.

The other option is to have it always truncated leaving the dot as a completely separate column but that means the selection dot will have reduced the display space for the text name in all cases.

- Michael

Hi Michael,

first, congratulation for V3! The new interface (UI) seems to work just fine.

I've tried the flow command on the file you provided and it working as expected. I then tried a surface to surface flow and here is what I get (Flow-test001). My intention was to deform a medallion from a flat surface to a rounded one. So I started simple but I didn't get what I was expecting. I assume something can be done to put the ring on the outside of the rounded surface. Also, as you can see both surface have the same Y length, so I expected the ring to have the same Y value as the original and since in the X direction the surface have different length I expected the ring to be strech.

Just to be on the safe side, I tried to make sure both surfaces had the same X and Y dimensions and this time I got what I wanted. (Flow-test002)

I think for cases like the first example, it would be nice to have somekind of option to flip the ring to the inside/outside and it maybe nice to have control over which direction (X,Y, maybe Z and XY, XZ, YZ, etc.) to stretch things beside having to do it by carefully (re) creating both surfaces.

Thanks and congratulation again,
Felix

Hi Michael

UI-size-thingy:

http://www.screencast-o-matic.com/watch/cXjZXa3uw

or




Truncate:
..thats really a tricky bit of usability-issue.
Normally i would vote for an Ellipsis (…) (bold or somehow visually enhanced) ..eventually a Guillemet (»)..but sometimes you have to deal with generated names and then this approach might not work as intended.





Cheers,
Marco

Hi Michael,

> What about a kind of hybrid behavior.......

> The other option is to have it always truncated
> leaving the dot as a completely separate column...

My vote is for either method as long as the name doesn't end up under the indicator dot.

Another thing I've noticed Michael, the icon fonts seem a little different to V2 and the Icons themselves seem to be softer or is that just me?

-

< seem to be softer or is that just me?
Maybe default screen regulates lights are not the same than your V2?

Hi Pilou,
I have both V2 and V3 on the same computer and screen.

-

""""""Another thing I've noticed Michael, the icon fonts seem a little different to V2 and the Icons themselves seem to be softer or is that just me?"""""""""

I have the difference too Danny.. Havnt look at the css or ini files for differences yet..

BTW:

I've felt it and it feels good! :o

Hi Danny,

> Another thing I've noticed Michael, the icon fonts seem a little
> different to V2 and the Icons themselves seem to be softer or
> is that just me?

It uses the same mechanism as v2 for scaling the icon images... But possibly the font size of the UI is slightly different than v2 on your system making a slight change in appearance with things like 2 pixels bigger or smaller.

Could you please post a screenshot of both v2 and v3 so I can see the difference? Please use PNG format to avoid any JPEG compression fuzz.

- Michael

Hi Marco, thanks for the UI size video. I seem to remember having a similar problem at one point before and I thought that I had gotten it fixed but apparently not all the way. Something to do with a timing issue for when the slider gets initialized. I'll examine it and see if I can get it fully solved for the next beta.

Are you use you don't want to leave the UI at size 30? That makes it really easy to read the text! ;)

- Michael

Hi Michael,
I had commented that I noticed it too... I was following through with your request to Danny and noticed that it was actually being drawn BETTER!!! (No different font or size.)

I dont know if this screengrab is large enough to show it.. V2 is on the right, V3 is on the left



Maybe the "difference we see, is that we are seeing a "Clearer" picutre of whats there??

Hi Felix, re: flow question

> I assume something can be done to put the ring on the
> outside of the rounded surface.

For now you can either sink the ring down to the bottom of the plane surface, or you can select the rounded surface and run the Flip command on it to flip its positive surface normal direction to the opposite side:
http://moi3d.com/2.0/docs/moi_command_reference10.htm#flip

Right now surface-to-surface flow will use the positive surface normal direction of the surface for the mapping, and with open surfaces this can often be pointing in a way that you didn't really expect - it's dependent on things like what direction (clockwise or counterclockwise) you drew some of your initial curves in or things like that. If you make things into a solid then it will orient surfaces to have the positive normal direction pointing to the outside of the solid.


> Also, as you can see both surface have the same Y length,
> so I expected the ring to have the same Y value as the original
> and since in the X direction the surface have different length
> I expected the ring to be strech.

So what is happening here is the U and V directions of the rounded surface are swapped from what you're expecting.

Every surface has a kind of UV space associated with it, similar to how texture mapping works in rendering. One direction of a surface's control point grid is the U direction and the other direction is the V direction. They're not really forced to be aligned in any special way to the world axis directions.

Flow actually maps things from one UV space to another UV space, it doesn't really map things in world x/y/z axis directions - the directions all come from this UV space.

So in this case the U direction of the plane is the horizontal one, but the u direction of the surface is the vertical one it ends up mapping from one direction to the other causing this distortion.

Right now I don't have any way set up to swap the UV axis directions so the solution right now would be to squish the plane so that it was horizontally the size that you wanted to map to the vertical direction of the rounded one.


> I think for cases like the first example, it would be nice to
> have somekind of option to flip the ring to the inside/outside <...>

Yeah I want to make some way to control it - that's an area that I still need to work on adding to the Flow command, I just didn't want to delay the beta any longer before finishing that up.

I'll probably be trying to make it work by paying attention to which corner you click nearby on each surface when you select them and then matching those corners together, that should make it so that you won't have to worry about u and v directions anymore.

For curve to curve flow that works already - it matches things starting with the ends that you clicked closest to but I didn't finish the equivalent thing for surface to surface flow yet.

- Michael

Hi Burr, so it does look there like a different font is being used - notice for example the 'a' character in "Draw curve" is a different glyph between those 2 screenshots.

Some change in font detection code in v3 apparently thinks that you don't have the Century Gothic font installed, that's the difference, so it's defaulting to Arial instead of the regular default font.

If you want v2 to look the same way, go to moi.ini and set

[UI]
UIFontName=Arial

and then it should look the same between the 2 versions font-wise anyway - can you give that a try to verify that?

- Michael

Hi Michael,
Yeah, the V3 was set to arial.. Setting them both to match made them equal...

Is there anything you want to know about my century gothic?? Like there are different values for display and embeddability attributes.

For arial I have "text" and for century I have "display" for the catagory and for font embeddability I have "editable" for arial and "installable" for century gothic...

Anyway, lett me know..

Hi Burr - re: century gothic - maybe what I will do is send you a couple of small test programs through e-mail and if you can run those to test if your Century Gothic is detected or not with a couple of different techniques.

But I'm not too worried about this issue at the moment, maybe we can wait until next week to try that?

- Michael

Hi Michael,

Re: fonts
What Burr is getting there is the same as on my computer I think I prefer century gothic.

BTW was your server down for most of the day ?

Cheers
~Danny~

Yeah, the whole day I tried MoI Forum but looks like server down :(

Good to see the site back up and running. Downloading V3 beta now. Looking forward to having a play with it.

Nigle / 3DKiwi

Word got out about the V3 beta and everyone tried to get it at once! (Oh to be so popular!)

Hi Michael,

I use flip on the rounded surface and the ring appeared on the right side of the curve but it is still stretch in V direction. I assume as you mention, it because the UVs are orthogonal from the base to target surface.

Since you already have the orientation picker would it be possible to use that to orient the UVs of each surface as you like (as an option maybe), this would provide maximum flexibility, ie. the UVs of base surface could be setup say north/south and the target UVs could be oriented north-est/south-west or whatever. If HISTORY was available one could even modify the original object by whatever mean. (control points, move, scaling, etc) and see in real time the effect on the final object. This would give us awsome creative potential.

Thanks,
Felix

PS. I tried history and it doesn't apply (unless it's me) and I tried to orient indirectly the UVs by rotating the surface, it works only in 90 degree steps sort of speek. Conclusion: history would be a big plus.