So Impressions

[angus_davidson]

Been playing with the 2014 trial and so far its been a pretty stable. The only new feature that's really of concern for me and my students in the new camera sequencer.
For the most part its pretty useful. I can see it been used to help with our previz , and it will simplify our project clutter as you no longer need to have 5 versions of a shot if you have 5 camera angles. The issue with motion blur isnt a biggie for us as with student projects most motion blur is added in post, but I can see how it would pretty much make it a useless feature elsewhere for any final shots. Although I don't think it should be to much of an issue for the devs to Hot Fix.

So far its been pretty solid , but nothing to really excite.

[rray]

Didn't get to do a lot with it yet, tested the new Syflex first, which works as advertised (and very fast!). Most important parameters are now per point so you could for example simulate things like cloth falling into water.

AA in HQ VP now off during interaction - makes sense but should be configurable not to be.

[Lord Futzi Voldemort]

Did anyone manage to set up a simple pedestrian crowd by the workflow described in the manual? I only get a bunch of sliding t-poses.

Edit: Jepp, they actually dehanced it.

[Lord Futzi Voldemort]

I don´t know what to say.
What have you done at Softimage the last year? Poking in your noses?
What is your support good for if you don´t fix those bugs?
The GOG is STILL calculated wrong!

What the .... are we paying you for? Playing? This is a severe bug! Now I'm really, really angry. This is sooo lame. This release is sooo lame. Shame on you Autodesk. You suck big time.

cog.jpg

Also the Strand multiplier bug in the hair op isn´t fixed. A bug for generations it would seem. Ridiculous.

I think announcing " exiting stuff " in softimage was meant as a warning for high bloodpressure patients.

[csaez]

The GOG is STILL calculated wrong!

Seems right to me... How should it work? Perhaps you're looking for the 'Center Of Bounding Box' instead of 'Center Of Geometry'?

[Bullit]

Csaez the pivot it is not certainly in the center of geometry. And in this solid the Center of Bounding Box is the same as the Center of Geometry.

[rray]

COG should be the average point position. Naming is a bit bad, it's more like center of gravity (of the points) so if one face is subdivided it will move over there.

But still, this is a bug since the points are distributed equally on 4 sides in this example, so the COG should be a little in positive Y, but should be centered in the top view.

Code: Select all

CreatePrim "Cube", "MeshSurface"
SetValue "cube.polymsh.geom.subdivv", 2

CreatePrim "Cube", "MeshSurface"
SplitEdge "cube1.edge[2]", 50, False, True, siPersistentOperation
SetValue "cube1.polymsh.splitedgeop.paralleledgeloop", True

this creates 2 identical cubes, cube#1's point COG is at (0,0,0) and cube#2's at (0.5,0,-0.5).. weird, looks like it depends on point order

[Lord Futzi Voldemort]

COG should be the average point position. Naming is a bit bad, it's more like center of gravity (of the points) so if one face is subdivided it will move over there.

But still, this is a bug since the points are distributed equally on 4 sides in this example, so the COG should be a little in positive Y but should be centered in the top view.

It's a completely symmetrical object with the same amount of points on all sides. The COG should be in the world origin. If you take out one edge loop, the COG is where it should be. I have scene, where the COG is so far of, you wouldn't believe it.
It's a confirmed bug. I reported it about 4 Months ago. Was I really the first one who saw this?
The same with Crowd FX. For some reason, now, when you load an actor, a crowd is automatically generated, breaking the rest of the workflow. Either no one tested it, or the reports were ignored.

Anyway, now you know how must be if you do beta testing, you report a bug and it is simply ignored.

So even support is of no use when you're on subscription.

Oh, by the way, it doesn't only happen in COG mode! This makes modeling in softimage completely unsafe.

[Chris_TC]

When was this bug introduced? I use COG all the time and never noticed it in 2012. Seems like a really big issue actually, I'd hate having to find workarounds for something I need all the time.

[Lord Futzi Voldemort]

So, I've not done anything serious in 2014 and have already found three severe issues.
You know why Autodesk has this naming scheme? Because what you buy 2013 is at the earliest usable 2014! If at all, given the trend...

[luceric]

FYI, about COG problem, which might have been there forever: Brent who did the modeling has never noticed it before and I cannot find anything logged about it in the bug database. (In other words, someone should log it)

[McNistor]

Did you move the cube in Object mode or you moved all the verts out of 0,0,0 position?

[Lord Futzi Voldemort]

FYI, about COG problem, which might have been there forever: Brent who did the modeling has never noticed it before and I cannot find anything logged about it in the bug database. (In other words, someone should log it)

FYI:

Mail from Softimage Support 11.1.2013

Quote:

Regarding [CaseNo:07946168.] COG in Softimage 2013 wrong (reported by me 3.1.2013 I believe)

Hi Achim,

Yes, I've noticed it happens in local mode too.
I have added this info in the bug.
Thanks.

Best regards,

Manny Papamanos
Autodesk Support Team

End Quote


So if you can't find anything, does that mean manny was lying? Or did you look in the wrong place? Or does the Autodesk support simply not work? There are many more confirmed bugs that have been left unattended. I guess that is because you guys can't find anything in your company anymore?

[luceric]

So if you can't find anything, does that mean manny was lying? Or did you look in the wrong place?

Alright I got it now, it didn't have COG in the title and I missed it (SOFT-8046)

[Lord Futzi Voldemort]

So if you can't find anything, does that mean manny was lying? Or did you look in the wrong place?

Alright I got it now, it didn't have COG in the title and I missed it

Well, hopefully someone from the softimage team will find it as well so we can have a hotfix as soon as possible, for 2013 and 2014.

[origin]

Ha didn't noticed it until i saw this topic, and im doing heavy subd modelling for 5 weeks now lol. I think the error gets smaller when there are more vertieces in mesh.