|
I think that sums it up pretty well.
I blame you Brad for throwing the word hypergraph in the mix too early :D
I am aware of the current development efforts and am very much looking forward to seeing them extended over time to give XSI a subtle but important and extended face lift, and I have no doubts it will sooner rather then later reach rigging.
The point about the schematic in its current form is not that it's useless, it's very, very far from being so. The problem is that some -very small- missing bits make into a view that is "easily made useless in some situations".
I find the nostaligic feelings for the old schmeatic out of place to be honest, for the very simple reason that modern rigging makes SI|3D and XSI an apples&oranges comparison.
Never in SI|3D you would as easily automate the creation and setup of as many nodes that so quickly make the use of an inorderable schematic impossible.
All of this could be fixed with one trivial addition of some decent layout tools.
A huge step forward from that (or even replacing that if need be) would be to open up the schematic more for API access.
A last ideal step would be to offer a frontend to managing those things that the schematic so efficiently represents, which are items of a limited range of qualities (scene items and not their properties), materials and the most basic relationships estabilished between those like hierarchies, constraints and material associations.
Of course doing more than that is best left to moondust, but that much is perfectly possible without changing a iota of how the schematic looks and works (especially given the simple approach to constraining in xsi that is always abstracted to an object2object relationship, even if the hook is in a property)
Forgive me if I might appear as an arrogant jerk at this point, but I believe that anybody thinking the schematic is "perfectly fine as it is" right now for rigging are bound to be people who simply have never worked in an environment populated and large enough to deal with hand-overs and revisions, and possibly haven't evolved/translated/extended their rigging tasks with automated procedures and custom components yet. Those things truly invalidate the schematic entirely at the push of a button, unless you're ok with spending literally hours patiently dragging nodes around that could number in the hundreds.
This argument is scarily reminiscent to me of that period when every lightwave user was telling other users that "hey, CA in LW is fine, stop complaining", and we all know what that ended like.
On Fri, Feb 29, 2008 at 8:33 PM, Bradley Gabe < withanar(at)gmail.com> wrote:
As with most other threads, I think we've gotten mixed up into multiple arguments. The first is we have some
people who say the schematic in XSI is useless, and those who disagree and are defending it's uses. It may be useless for some people, but for those who do use it such as myself, it is absolutely critical for production, every hour of every day.
The second argument is about ways to improve the schematic, and the proposed ideas come from user experience with other applications, including si3d, Maya, Houdini. Those who don't use the schematic at all or feel it is "useless" seem to be more likely to want to replace the schematic wholesale with something that looks like Houdini's node graph, or the Hypergraph. I'd rather Softimage worked on their own improvements based on XSI's non-linear workflow which would hopefully consider the modes in the operator stack, among other things.
As far as improvements go, it also seems there are two areas, aesthetic improvements such as node size and layout, and functionality improvements such as the ability to rewire operator connections. Aesthetic changes don't require updates to XSI's core the way many functionality improvements do, which is why we've seen a few changes in aesthetics of the schematic in the past few XSI updates. We haven't seen any major udates to functionality such as the ability to rewire operators, not because Soft doesn't want to, but because what we already have in the schematic is likely a manifestation of XSI's current core architecture.
We should all know by now from what we've seen at siggraph and other user groups that Softimage is already working on core upgrades for next generation functionality. Hopefully, such work will also lead to more radical improvements for the schematic, but in the mean time, some aesthetic improvements could help usability.
|
