PDA

View Full Version : Just a few wishes



hrgiger
02-22-2003, 01:13 PM
Just a few features I would love to see in upcoming versions of Lightwave.

1. Camera switching. I know there is a plug-in to switch cameras during a render but I would like to see it implemented in LW. This should be an easy one.

2. Crowd simulation. Any type of AI for multiple characters in a scene, hopefully with some sort of instancing to save on RAM.

3. Sound tools. Just a basic editor to do things like sound mixing, fades, reverbs, etc.... plus being able to render that sound into your animations.

4. Some sort of interface for joint morphing. You know, this one's been asked a hundred times but you know we'd love it. You know, bend the bone, fix the joint and all in the same screen. Imagine that. Maybe even some sort of corrective lattices, as long as there was a way to setup it up so it was automatic with the bending of the bone.

5. Aura 'lite'. A paint tool that would let us paint across UV seams or directly on the model.

6. Better FX. I think everyone can agree that taking FX to a new level would be a welcome addition. Including and especially, softbodies, liquids, and cloth. Maybe not real flow or anything but better then the current HV solution we have now.

7. Send Lukasz and Pawel a check and put ACS4 into LW. I already own it, but I just feel sorry for those who don't have it. It's an amazing tool.

hrgiger
02-23-2003, 01:05 AM
Something that I was also thinking about today... I think it would be good if we could make any lightwave object into a light source. This would be good for making things like light bulbs or any luminous object that normally casts light. It would be almost like a point light but would cast light off in the direction of the object normals.

Mylenium
02-23-2003, 06:06 AM
1. Camera switching - go for it! I think this would be more useful for previews, though.

2. Crowd simulation - would be nice but won't happen. It would require too much research and programming resources. Also this would be a feature only few people would ever use. I'd vote for smarter particles so we can attach objects to them and have some sort of interaction. Perhaps we will have some day the long promised free version of Massive to play around (possibly after they wrap production on LOTR). Instancing would be very welcome on the other hand.

3. Sound tools - No! Leave that to the Pros. Only support for more formats and other sampling rates would make sense to me.

4. Joint morph/ Lattices - Yes! Perhaps even more really useful deformers (independent from objects, native splines, Clusters...)

5. Painting - well, for those who need it...

6. More Dynamics - Yes! (No comments here, 'cos it's been banged about too often already)

7. ACS - Yes, but on the one condition that they keep developing it and do not stop after they got their cheques.

Mylenium

faulknermano
02-23-2003, 09:04 AM
Originally posted by Mylenium


7. ACS - Yes, but on the one condition that they keep developing it and do not stop after they got their cheques.

Mylenium

money has this certain sedative effect, dont you think? :D

Arte
02-23-2003, 01:16 PM
Originally posted by CIM
Newtek needs to develop it's OWN tools, not just slap some crummy plugin into Lightwave, like they have been doing.

- It's not a real solution, just a shortcut.
- Plugins are generally unstable (look at Motion Mixer or Spreadsheet).

Ppl. need to get a clue and stop requesting the flavor of the month type things. Who wants an old, barely working crap car with a new paint job, when they could have a car that runs well?

IMO it would be far better if the plug in system were more streamlined so that plugins could not bring down the system anyway. But bug fixing is a major no brainer. NT need a kick up the *** on bug fixing. The crashes on certain things are not justifiable close to a year after 7.5.

As for developing their own tools, that is a bit too much CIM. Not even Microsoft has that kind of manpower:)

What do you think of liscencing technologies from others and using those technologies which are ready to go with modifications instead? IIRC they did this with a lot of the app. However the liscence must allow free development by NT and not end up stuck with dead end technology.

BTW I've looked at NT's development staff requests. LW is getting more dynamics, a new interface and more from the requests.

We should get Proton to moderate a list of things that need fixing and adding, moderated by us all on what can and can't be done already.

X

hrgiger
02-23-2003, 03:39 PM
Cim,


What plug-ins are you talking about? ACS4? Have you used this? If anybody does animation, it's a godsend. Rigging characters takes a lot of time but with ACS4, you can do it in about 10 minutes. Your main characters might need more customization in the rig then the 10 minute rig in ACS4 but you can always add onto the ACS4 rig and save it for future characters. But I think the real beauty in ACS4 is for secondary characters in a scene that you don't want to spend weeks setting up all their controls.
XSI has a prebuilt rig to put into characters so I think it would be a good feature for Lightwave.
And if the (crummy) plug-in works, why not slap it into Lightwave?

ToonShady
02-23-2003, 09:24 PM
:D haha I love having CIM around. You basically say things that I have too much humility to express. But I agree, just what is it that is more important to the developer's agenda. For example, is it to create half a dozen more follower clones that only works for a few occasions, or tools that hardly anybody would use? It's nerve-racking to find tools that don't solve problems, or it's broken, or not robust enough to handle wide range of usage, or simply no solution to tackle a problem.

hrgiger
02-23-2003, 10:57 PM
I think Cim, that you assume that everybody wants to spend their time rigging. I personally would rather spend my time animating. It's not a matter of laziness, it's a matter of rigging being so gd boring, not to mention time consuming. Unless you want to get a job as a rigger, I don't even see why anybody would want to spend any more time on it then they had to. I'll take 10 minutes over several hours any day.

I didn't like Saslite at first either. I mean, I bought Sasquatch so I wouldn't have to use Saslite. But a few times, I've given people advice on how to use Saslite for grass or hair and I've gotten some pretty nice results with it. Much better then if we didn't have it at all, and once again, faster. Personally, I'm glad to see a lot less bald models in the gallery.

Sometimes, you have to come up with a workaround to solve a particular problem. Workarounds are everything in 3D no matter what app you have. Plug-ins, even if they aren't integrated, can sometimes be the best workaround there is. The more, the better I say. Don't get me wrong, I would like to see everything flow with lightwave seamlessly and perfectly, but plug-ins are just more options.

Mylenium
02-23-2003, 11:36 PM
Hi everyone,

I think the main problems with plugins are:

1. Poor or missing documentation - even today there are plugins that are not mentioned with a single word in the manual.

2. Non-streamlined/ crappy interfaces - you always have to learn from scratch because everyone has his own view of the perfect interface (I personally think most of those programmers should not be allowed to design interfaces even such gurus as Mr. Worley)

3. Lack of interactivity/ predictability and missing stability - due to limitations in the current API (no full access to all data via plugins, multithreading is not supported by many image filters etc.) and bandwidth issues you just sometimes have to spend too much time on tweaking settings. Whether it is displacement plugins ignoring bones or PFX and HVs crashing right in that moment where you found the right settings - it is sort of an annoyance.

4. Some better integration - currently many plugins require to many steps to activate them (e.g. displacement and pixel filter for SasLite - already to steps). I wish everything was more 1-click. Also this concept of slots for plugins seems a bit outdated. Plugins should be able to appear more like normal functions in corresponding menus.

Mylenium

faulknermano
02-24-2003, 12:56 AM
Originally posted by CIM

Most of Lightwave's features are pure crap because of just being plugins. There are so many drawbacks to them that it practically outweighs their use.

so this can be, more or less, taken to mean 'most of LightWave's features are useless', since lightwave runs off plugins most of the way.

if so, i think you're overstating your case. ;)

faulknermano
02-24-2003, 01:02 AM
Originally posted by Mylenium

1. Poor or missing documentation - even today there are plugins that are not mentioned with a single word in the manual.


yes.. i'm pretty disappointed as well with chapters like ParticleFX. hardly a word of explanation on some parameter (especially the ones i needed to find out).



2. Non-streamlined/ crappy interfaces - you always have to learn from scratch because everyone has his own view of the perfect interface (I personally think most of those programmers should not be allowed to design interfaces even such gurus as Mr. Worley)


well, it's only a fair question of "who then?"



3. Lack of interactivity/ predictability and missing stability - due to limitations in the current API (no full access to all data via plugins, multithreading is not supported by many image filters etc.) and bandwidth issues you just sometimes have to spend too much time on tweaking settings. Whether it is displacement plugins ignoring bones or PFX and HVs crashing right in that moment where you found the right settings - it is sort of an annoyance.


i wouldnt go as far as saying it's about the API that causes such crashes directly. it's up to the programmer to know the API and work with it. i dont use the API directly, so i cant say that it should be improved and that it's 'good enough.' for what it's worth, i know lscript, and i have a couple of suggestions for that. :)



4. Some better integration - currently many plugins require to many steps to activate them (e.g. displacement and pixel filter for SasLite - already to steps). I wish everything was more 1-click. Also this concept of slots for plugins seems a bit outdated. Plugins should be able to appear more like normal functions in corresponding menus.


i agree. if lw8 sticks with this sort of segmented plugin architecture, some extensive work must be made to enable the user overall access of a plug that works on mulitple architectures. not to mention making it easier for lscript-writers (or plugin programmers) to do such things with their code as well.

faulknermano
02-24-2003, 01:27 AM
Originally posted by ToonShady
It's nerve-racking to find tools that don't solve problems, or it's broken, or not robust enough to handle wide range of usage, or simply no solution to tackle a problem.

how about this: my coworker here simply explained to me how he will do a tank tread animation in maya. it was simple: and no expressions needed. he would use Spline IK on a chain of bones that are 'rigid bounded' on the tracks. then he would just move the main parent bone along the spline (he could use a path constraint and animate it via channel). the point is: it is so simple because the whole system is rigged to interact with each other. the Spline IK does most of the work, but the motion path constraint helps in keeping the tracks from jumping.

in a word: there are already several 'basic' features that have been _proven_ very FLEXIBLE, well-rounded, and therefore useful in animation. in this example, Spline IK, in my example, was one of them. but of course, the implementation must allow it to interact with, or be interacted by other elements.

Mylenium
02-24-2003, 02:57 AM
Hi Lernie,

Who should design interfaces? Well, there are people who make a living of just that. You see there is programming and there is usability. Both are quite often contrary and those people try to solve this dilemma. Think of it like you would design a car that looks perfect but you engineer tells you "Can't do it" 'cos from a technical point he very much prefers a simple geometric object. I think especially you in the US will appreciate that. Hell, there even are already laws for accessibility and usability - a thing we in Europe are at least 2 years away. So please NT, pay one of this guys ore at least listen to the users. The same goes for the docs. Here in Germany we have guys called "Ingeneur für Nachrichtentechnik, technische Dokumentation" (engineer for communication technology, technical documentation) that are supposed to know a lot about this. Surely this may not be important in the states because you have no DIN and such but our "Deutsche Gründlichkeit" is not always just for show.

Mylenium

Arte
02-24-2003, 04:43 AM
Originally posted by faulknermano
how about this: my coworker here simply explained to me how he will do a tank tread animation in maya. it was simple: and no expressions needed. he would use Spline IK on a chain of bones that are 'rigid bounded' on the tracks. then he would just move the main parent bone along the spline (he could use a path constraint and animate it via channel). the point is: it is so simple because the whole system is rigged to interact with each other. the Spline IK does most of the work, but the motion path constraint helps in keeping the tracks from jumping.

in a word: there are already several 'basic' features that have been _proven_ very FLEXIBLE, well-rounded, and therefore useful in animation. in this example, Spline IK, in my example, was one of them. but of course, the implementation must allow it to interact with, or be interacted by other elements.


Spline IK for tank treads?

No offence but there is a quantum leap in terms of what you consider easy and what others do. Ugh feels like I just saw the excorsist in a graveyard at Midnight on Halloween.

X

faulknermano
02-24-2003, 10:16 AM
Originally posted by Arte
Spline IK for tank treads?

No offence but there is a quantum leap in terms of what you consider easy and what others do. Ugh feels like I just saw the excorsist in a graveyard at Midnight on Halloween.

X

true. but how would you do it? mesh-through-laid-out-bones? what if you dont want the bones to deform the tracks? i know of two ways to do it in lw, btw, so it doesnt scare me either way. the point is flexibility. after learning such a technique you cant convince me that that isnt a _big_ step forward.

faulknermano
02-24-2003, 10:24 AM
Originally posted by Mylenium
Hi Lernie,

Who should design interfaces? Well, there are people who make a living of just that. You see there is programming and there is usability.

well, assuming we're talking about 3d plugs, would that still be possible. i mean, would worley labs enlist a help of a newtek UI designer... ?? hmm... maybe they would. but is that something they could be bothered with?

Arte
02-24-2003, 10:49 AM
Originally posted by faulknermano
true. but how would you do it? mesh-through-laid-out-bones? what if you dont want the bones to deform the tracks? i know of two ways to do it in lw, btw, so it doesnt scare me either way. the point is flexibility. after learning such a technique you cant convince me that that isnt a _big_ step forward.

What scares me is not the technique, but what it is used for:)

If I wanted to do things the hard way, it can be done with particles you know:) Then again particles can do it easily too now I think about it. Unless we have crossed wires and I have misunderstood what you are after:)

X

Lightwolf
02-25-2003, 10:15 AM
Originally posted by faulknermano
well, assuming we're talking about 3d plugs, would that still be possible. i mean, would worley labs enlist a help of a newtek UI designer... ?? hmm... maybe they would. but is that something they could be bothered with?
Well, may be NT should lay out "strict" rules for the UI and give good examples within LW of how to use them.
Anybody remember the user style guide for the Amiga ? :)

It won't work of course if the included third-party-demo-plugin #1 looks and works entirely different from third-party-plugin #2 from another author (hm, Like Saslite and MD...).