PDA

View Full Version : Lightwave rendering engine opinions


papillon68
08-14-2005, 05:39 PM
Hi all, I'm very new to the world of 3D, and i have read on more than one magazine that Lightwave rendering engine is not comparable (in terms of quality) to 3ds max Mental Ray or Brazil rendering engines. They state that LW rendering engine is some steps back and that would need a major upgrade. What is your opinion about it ?
I have seen some LW renderings that seems pretty much photorealistic to me, but not being an expert i cannot tell by myself.
Thanks

Lewis3D
08-14-2005, 07:12 PM
Hi !

Are you SURE you read it right ? MAybe you missed some things in those magazines. LWs BUILT in renderer is excellent in terms of Quality no matter to what you compare it. Only thing where it is little behind is rendering speed (mainlly mc radiosity) but that is subject to change when LW9 hits the street.

telamon
08-14-2005, 09:24 PM
you can find a lot of stupid words in Magazine.

LW's renderer is good and the quality of the pic is got at 90 % by the artist.

Finally, Mental Ray is a fantastic renderer but setting it up is far from obvious. LW's renderer might look technically a bit weaker but so simpler to set up.

My last word, Brazil is an expensive third party renderer, it does not come shipped with 3DS Max.

c-g
08-14-2005, 10:16 PM
LW's render engine is very good but it is showing its age. If you deny that you are living in a vacuum. Take a look at some of the features of the other render engines mentioned. There are a LOT of this LW doesn't do or just does do as well as the others. With that said, not every render needs all the bells and whistles. LW still pulls its own weight but changes are desperatly needed. The developers know it needs to advance and they will make those changes.

papillon68
08-14-2005, 11:15 PM
LW's render engine is very good but it is showing its age.

That's exactly what i've been reading on some magazines, they were defining LW rendering engine obsolete. They even advice to use LW just for modeling, and use other engines for rendering (3ds max for example).
I hope i didn't start a flame war, i'm just trying to understand ...

Lyr
08-14-2005, 11:30 PM
That's exactly what i've been reading on some magazines, they were defining LW rendering engine obsolete. They even advice to use LW just for modeling, and use other engines for rendering (3ds max for example).
I hope i didn't start a flame war, i'm just trying to understand ...

It is indeed showing it's age in a big way. However the changes that the rendering engine is undergoing are very promising, and NT has the right people working on the rendering engine. I don't think it is unreasonable for them to make a comeback in this area. Over at spinquad in the lw news section there are some details of the changes that are taking place.

blaqDeaph
08-14-2005, 11:41 PM
I think that the core of LW's rendering engine has not changed much since 6.5 or so. With that being said however, I'd choose LW's engine over 3ds Max's any day. Even Maya's Mental Ray is rather daunting on first glance (and hence easily puts off beginners). Yes, it's got a lot of options, but half the time, you dont even touch most of them.

Overall, I think that LW's rendering engine offers a good compromise between functionality and ease of use, although the speed could certainly be improved (as many people pointed, updates are promised in LW [9]

KillMe
08-15-2005, 02:01 AM
it might be showing its age slightly in certain areas ( most noteable speed ) but its still one of the if not the easiest to get great results with - with the speed improvements that they keep going on about for v9 that should be solved and then they will no doubt add afew bells and whistles jsut to keep people happy too

if you have doubts about its renderer go to the lightwave gallery on the newtek site if you feel that you can do better if you had afew more options to tweak then maybe you do need something more but if you cant achieve results liek taht in lightwave you surely will struggle to get them out of another renderer

dmaas
08-15-2005, 02:37 AM
In my opinion, Lightwave has the best renderer that doesn't cost extra for each CPU. There are certainly better renderers out there, but you'll have to pay hundreds or thousands of dollars per render node for them.

I think Lightwave's biggest strength is the simple color/diffuse/specular shader system. It's less complicated than most other packages, and still allows you to create realistic surface appearances easily.

Lightwave's biggest weaknesses, rendering-wise, are texture filtering and AA/motion blur sampling. Basically these prevent you from getting the super-smooth "creamy" look of RenderMan-class renderers, and you can't use more texture data than will fit into RAM. Lightwave's high rendering speed compensates somewhat, since you can always render at higher resolution and resize down to get better image quality.

RobertoOrtiz
08-15-2005, 03:28 AM
Hi Dan, it is nice to see you on these neck of the woods.


Thanks for your reply.

-R

IC12
08-15-2005, 10:04 AM
This has been a hot topic for a long time. LW's built in renderer used to be way ahead of the competition. It had features not found on any other medium range (and a lot of high end) engines and then progress just stalled.

3DS Max's engine improves all the time but it's still nowhere near the quality of it's plug-in options (V-Ray etc) and it has only recently got to the stage where it could be compared to that of LW feature-wise. Quality is another discussion entirely.

No matter how you look at it, you can't get better for the money.

colkai
08-15-2005, 10:07 AM
Basically these prevent you from getting the super-smooth "creamy" look of RenderMan-class renderers
Hmm, for some reason, that phrase made me think of beer - shows you how my brain works. :p

Lightwaves renderer is behind the "big guns" in some areas as folks have said, but make no mistake, it is a very good renderer and still preferred in some studios for the look it gives.

Other renderers are now supposed to be coming onto the market with the benefit of unlimited free render nodes, but LW is currently the only package for sale which gives you that and is easy to put together for folks coming to the 3D market.
To be honest, if you're new to the 3D market, I'd be not so concerned with if the renderer was the best in the world, because no matter how good the renderer, if the work is not right, it won't make any difference.
You can look through forums and find stellar images from pretty much any package, but that's more to do with the artist understanding the texturing and lighting.

c-g
08-16-2005, 02:36 AM
That's exactly what i've been reading on some magazines, they were defining LW rendering engine obsolete. They even advice to use LW just for modeling, and use other engines for rendering (3ds max for example).
I hope i didn't start a flame war, i'm just trying to understand ...

Showing its age yes. Obsolete, not even close.

Freak
08-16-2005, 03:09 AM
I think there are two ways to look at this.....


Technically, LW's renderer is behind on some of the fancy terms, familiar in other renderers..
Sub pixel poly displacment, native SSS, and some AA and arguably motion blur etc..

In reality, LW's renderer is able to make nice looking images, that are often
just as good, if not better than those renderers, that cost more and are node locked.
And indeed compared to Maya's scanline renderer, and Max's, and C4D,
LW is far superior........

In comparison to MR i think it's just as good in 99% of cases,
and much easier to get great looking images easily.

Brazil, no doubt excellet...... This was the first renderer, that i could visually
look at, and instantlly know, it was a Brazil render, and superior to LW's......
However, price, node locked, +Max, and difficulty in use, again... Make LW a nice comprimise.

At the end of the Day, there are very few things LW can't make good.....
And it's creators, were named in the 10 most influential CG programmers, because of it.
The new team, is adding new features and seperated the renderer to it's own .dll
making further enhancements, easier....... LW works well, and for the price can't be beat.

3DDave
08-16-2005, 03:29 AM
Max, Maya and Softimage dropped their "in the box" rendering engines in favor of Mental Ray. Humm, why would they do that if their engines were well respected.

Lightwave on the other hand has been kicking out excellent renders for years, look at the list of films and TV shows from Episode 2 to Battle Star Galactica.

plotz
08-16-2005, 04:25 AM
[i]is not comparable (in terms of quality) to 3ds max Mental Ray or Brazil rendering engines. They state that LW rendering engine is some steps back and that would need a major upgrade.

Keep in mind that many magazine reviewers don't really have the time or resources to do in-depth testing and comparisons. This often leaves them in the position of comparing "checklist" features like sub-surface scattering or ambient occlusion between software packages. (i.e. "Package X has SSS and package Y doesn't. This means Package Y is behind the times and showing it's age.") While this does give some indication of a packages features it doesn't really tell you how well it works in a real production situation.

Any time you read a review where a package is said to be inferior look for follow up explanations and facts. For instance if the reviewer says: "I applied a 4K texture map to a model and rendered in package X and Y. Package X crashed six times and package Y performed flawlessly." Then you've got some real results to go on.

If a reviewer makes blanket statements about the quality or usability of features without giving any practical examples then you have to take those statements with a grain of salt.

gerardo
08-17-2005, 01:50 AM
Consider also the "how we use the tools" has a great impact in final results and render times.
btw, in this thread:

http://forums.cgsociety.org/showthread.php?t=197564

we can see different versions of the same scene solved with Brazil, MentalRay, Maxwell and LW



Gerardo

Jarrede
08-17-2005, 02:03 AM
I personally think LW's renderer is showing some serious signs of age. Other packages, although they've "dumped" the native support on the rendering side it's nice that they have a choice when it comes to renderers. LW doesn't offer 3rd party support for the other high end renderers, Fprime is amazing and fast...if they'd only incorporate it, so we'd have network control it can be used on farms....

Other renderers offer photon control, true sub-surface scattering, and motion blur that doesn't look "stepped". While at times it's a chore to setup simple tasks such as radiosity, they do offer more control.

LW's native renderer is still very competent, but for now I'm using a lot of 3rd party plugins to hold it together.. and I think Newtek should give us the option to use other render solutions outside of Fprime.

telamon
08-17-2005, 02:11 AM
It is to be noted that the integration of third party renderers is not always perfect. Look at Maya,. Their SubD cannot be rendered by MR. Therefore they have to avoid SubD in the pipeline. And it is not the only annoyance.

Concerning future, according to Sig words, it seems NT has externalised the render core. Any third party external renderer could plug in LW as long as the future SDK allows it.

Freak
08-17-2005, 02:57 AM
I personally think LW's renderer is showing some serious signs of age. Other packages, although they've "dumped" the native support on the rendering side it's nice that they have a choice when it comes to renderers. LW doesn't offer 3rd party support for the other high end renderers, Fprime is amazing and fast...if they'd only incorporate it, so we'd have network control it can be used on farms....

Other renderers offer photon control, true sub-surface scattering, and motion blur that doesn't look "stepped". While at times it's a chore to setup simple tasks such as radiosity, they do offer more control.

LW's native renderer is still very competent, but for now I'm using a lot of 3rd party plugins to hold it together.. and I think Newtek should give us the option to use other render solutions outside of Fprime.


Umm, They do! Maxwell Render, Fprime, Yafray, Kray, Hikari, are alterantive renderers....

LW9, has stated the .dll has been seperated, which indicates, they are allowing
more render access to 3rd parties, in upcoming releases.....
Renderman, and Maya both have programs like Pointoven, and Beaver Project....
Which can output all relevant files to relevant programs to allow rendering in different
engines... Maya and XSI are still intergrating many of their native features, to work
100% within MR, in the exact same way LW and FPrime are.....

MR and Brazil are dedicated renderers, LW is a complete Modeling, Rendering, animation, program, with much smaller pockets......
1 Node licence of MR or Brazil, is more expensive than LW's 999 / Mac/PC nodes....

Like i said before, it might not have the over hyped native SSS, features....
But iv'e still seen just as many nice images, out of good lighting and gradients in LW,
to give the same of better results of SSS in some other packages....
People seem overly obsessed, on superlatives it's missing, rather than the images it produces.

LW was the first HDRI full floating point, render engine with Radiosity....
And that was only 6.5.. Anyway....
Iv'e seen people make impressive images, in C4D, LW, PovRay, and even MAX.... .EEK!

And as FPrime, is clearly lightyears ahead of the crowd, in terms of instant previews,
and faster radiosity....

Jarrede
08-17-2005, 03:59 AM
Umm, They do! Maxwell Render, Fprime, Yafray, Kray, Hikari, are alterantive renderers....


What about the other popular commercial engines? Are all of those listed above feasible to incorporate into a pipeline?


LW9, has stated the .dll has been seperated, which indicates, they are allowing
more render access to 3rd parties, in upcoming releases.....
Renderman, and Maya both have programs like Pointoven, and Beaver Project....
Which can output all relevant files to relevant programs to allow rendering in different
engines... Maya and XSI are still intergrating many of their native features, to work
100% within MR, in the exact same way LW and FPrime are.....


Pointoven doesn't support camera data transfer last I checked, which is a relevant file to me. Also using those plugins requires extra work integrating the data, which usually comes across as point data transfers that carry large file sizes with them. I'm not discrediting them, I use point oven in my pipline, it's just cumbersome when you have large scenes.


MR and Brazil are dedicated renderers, LW is a complete Modeling, Rendering, animation, program, with much smaller pockets......
1 Node licence of MR or Brazil, is more expensive than LW's 999 / Mac/PC nodes....


I'm tossing pricing aside when I speak of renderers, I got the feeling that the original poster wasn't concerned about price, but quality.


Like i said before, it might not have the over hyped native SSS, features....
But iv'e still seen just as many nice images, out of good lighting and gradients in LW,
to give the same of better results of SSS in some other packages....
People seem overly obsessed, on superlatives it's missing, rather than the images it produces.


I'm not obsessing about features here, I want more control, flexibility, and freedom from a render engine. Often times I have difficulty getting things to look the way I want them and I feel the features I listed could help give me that extra control, flexibility, and freedom.


LW was the first HDRI full floating point, render engine with Radiosity....
And that was only 6.5.. Anyway....
Iv'e seen people make impressive images, in C4D, LW, PovRay, and even MAX.... .EEK!


A good artist can make impressive images with just about anything, it's mainly a matter of how much time it took them to get there, and what kind of sacrifices they made along the way.

They were the first,...but it seems like when they add a feature they leave it alone for a few builds instead of expanding on them.


And as FPrime, is clearly lightyears ahead of the crowd, in terms of instant previews,
and faster radiosity....

I wish Fprime was directly incorporated into LW's renderer, Fprime really is what Viper should be.

PaZ
08-17-2005, 01:05 PM
That's exactly what i've been reading on some magazines, they were defining LW rendering engine obsolete. They even advice to use LW just for modeling, and use other engines for rendering (3ds max for example).
I hope i didn't start a flame war, i'm just trying to understand ...

Beware of software enthusiast, ever.
LW's renderer is quite obsolete, it has too little control over almost everything, its lighting/shadowing tools are very weak and GI engine is the slowest on mid and hi-end area.
Texture filtering and antialiasing/Motion blur are behind the market, too. The only pro-level section is surfacing, LW's surface editor lacks of a node editor (coming soon in 9.0) but it's more versatile and powerful than other software's ones.
This is a realistic snapshot of current LW engine, not much more to say.
Now, you should consider what they're doing for release 9 - it seems they're really revamping the whole thing, finally. It's claimed to be a lot faster and we'll get a custom-version of subpixel displacement, another major absence in current LW's engine.
But the biggest new is engine has been extracted from the core, so soon you'll be able to render with the renderer you like the most, as soon as they'll port them for LW (Vray, brazil, Finalrender, Mental ray and so on). Dont forget we already have two major savers which allows LW to be compared to newer engines, Fprime and Kray.
With a good integration of these (which seems to be near) LW will have a pretty good rendering arsenal.

Paolo Zambrini

bt3d
08-17-2005, 02:58 PM
The quality of the output depends heavy on the artist as many mentioned before. Some like it to have more control, some like it more simple. Anyway, every renderengine has it's strength and weakness. So either you will find a workaround if you want to stick with a special renderer or you will take another one if it fits better. It depends on your requirements/jobs which renderer will be ok for you in most situations. I personaly don't know a renderer which works great in all situations.

One thing I can't support is PaZ statement about the surface editor. I don't think it's more versatile and powerful than other software's ones. For me it's not flexible enough. There are surely packages with a weaker surface editor but definitly some with a much better. But once again: some like it simple, some like it to have more control. It depends on you and your workflow if it will fit to the things you want to do.

It's definitly the best to try different demoversions to build up your own opinion, because it can be only a start to look at some magazine reviews or galleries.

Just my two cents
Tim

RobertoOrtiz
08-17-2005, 02:59 PM
Beware of software enthusiast, ever.


Tone...

Your comments are very good, but the opening statement is a bit too much (go for the troath))


-R

PaZ
08-17-2005, 03:40 PM
Tone...

Your comments are very good, but the opening statement is a bit too much (go for the troath))

-R

I apologize if i sounded polemic there; i just think in every forum there are people (even competent and kind people - not a "judge" itself) which are too much "software-biased", so their comments are to take with a bit of salt.
I use LW daily from many years, i appreciate its simplicity but i could never answer things like "what? Lw rendering not good ? are you mad?" or similiars. It's simply too distant from reality, so i think we should keep an eye on both excess directions. There are too polemic biased people but also too enthusuast ones, and they both are dangerous for the beginner-reader in different ways.

Regarding surface editor: sure there are more powerful solutions, but i find LW's one to be one of the best compromise between "snappyness" and quality. Overall (and beside my opinion - i hate Maya's surface editor, for example), Surface Editor is one of the best sections of LW's rendering pipeline.

Paolo Zambrini

bt3d
08-17-2005, 07:06 PM
Regarding surface editor: sure there are more powerful solutions, but i find LW's one to be one of the best compromise between "snappyness" and quality. Overall (and beside my opinion - i hate Maya's surface editor, for example), Surface Editor is one of the best sections of LW's rendering pipeline.
Ok, I accept your opinion. My opinion differs a little bit, because I never liked for example the layer system much (but hey, maybe I never understood it :). I have used LW from version 5.6 to 7.5 for many jobs over the years (still use it sometimes) and think in particular in conjunction with for example tools like G2 the surface editor is a good solution but I personaly like it to have more access to the elements of a surface/shader and the option to plug them together in a more flexible way. But this is my personal opinion. As I mentioned above it depends on your workflow and what you want to do.
Tim

Jarrede
08-17-2005, 07:50 PM
"But once again: some like it simple, some like it to have more control. It depends on you and your workflow if it will fit to the things you want to do."

It is possible to have both in one package imo, it all depends on how you present the tools to the artist. Keep it simple on the outside, but if you wanted to dig deep enough the complexity/ control is there.

and to comment on the surfacing, I was thinking of ways they could easily keep a similar structure and incorperate a node based system. I think Maya's surfacing is a prime example of a poor way to present a powerful tool to an artist.

PaZ
08-17-2005, 07:56 PM
Maybe we're saying similiar things; deep access is not one of LW's strenghts (in any area of the software) so i agree with you on this. It's just the balance of time needed/quality i like so much in Lw's surface editor, but i agree there are quite a lot holes in it, from clipping not being here to not animatable gradient and non-customizable gradients imput, from bottleneck of shader architecture to deep texture nesting capabilities (which only a node based editor can give), from smoothing system being so dumb to the total lack of a bunch of features.
But... when it comes to mid-complexity materials like the ones i deal with in 99% of the cases, i find it so simple and straightforward. Or maybe i'm so used to it ;)

Paolo Zambrini

bt3d
08-17-2005, 08:25 PM
Jarrede:
I agree with you, it should be possible to have both: a simple way to create surface looks and a more complex if you want to go in depth. But I haven't seen a software package which does this in a good way until now (ok, to be honest, I don't know all packages ;) Maybe LW9 will do this, we will see.

Paolo:
I think you are right, we talk about the same thing and I agree that LW is a powerful and more or less easy tool for many tasks. If it fits to a job there is surely no reason not to use it. And as mentioned before I also still use it sometimes.

Tim

Jarrede
08-17-2005, 08:50 PM
"But I haven't seen a software package which does this in a good way until now (ok, to be honest, I don't know all packages ;) Maybe LW9 will do this, we will see."

I haven't seen a complete one either, just bits and pieces of tools that are done that way. But I think it's totally possible if they put the time and effort into making it so. :)

Freak
08-17-2005, 11:00 PM
What about the other popular commercial engines? Are all of those listed above feasible to incorporate into a pipeline?

I'm tossing pricing aside when I speak of renderers, I got the feeling that the original poster wasn't concerned about price, but quality.

I'm not obsessing about features here, I want more control, flexibility, and freedom from a render engine. Often times I have difficulty getting things to look the way I want them and I feel the features I listed could help give me that extra control, flexibility, and freedom.

They were the first,...but it seems like when they add a feature they leave it alone for a few builds instead of expanding on them.

I wish Fprime was directly incorporated into LW's renderer, Fprime really is what Viper should be.

LW has very good pipelines, for Maya, Modo, and C4D, PRMan, and to a lesser extent Max, & XSI. So you can buy MR with Maya, and use LW with beaver project, or point oven, etc.
So it does fit in...... Would MR ever be made for LW anyway? Thats a question for
MR, not NT.... When Maya and XSI have chucked their renderers, for a dedicated
MR renderer, you can see if these companies with deep pockets, can't keep up with MR,
than LW, won't always have feature X, or option Y....

Considering, you could buy Maya with MR, for the same price as MR for LW, would be.
And then consider both Maya and XSI's MR renderer doesn't support all of their
functions, in the same way Fprime doesn't support shaders in LW.
I don't think really LW's doing too badly, considering...... But hey, we would all love more!

Price is an important, and critical component to business... Having a full time
Rendering team programming MR, and being sold with Maya, and XSI, must give
that an advantage.... You must be realistic with a price vs performance ratio....
If a magazine is comparing oranges, and apples, then the magazine has the issue.

We all want more control and features and speed, and it is coming.....
Keep in mind, that a render engine built by a guy, who no longer works there,
it will take a little time, for someone esle to step up to the plate.

But even then, NT have to update SubD's, Surface Panels, Hypervoxels, etc... etc...
You can't expect a major update to all major modules every update..... It's not feasible...
Anyway i'm noticing improvements, under the hood, which within in a few versions
will allow easier modifications, faster....

I agree, Viper was a missed opportunity for LW, and Fprime still is, until everthing
can be previewed in the preview window.....

But let us not forget, apart from Modo201, FPrime and Lightwave are the only realtime renderers, capable of a lot of things none of the others are, including Brazil or MR......
Iteritve rendering, instant previews... etc... We are ahead in this respect....

Cheers,

telamon
08-17-2005, 11:29 PM
Completely true but FPrime is not programmed Newtek. The only point for which some people think LW is ahead today is a third party plugin partly integrated.

Freak
08-18-2005, 12:00 AM
Completely true but FPrime is not programmed Newtek. The only point for which some people think LW is ahead today is a third party plugin partly integrated.

Nothing wrong with 3rd parties giving us cool tools, especially when we
have the very well respected Steven Worley desinging exclusively for LW...

MR is a third party renderer for Maya, in the exact same way FPrime is for LW....
Brazil, and MR for Max are too......

As i said, if you look at the trend, outsourcing the renderer is common,
as it give Alias, or Avid etc, to concentrate on making better HV or crowd plugins etc.
So it can be really helpful to have Worley, taking some of the design of the renderer...
This way, NT can concentrate on other areas, and look at what areas of the Worley
renderer are popular, and implent only needed changes, to their own renderer....

I think it's sometimes good to be the last one to add feature X,
as it often makes sure it's the best implementation, of the technology.

It's all about balance, LW8 was a little disappointing, but LW9, and beyond,
seem to of gained some direction.... So time will tell where we end up.
No software, will ever do everything better than, everything else....

Cheers,

ThE_JacO
08-18-2005, 01:59 AM
It is to be noted that the integration of third party renderers is not always perfect. Look at Maya,. Their SubD cannot be rendered by MR. Therefore they have to avoid SubD in the pipeline. And it is not the only annoyance.


I'm afraid that's an entirely false statement.
in first place LW's engine doesn't even do adaptive SDS, so even if maya didn't support them it would be a moot point.
what you think of as subD really is a crude geometry approximation obtained by subdividing and smoothing that is hidden from the user, and that handles to the engine a normal set of polygons.
all the apps with MRay integrated support that just fine, Maya also supports MRay's Mental Matter module, which gives access to adaptive subdivision surfaces rendering the limit surface through micropolygons, which is not even remotely possible in any other mass diffused engine other then MRay and RMan compliant ones.


Concerning future, according to Sig words, it seems NT has externalised the render core. Any third party external renderer could plug in LW as long as the future SDK allows it.

I'm afraid people have heard the words SDK, core, open architecture etc. way too many times, but very few people know what's actually implied in integrating a rendering engine into another app.

SUPPORTING a rendering engine is easy, if you didn't mind the performance hit, with a barely decent SDK and API you could write a fully featured exporter of RIB files (RMan support) in a purely interpreted language like any scripting language is.
INTEGRATING an engine is a lot harder then that and it factors many more variables.

last but not least, entirely separating a rendering module from the heart of the application COULD have repercussions of a certain magnitude, it's not always all good.
if it becomes an external engine, no matter how solid the connection to the hub or to layout, things like duplicated scene data VS direct fish&dump, data transfer issues etc. all pop up, when they didn't before.

dmaas
08-18-2005, 04:55 PM
Supposedly Lightwave 9 is going to have adaptively-tessellated subdivision surfaces.

annaleah
08-18-2005, 05:23 PM
Supposedly Lightwave 9 is going to have adaptively-tessellated subdivision surfaces.

Do you have a link to that statement please?
Im searching the docs from newtek and see nothing.

Nevermind...got it.

Phyrea
08-18-2005, 05:56 PM
last but not least, entirely separating a rendering module from the heart of the application COULD have repercussions of a certain magnitude, it's not always all good.
if it becomes an external engine, no matter how solid the connection to the hub or to layout, things like duplicated scene data VS direct fish&dump, data transfer issues etc. all pop up, when they didn't before.
Using the Hub to connect LW to its rendering engine? Come on that's ridiculous -- not at all what NewTek has done by separating the rendering core. What they've done doesn't require sending data back and forth through the Hub or some other connection software. If you open up LW right now with the Hub disabled and press F9 you'll still get a render. The render core has been extracted for 8 months now, the user's experience has only gotten better. Don't you think we would have noticed some extra "connection" software loading when we render? This isn't like Maya and MR - LightWave's renderer is integrated.

pelos
08-18-2005, 07:04 PM
think like this,
why is not other render for LW?

maya has maya software, and maya hardware and now maya mental ray, also tortuga, developed by a single guy.
max? has vray. brazil, mental ray

lw has lw render! thats it!


but the true is that is very easy to set up a scene and render, but the time need to be faster and with more option. another radiosity like messiah studio, that you can chose from monte carlo and photons =)

Freak
08-18-2005, 11:00 PM
think like this,
why is not other render for LW?

maya has maya software, and maya hardware and now maya mental ray, also tortuga, developed by a single guy.
max? has vray. brazil, mental ray carlo and photons =)

As i mentioned before, Lightwave has Yafray, Kray, Fprime, Maxwell Render Hikari, at the moment.....Maya has Mental Ray, I don't see what your are talking about!
(Fprime is also made by a small team, as is Maxwell Render)

Maya also has a crap, sofyware scanline renderer, a crap hardware renderer,
and now a nice half intergrated version of MR......

And the whole reason, NT have removed the Renderer to it's own DLL
is to make it easier to updater and implement changes to it's....

bt3d
08-19-2005, 09:53 AM
Freak, although I don't use Maya I think your statement is a little bit onesided. To name some examples:
Maya has also connections to finalrender, vray (is on the way), turtle and last but not least there is a plugin for Maxwell, which exists longer than the one for LW. And finaly it has a better support for renderman renderer (Ok, LW also has one or two plugs to export to rib, but I think you will agree that the maya connection might be more comfortable). Anyway, don't get me wrong, I don't want to start any flamewars. As I mentioned before I don't use Maya at the moment, and I don't see any reason why I should use it in future for the moment. But I definitly don't like it to see only onesided statements, let's keep it objective.
Tim

ThE_JacO
08-19-2005, 09:57 AM
there's been heaps piled upon heaps of mis-information to tell the truth.
from hints that maya's MRay integration doesn't support SDS to saying that MRay in XSI doesn't support all of XSI features (when MRay in XSI IS THE rendering engine for the app, including shared scene data, which is not true for any other application around).
posts over posts of people talking of isolated dlls(which makes no sense as a dll is a purely windows thing and it simply means you isolated part of the elements into a library at compiling time), SDK, core issues and all that but who, obviously, have no knowledge whatsoever of a dev process.

the funny thing is that the most fervant people in comparing and throwing out one-way statements are also those who, just as obviously, never used the X "rival" applications they are featuring in the comparison.

luckily there are still well equilibrated users and opinions who don't mind sticking to facts, and still manage to come out positive about the whole deal.

and with this I'm out of this thread before I ignite something.

PetterSundnes
08-19-2005, 11:50 AM
As I see it, LWs "simple and limited" render technology is also its strength. Having used XSI for some interior lightmapping, it is great to be able to create complex shaders and let some models cast GI light, while others only recieve GI and such, but in the end you might end up fiddling with settings rather than getting work done.

When the renderer hits a limitation you can not work around, then it would be nice if it was more flexible and configurable such as MR.

Btw, a coworker is a betatester for the Max renderer called BusyRay, its extremely simple in terms of settings, which is useful sometimes:
http://www.busyray.com/images/uishot.gif

Chuck Baker
08-19-2005, 02:03 PM
last but not least, entirely separating a rendering module from the heart of the application COULD have repercussions of a certain magnitude, it's not always all good.
if it becomes an external engine, no matter how solid the connection to the hub or to layout, things like duplicated scene data VS direct fish&dump, data transfer issues etc. all pop up, when they didn't before.


there's been heaps piled upon heaps of mis-information to tell the truth....
posts over posts of people talking of isolated dlls(which makes no sense as a dll is a purely windows thing and it simply means you isolated part of the elements into a library at compiling time), SDK, core issues and all that but who, obviously, have no knowledge whatsoever of a dev process.

...
luckily there are still well equilibrated users and opinions who don't mind sticking to facts, and still manage to come out positive about the whole deal...

Your comments seem to indicate that you are not in possession of all the facts either, at least as regards the separated renderer in LightWave 3D. We are long past the stage of "If it becomes..."; the renderer has been extracted from Layout and this extracted renderer shipped as of version 8.2; as cited elsewhere in the thread, that would be about 8 months ago. It has none of the issues that you suggest with regard to memory and communication, and its operation has been so transparently unchanged from the user perspective that in fact this major core restructure drew no public comment, with the overwhelming majority of users not aware of the change until we publicized it just a few weeks ago during SIGGRAPH. The folks who have mentioned status as a DLL are correct - on Windows platform it appears as a DLL file, and of course on Mac as an SHLIB file.

We've documented this and other core work for the 8.x and 9.0 cycle here:

LightWave's Future (http://www.newtek.com/lightwave/lwfuturedev.php)

red_oddity
08-19-2005, 02:50 PM
Maya also has a crap, sofyware scanline renderer, a crap hardware renderer,
and now a nice half intergrated version of MR......

Horse crap, just because some people don't have the creative talent to create pretty pictures with as few clicks as possible with a standard rendering engine doesn't mean it's a shitty render engine (i've done some very nice renders with it and the hardware renderer has often been a blessing.)

How did that saying go? It's a poor craftsman that blames his tools?

Chewey
08-19-2005, 03:42 PM
Your comments seem to indicate that you are not in possession of all the facts either, at least as regards the separated renderer in LightWave 3D. We are long past the stage of "If it becomes..."; the renderer has been extracted from Layout and this extracted renderer shipped as of version 8.2; as cited elsewhere in the thread, that would be about 8 months ago. It has none of the issues that you suggest with regard to memory and communication, and its operation has been so transparently unchanged from the user perspective that in fact this major core restructure drew no public comment, with the overwhelming majority of users not aware of the change until we publicized it. The folks who have mentioned status as a DLL are correct - on Windows platform it appears as a DLL file, and of course on Mac as an SHLIB file.

We've documented this and other core work for the 8.x and 9.0 cycle here:

LightWave's Future (http://www.newtek.com/lightwave/lwfuturedev.php)

Thanks for taking the time to set the record straight with "correct" facts.

Freak
08-19-2005, 11:20 PM
Freak, although I don't use Maya I think your statement is a little bit onesided. To name some examples:Maya has also connections to finalrender, vray (is on the way), turtle and last but not least there is a plugin for Maxwell, has one or two plugs to export to rib, but I think you will agree that the maya connection might be more comfortable).But I definitly don't like it to see only onesided statements, let's keep it objective

I think i'm being objective (but then we all do) I agree Maya is setup better for
3rd Parties to program, and is designed for large studio's to add a programing team
too, to add features in Melscript or C++.... It's architecture is better setup for external
renderers...... (Because from the start, they knew, they could not afford to play catch-up
to more established renderers) I think the original poster i was reffering too, was being
one sided, Lightwave does allow other render engines to be made, Maxwell, and FPrime
and the others i mentioned are proof of this, Newtek have acknowledged that this is
an area they are aware of, and the connection is being isolated and improved to achieve this.

Unlike what "thE_JacO" seems to think Maya and Max do suffer from many of the
same SDK disadvantages that LW does, And indeed Maya7 again integrates more
of it's native features to work in MR, in there latest release....
In the same way NT, made it possible for Worley, to render Volumetrics etc...

And again"thE_JacO" seems to make "no sense" when talking about the .dll
isolating the renderer, shows that NT have already taken steps to improve
the crits, that people are laying..... Seperating the renderer, that for 10+
years has been added to and threaded amongst the actual programs
is not just an overnight job, but it's already done which means adding
to it, or improving it's functions, are now much easier to achieve......

Better AA routines, Faster Rendering, Improved Vector Blur, are some of
the latest improvements, and OGL2 All within 8.2 - 8.5
Optimized platform specifics can now be made to Mac/PC/Intel/Amd/64bit etc...
You don't get those minor updates in Maya... You pay for them in the next full version.

Would FinalRay, or Mental Ray, or any other major Renderer, be ported to LW, even if this
connection and SDK was improved to allow major integration possible anyway?
Brazil, is the only engine i have seen make public statements, that one day it would
consider writing a LW version at some stage. (this was 2+ years ago)

At the end of the day, All programs will have advantages somewhere.....
I don't believe LW's renderer is in anyway bad... (It's not perfect)
For those saying LW is missing out, and can't reach the quality of FR, or Vray,
or MR etc.... Should show use examples.

But we are all entitled to our opinions, and thats mine....

(i've done some very nice renders with it and the hardware renderer has often been a blessing.)
How did that saying go? It's a poor craftsman that blames his tools?

Huh? I didn't say anything about not been able to produce, nice images...
As iv'e stated already in this thread, All renderers can make nice images...

But MR was added to Maya, to compete with the likes of FR, LW, Brazil etc...
because their scanline renderer's, like Max's is generally of lower quailty....
MR is a nice renderer, Brazil is too, Maya's other two, are not as nice....

Jarrede
08-20-2005, 07:13 AM
LW has very good pipelines, for Maya, Modo, and C4D, PRMan, and to a lesser extent Max, & XSI. So you can buy MR with Maya, and use LW with beaver project, or point oven, etc.


"very good" is your opinion. I'd say it's "workable"



Price is an important, and critical component to business... Having a full time
Rendering team programming MR, and being sold with Maya, and XSI, must give
that an advantage.... You must be realistic with a price vs performance ratio....
If a magazine is comparing oranges, and apples, then the magazine has the issue.


Is this statement based on my previous points? I find this somewhat of an insult that you would assume I don't understand the importance of price and where it falls into trying to market your product. I stated I wasn't considering price because the original poster didn't mention anything about price, only quality. I understand what your point is, but if it's brought on from my "price" statement your assuming I'm completely ignorant on the subject because you may have misread my post.


We all want more control and features and speed, and it is coming.....
Keep in mind, that a render engine built by a guy, who no longer works there,
it will take a little time, for someone esle to step up to the plate.

But even then, NT have to update SubD's, Surface Panels, Hypervoxels, etc... etc...
You can't expect a major update to all major modules every update..... It's not feasible...
Anyway i'm noticing improvements, under the hood, which within in a few versions
will allow easier modifications, faster....


Hey, if the dudes gone, that's not my problem. Why is it not feasible? WHY? I'm not expecting an update to ALL modules, just it's animation tools and rendering. They don't have to completely re-vamp the whole thing, there's tons of features I'd like to see. I don't see why it's such a problem to demand more from the software we use.



But let us not forget, apart from Modo201, FPrime and Lightwave are the only realtime renderers, capable of a lot of things none of the others are, including Brazil or MR......
Iteritve rendering, instant previews... etc... We are ahead in this respect....

Cheers,

Ahead, but behind at the same time. Fprime won't render over clusters, I love it's speed, but it's still showing me the same quality renders for the most part.

LW's best renderer, subsurface scattering plugins, Hair plugins, rigid body and soft body dynamics are all 3rd party plugins integrated into a weak foundation. It's time to demand more, not come up with excuses "oh well the guy is gone" am I supposed to pay just as much for the hacked upgrade becasue of that? Do they expect us to?

LW9 is going to determine if I will continue to use LW, after 7 years, thats what it's come to with me.

Freak
08-20-2005, 09:30 AM
LW's best renderer, subsurface scattering plugins, Hair plugins, rigid body and soft body dynamics are all 3rd party plugins integrated into a weak foundation. It's time to demand more, not come up with excuses "oh well the guy is gone" am I supposed to pay just as much for the hacked upgrade becasue of that? Do they expect us to?

LW9 is going to determine if I will continue to use LW, after 7 years, thats what it's come to with me.

If you don't like Lightwave, or would prefer to use the competition, why wait till Lightwave 9?
Go now, there is nothing to stop you, i'm sure it's easy to find a complete 3D package
that has the feature set, and ease of use combined with price or LW?

999 CPU's Mac/PC Modeling, Rendering, and Animation..... For Cheaper than Brazil.
which is just a 3rd party renderer for Max....
Emmy's and Academy Awards credits.... And a great community,...

As someone who just bought Maya7, i'm really disappointed, no Viper for FX,
no Fprime for instant everything, and new stuff like the ToonShader....
Will not work with the Hardware, Vector, or Mental Ray Renderer.....

Like so many other MayaFX, it will only work with the Software renderer.
So if you think the grass is greener, more power to you..
It's a constant game of leap frog....

And for the first time, in a while LW is making some inroads into
If you think after 7 years of your expertise, LW is not for you,
why not move to another application?

We could argue all day, but really there is no point....
I think people get caught up with other people have,
and forget would they already do have...
FPrime increased my workflow speed, by 1000%..
Modo201, is so far the only one to catch on to the idea.
And it will take many years to mature, beyond LW
And it costs more too....

Shade01
08-20-2005, 10:07 AM
This conversation gets worse as this thread gets longer.

3...

Jarrede
08-20-2005, 10:35 AM
If you don't like Lightwave, or would prefer to use the competition, why wait till Lightwave 9?
Go now, there is nothing to stop you, i'm sure it's easy to find a complete 3D package
that has the feature set, and ease of use combined with price or LW?


I never stated I didn't like LW, I love LW, it's paid my bills for a long time. It's just showing signs of age recently, and I wanna give LW9 a chance before I decide I wanna replace it. They've got a nice feature list going, I just hope it's implemented well.


As someone who just bought Maya7, i'm really disappointed, no Viper for FX,
no Fprime for instant everything, and new stuff like the ToonShader....
Will not work with the Hardware, Vector, or Mental Ray Renderer.....


for me, I had to get over the "Maya hump" after bout a few months, it started to grow on me. IT's biggest problem in my opinion, is it lacks consistency in the way they present tools. Painting skins is as easy as pie, yet, building a box and adding sections requires numeric inputs.

After a year and a half using Maya at a game company, I came to the conclusion that while at times simple tasks can be cumbersome, and I wouldn't recommend anyone try modeling with it straight outta the box, it truley has some awesome power under that hood.

Have you played with the rigid and soft bodies? The Character tools? How bout the "undo anything" undo ? :)


Like so many other MayaFX, it will only work with the Software renderer.
So if you think the grass is greener, more power to you..
It's a constant game of leap frog....


I'd never commit myself to just one package right now, or, I've never found a single package good enough to commit to, I haven't spent a lot of time with them all though, I just like to keep my options open. It's not the software I use that my employers valued, ileast...a few of them did...



And for the first time, in a while LW is making some inroads into
If you think after 7 years of your expertise, LW is not for you,
why not move to another application?


I already have moved to other packages due to the current packages inherent disadvantages that I felt was holding my work back. But I see no reason why their can't be one package that makes me happy, it's possible, I can't think of any technical limitations right now, it's all money and time.


We could argue all day, but really there is no point....
I think people get caught up with other people have,
and forget would they already do have...
FPrime increased my workflow speed, by 1000%..
Modo201, is so far the only one to catch on to the idea.
And it will take many years to mature, beyond LW
And it costs more too....

I'm not caught up with what anyone has, nobody has Modo 201 right now and I'm looking forward to it. Fprime did increase the heck outta my workflow, if I switched, I know I'd miss it. I'm not forgetting what I have, I use LW everyday, that's why I'm hear pleading and discussing my needs to make it better. If I didn't care, I wouldn't be here right now. I get no satisfaction from bashing software just to do so. The past year my LW experiences have felt as if I'm constantly fighting with workarounds and sacrifices to get the results I want, that I've been fighting and wasting time with technical issues that shouldn't be there, I don't feel challenged anymore, I feel as if I've hit a wall in certain areas, while areas have been improving there's fundamental things that just keep getting overlooked with every new build. I'm not going to limit my work and my toolset to one thing just because I know it almost inside and out, and have been using it for so long, I want to keep my options open.

I think people are reluctant to try new things because they fear and they know that their skills will be reduced to almost nothing, they've come so far with knowledge of one package, just to throw it all away and start over. But they fail to realize that besides the fact they're limiting themselves, the true value in that person is the ability they used to learn and get to where they are, not what they already know. It's kinda like an artist putting his value on a piece of work he did, when the true value is the talent he has to create those images. I'm not saying your that type of person, I just stating that with some people it's difficult to convince them that we deserve better, and we should demand better because by admitting there's faults in what we use, the thought that we may have been cheating ourselves out of greater experiences just doesn't settle right with most people. ...I donno if that made any sense at all...if it didn't, I apologize.

anyway's, I don't mean to sound anti-lightwave, I just think if we don't demand a lot from the people who provide us with these tools...well...were not gonna get a lot, because if they could sell us LW 5.6 right now, they would. :) When people ask me about LW's renderer I tell them it's very robust, but showing signs of age, and could use a lot of improvement.

we've got some nice toys, but...most of them are 3rd party which makes me wonder where Newtek has their priorities right now.

So I'll re-phrase my opinion, lets throw all other packages aside and just focus on LW, it's difficult to look at problems objectivley when they seem to be global 3d issues and other packages have tired or come up with a solution. It's not that I'm claiming "Hey! X software has this feature! LW needs it!" I'm saying something more along the lines of, "For the life of me I spend a heck of a lot of time getting skin to look like skin, rigging characters, and getting flesh to behave more like real flesh, and I've got some suggestions that may help me, along with others on this...what kind of solutions have other people come up with? is it working? and is it planned for the tool I use? If not, why not?"

Jarrede
08-20-2005, 01:48 PM
You know what I've heard Freak?, aside from my own opinions, and I'm not saying I agree with it, but I saw his point. I've heard directly from the mouth of a president at a well known effects company say they won't consider using LW because if they don't like it's renderer, they're screwed, they won't have any feasible options to run to, and they can't use other render solutions they're already familiar with. They have a large programmer staff writing custom shaders to pipe into their render solution, I couldn't tell him that LW offered that kind of flexibility. I tried convincing him it's native renderer isn't that bad, but he couldn't make the commitment to something knowing that the possibilities are limited. I know you've mentioned many others, but do all of those really offer a viable solution that can meet a companies needs?

If Newtek has mentioned they recognize the issue, and their planning on putting some time to it, then you've got no argument with me.

But, I must say, he would have never even thought about LW unless I had shown him Fprime.

Chuck Baker
08-20-2005, 02:40 PM
I've heard directly from the mouth of a president at a well known effects company say they won't consider using LW because if they don't like it's renderer, they're screwed, they won't have any feasible options to run to, and they can't use other render solutions they're already familiar with. They have a large programmer staff writing custom shaders to pipe into their render solution, I couldn't tell him that LW offered that kind of flexibility. I tried convincing him it's native renderer isn't that bad, but he couldn't make the commitment to something knowing that the possibilities are limited. I know you've mentioned many others, but do all of those really offer a viable solution that can meet a companies needs?

If Newtek has mentioned they recognize the issue, and their planning on putting some time to it, then you've got no argument with me.

But, I must say, he would have never even thought about LW unless I had shown him Fprime.

LightWave has been used as an element in pipelines that end at every other professional renderer available for years and years. LightWave's renderer has also been the target renderer in multi-application pipelines for just as long, putting pristine photo-real images on the motion picture screen, on television and in high-end commercial work, including in several Oscar VFX winners and plenty of Emmy winners. You mention up above that you personally feel the solutions that have made this possible are "workable" rather than "very good"; but a good many folks find them more than merely workable and the capabilities will have lots of room to grow with the architectural and SDK changes we have and are continuing to implement relative to rendering and pipeline integration.

And again, there's no "if" to it - we've described what's been done in the 8.x series and what is being worked on for the v9.0 release in the link I provided in my previous post. I hope you'll review it and familiarize yourself with what we've done so far and have planned to do, and any feedback you'd like to provide on features you'd like to see will certainly be appreciated, here or at lwfeatures (at) newtek (dot) com or on our feature request section on the NewTek forums.

Zarathustra
08-20-2005, 03:05 PM
This is a long thread. Sorry, I haven't read the whole thing. I see Shade01's post and figure why bother, it's probably nasty back there. Anyway, I think it's important to consider when reading LightWave's renderer has also been the target renderer in multi-application pipelines for just as long that the 999 node rendering allowed is a HUGE reason why LW is a rendering option in a lot of pipelines. You can't tell me it's used solely due to quality. The quality is pretty decent but it's the nodes that sell it. If it had less than 3 nodes/license then it's usage might drop (although the new lower price would ease that a bit).

9's render engine is said to have all kinds of improvements. That's great, because if other options spring up that allow huge numbers of nodes and are comparably priced then quality will become the major selling point. Ok, ease of use is a big deal, too.

Jarrede
08-20-2005, 03:40 PM
LightWave has been used as an element in pipelines that end at every other professional renderer available for years and years. LightWave's renderer has also been the target renderer in multi-application pipelines for just as long, putting pristine photo-real images on the motion picture screen, on television and in high-end commercial work, including in several Oscar VFX winners and plenty of Emmy winners. You mention up above that you personally feel the solutions that have made this possible are "workable" rather than "very good"; but a good many folks find them more than merely workable and the capabilities will have lots of room to grow with the architectural and SDK changes we have and are continuing to implement relative to rendering and pipeline integration.


You don't need to try and sell me on what kind of "pristine photo-real images" LW has created, I've been using it as a professional in games, film, broadcast, and themepark rides for almost 7 years now, it's not as long as most of the vets here, but it's enough to know about it's "pristine photo-real images" Lightwave is a good package, it just hasn't met all of my needs recently so I've felt the need to express my opinions, and listen to others. Awards and Oscar's carry very little weight to me when it comes to choosing what software I'm comfortable with. It's great to hear that the SDK can open more doors, but am I to be pleased with those current solutions simply because "a good many folks" may not share the same opinion? Do I really come off as that impressionable? I also know a "good many folks" who share similar views to mine, ones who are also recognized by the industry, but I'm not going to use it as a persuasion tool, nor am I going to be satisfied just because others are. Don't confuse talented artists and companies that win awards for perfect software. It takes a talented user to put those "pristine photo-real images" up on anything, far more than any perfect software can do it. Most of the time it just makes the trip a bit shorter.

The current 3rd party application bridges in terms of animation transfer are "workable" to me because they function, but I think they're a pain to use at the moment and have many areas that are in need of improvement. I'd rather put my trust in that solution to Newtek than a 3rd party company with questionable future support. People most often have to pay for these solutions with the concern of functionality, where as, if it came directly from Newtek a certain level of functionality can be expected. This would also help protect companies investments in their software.

Point Oven uses pointdata to get information across, much like the old messiah did, this can become cumbersome when working with large scenes, or multiple characters. The files can get too large for most machines ram capabilities depending on the length of the shot. It also doesn't transfer camera data, and they're very limited to what types of information is translated. To my knowledge, the Beaver Project operates with a similar type of data. I understand the technical limitations involving the data transfer right now, and know it's a large task to ask of. Idealistically I'd like to not even have to consider an outside application-bridge solution. The main reason this is done is because other tools are better equipped to handle the job in time, so much better equiped that even the time it takes to transfer data is worth it.

So basically what you've told me is, Newtek doesn't currently plan on creating this bridge themselves, but because of the new SDK changes, 3rd party developers can potentially come up with better solutions.

While this shows promise, leaving it up to 3rd parties is not always the best solution.

The real topic at hand here isn't incorporating other packages animation etc. back into LW, but offering more flexible rendering solutions, and plans to make it possible for companies that produce MR, and Brazil as an alternative to LW's renderer, along with adding new possibilities and better quality to LW's native renderer. If your telling me bringing the SDK to the front is going to help make those possible, I'm glad to hear it.


And again, there's no "if" to it - we've described what's been done in the 8.x series and what is being worked on for the v9.0 release in the link I provided in my previous post. I hope you'll review it and familiarize yourself with what we've done so far and have planned to do, and any feedback you'd like to provide on features you'd like to see will certainly be appreciated, here or at lwfeatures (at) newtek (dot) com or on our feature request section on the NewTek forums.

Well, after reading the feature list again, it looks promising given those features are incorporated well. While it's great to see some old weaknesses in LW have been addressed like OPENGL, a node based material editor, and finally SDK (which to me means easier outside implementation of custom tools) unless I missed something on the list, the render issues I was hoping would be resolved were not mentioned, like better motion blur (I recently spoke to a fella who claimed they were considering searching for an outside render solution because LW's motion blur couldn't meet their needs), subsurface scattering, an RSA buffer saver output interface with Z depth outputs etc, better EXR and RLA support, and independent ray recursion limits for different types of rays, the increase in speed is very nice, but honestly LW's rendering speed doesn't bother me as much as the other issues. I'll be sure to leave the requests on Newteks site, in the meantime I've posted a few topics in the "how to improve lightwave forum" along with other valid requests from different users.

So, with your knowledge of LW's future, what would you tell the president of the company I spoke to, about his concerns implementing LW in their pipeline?


9's render engine is said to have all kinds of improvements. That's great, because if other options spring up that allow huge numbers of nodes and are comparably priced then quality will become the major selling point. Ok, ease of use is a big deal, too.

Very big deal, and I agree. A while back I was at a company considering switching some of their pipline to a different package, and the only feasable render solution at the time was 1000k a node, per CPU. Having a stack of 350+ cpu's, it would have cost them 350,000, a price that wasn't worth it. The software seller wasn't willing, or unable to budge on the price because it was a 3rd party render solution and the prices were controlled outside their company.

Since then, it's changed, but I knew it wasn't even worth bringing that number to my boss.

Really for it's cost, LW's native renderer is still one of the best...but I'd like better. :)

Personne
08-20-2005, 03:52 PM
This is a long thread.

Yes and with long text too.

Jarrede
08-20-2005, 03:57 PM
Yes and with long text too.

how do you make it shorter? :)

I've been up all night, so...I see like sentences and words...if they don't make sense, I'll fix'em tomorrow.

and outta the blue...

photorealism bores me.

really, I don't want to sound like an anti-lightwave person, but I know now that my statements will probably appear that way...

Something that I'm interested in people elaborating on,

"The actual core of the LightWave rendering engine has been replaced with a modern implementation that reflects some of the latest developments in the CG industry. This new foundation allows for expansion of new rendering technologies, and lays a very strong foundation for the future."

Does this mean that Newtek may possibly rely more on 3rd party developers to create these expansions now that it's become available to them? How early can we expect to see some of these new render technologies outside the raytracing and speed improvements?

Does this mean LW9 is now open to outside 3rd party renderers if they choose to support it? The statement can mean many things...

answers to these questions may contain what I've been hoping for...

Chuck Baker
08-20-2005, 10:41 PM
Anyway, I think it's important to consider when reading that the 999 node rendering allowed is a HUGE reason why LW is a rendering option in a lot of pipelines. You can't tell me it's used solely due to quality. The quality is pretty decent but it's the nodes that sell it. If it had less than 3 nodes/license then it's usage might drop (although the new lower price would ease that a bit).

9's render engine is said to have all kinds of improvements. That's great, because if other options spring up that allow huge numbers of nodes and are comparably priced then quality will become the major selling point. Ok, ease of use is a big deal, too.

If LightWave's renderer couldn't produce the required quality, no number of free render nodes would have gotten it to where it has gotten to in the industry. The combination of rendering quality and free render nodes has indeed been a foundation of the product's success. That does not make it inappropriate to discuss the quality on its own, and again, for the group of folks reading this discussion, there are plenty of folks in the industry who have an opinion different than what you express - they find the quality better than just "pretty decent" and are not settling for it, but choosing it over the other available options.

Jarrede
08-20-2005, 11:01 PM
If LightWave's renderer couldn't produce the required quality, no number of free render nodes would have gotten it to where it has gotten to in the industry. The combination of rendering quality and free render nodes has indeed been a foundation of the product's success. That does not make it inappropriate to discuss the quality on its own, and again, for the group of folks reading this discussion, there are plenty of folks in the industry who have an opinion different than what you express - they find the quality better than just "pretty decent" and are not settling for it, but choosing it over the other available options.

To them it's above the competition for their needs, I see your point, but that doesn't rule out the possible need to improve it. Just because I think it's the best offering at the time, doesn't mean I don't feel that adding more isn't needed, or that the once dominant native renderer in the 3d industry is in danger of becoming just another good renderer for the price if advancements aren't planned in it's future.

I'm not sure what I got from that statement about lw9's renderer, it can mean many things...I'm not going to elaborate and assume, only to have my expectations let down.

K. Scott Gant
08-20-2005, 11:02 PM
Getting back to the original poster with the original question. In my humble opinion (IMHO) the best renderers I've ever seen, consistently that is, are Pixar's Renderman and Lightwave. Yes, to me Lightwave can be held up against Renderman. Sure, it may not have all the bells and whistles the new renderers have and it's a little slow now compared to them, but I'm talking sheer quality.

Having said that, the main men behind the quality of Lightwave's renderer have gone on to build another application, but only time will tell if the new rendering engine (which Allen has already said will eventually be able to use as a renderer for other apps...dont' know if that would involve a plug-in for Lightwave or Maya or Max etc etc...) will be as good as if not better than what we have out there now.

c-g
08-20-2005, 11:15 PM
So basically what you've told me is, Newtek doesn't currently plan on creating this bridge themselves, but because of the new SDK changes, 3rd party developers can potentially come up with better solutions.

While this shows promise, leaving it up to 3rd parties is not always the best solution.

That is the ONLY reason why Maya is ahead of the pack. Because instead of creating a really good application ,they created an open applications. How many big studios (since we seem to keep using them as the ultimate user) are using Maya with no custom shaders, no custom plugins or ready for this...Not one SINGLE mel script that didn't come with the base program? Now with Maya the 3rd party is usually the end user because most of the time it requires more part artist/part developer. If someone is sitting down banging out those crappy NURBS heads we see so often, no you don't need to program. In LW if you are banging out SDS heads, quality aside, you don't need to be programming either. Back to rendering the current (I shan't call them new anymore) are working on the render engine. I can't believe how many people are saying how poor something is this far before its release for LW9. NewTek knows there is room for improvement, I'm sure they hear about it every day.

Jarrede
08-20-2005, 11:35 PM
Back to rendering the current (I shan't call them new anymore) are working on the render engine. I can't believe how many people are saying how poor something is this far before its release for LW9. NewTek knows there is room for improvement, I'm sure they hear about it every day.

PLease not that I stated "leaving it up to 3rd parties is not ALWAYS the best solution."

I understand your point about Maya, and agree.

I'm not claiming that other packages have the solution here, I was trying to entertain different ideas... nor did I claim LW's renderer was poor, LW9 wasn't in the discussion, the current LW renderer was.

I'm also not assuming their so naive that they believe there's no room for improvement, I'm stressing that improvement should be made in areas I, and others find valuable.


To sum it up, IMO, LW's current native renderer is good, reliable, and 85% of the time gives me the results I want, But I think recently it's been loosing a bit of ground, it's by no means poor, or crap, it's just not adapting fast enough recently to satisfy my needs.

LW9 may change my opinion completely, but since the details are still hazzy I only have 8 to base my judgment from.

Chuck Baker
08-20-2005, 11:42 PM
but am I to be pleased with those current solutions simply because "a good many folks" may not share the same opinion? Do I really come off as that impressionable? ... I also know a "good many folks" who share similar views to mine, ones who are also recognized by the industry, but I'm not going to use it as a persuasion tool, nor am I going to be satisfied just because others are.

Expecting you to change your view was actually not the point of my comment. You expressed your opinions and discussed similar opinions you've heard from others, and I felt the group reading the discussion should hear that there are other opinions as well.

As for persuasion tools, it seemed to me that you are expressing your point of view as regards aspects of the usability of LightWave in your posts, and that the story of "the head of a VFX company that you know" was intended as persuasive support for that point of view. Given that, that's why I felt it was worthwhile to mention that there are alternative views.

That doesn't mean that NewTek isn't going to take your comments or the reservations of the party you describe into account - but in public discussions there is more to account for than simply accepting the feedback, as I'm sure you must be well aware.


Don't confuse talented artists and companies that win awards for perfect software. It takes a talented user to put those "pristine photo-real images" up on anything, far more than any perfect software can do it. Most of the time it just makes the trip a bit shorter.

I'm very aware that it is the talented and inventive artists who earn the awards - and there should be no onus on mentioning that our product was chosen for use by such folks for their award-winning projects nor any suggestion that we lack the proper perspective on this.

So basically what you've told me is, Newtek doesn't currently plan on creating this bridge themselves, but because of the new SDK changes, 3rd party developers can potentially come up with better solutions.

Apologies, I don't see how any of my comments could be construed in such a way. I did not speak in any specifics to our development plans in that regard and have no information that the LIghtWave team is ready to release on the topic. There's no basis on which to make such an assumption.


Well, after reading the feature list again, it looks promising given those features are incorporated well. While it's great to see some old weaknesses in LW have been addressed like OPENGL, a node based material editor, and finally SDK (which to me means easier outside implementation of custom tools) unless I missed something on the list, the render issues I was hoping would be resolved were not mentioned, like better motion blur (I recently spoke to a fella who claimed they were considering searching for an outside render solution because LW's motion blur couldn't meet their needs), subsurface scattering, an RSA buffer saver output interface with Z depth outputs etc, better EXR and RLA support, and independent ray recursion limits for different types of rays, the increase in speed is very nice, but honestly LW's rendering speed doesn't bother me as much as the other issues. I'll be sure to leave the requests on Newteks site, in the meantime I've posted a few topics in the "how to improve lightwave forum" along with other valid requests from different users.

So, with your knowledge of LW's future, what would you tell the president of the company I spoke to, about his concerns implementing LW in their pipeline?

[snip]

Very big deal, and I agree. A while back I was at a company considering switching some of
Really for it's cost, LW's native renderer is still one of the best...but I'd like better. :)

Better is on the way. Vector Blur has already been addressed, and motion blur and a good many of the rest of the issues you mention are in planning to address once the main core changes for v9.0 are in place, as they will be very dependent on those changes; the list of features for v9.0 is preliminary, subject to change without notice, and in particular not exhaustive as regards the potential features that may ship in the product. Then, of course, there will be the other releases in the 9.x series.

As for the fellow you are talking to, if there's a way he can get a look at how some of the current mixed pipelines are managed and evaluate LightWave's renderer directly to see if it can produce the look or looks he wants for their projects, that's really the best way to find out if it will fit his needs. Hands on with what's here now is always the best way to make such a decision for current and immediate future needs.

Chuck Baker
08-20-2005, 11:48 PM
To them it's above the competition for their needs, I see your point, but that doesn't rule out the possible need to improve it. Just because I think it's the best offering at the time, doesn't mean I don't feel that adding more isn't needed, or that the once dominant native renderer in the 3d industry is in danger of becoming just another good renderer for the price if advancements aren't planned in it's future.

I'm not sure what I got from that statement about lw9's renderer, it can mean many things...I'm not going to elaborate and assume, only to have my expectations let down.

Very practical comments and very much what we know to be true - the application has to grow constantly, and the renderer in particular has to advance very quickly at this point. The team is hard at work on that.

Jarrede
08-21-2005, 12:07 AM
As for persuasion tools, it seemed to me that you are expressing your point of view as regards aspects of the usability of LightWave in your posts, and that the story of "the head of a VFX company that you know" was intended as persuasive support for that point of view. Given that, that's why I felt it was worthwhile to mention that there are alternative views.


It wasn't intended to support my view, in fact I mentioned I didn't agree with it, but recognized his position. I was trying to see if anyone had answers regarding his concerns, and asked you what you would tell him to put his mind at ease regarding future LW versions.

I didn't like not having answers to these things, but I can't help to ask the question "why" do I not have answers. If it's my knowledge, then that can be fixed. If it's limitations in the software...well there isn't much I can do about that.


That doesn't mean that NewTek isn't going to take your comments or the reservations of the party you describe into account - but in public discussions there is more to account for than simply accepting the feedback, as I'm sure you must be well aware.


I don't understand what your trying to say.



I'm very aware that it is the talented and inventive artists who earn the awards - and there should be no onus on mentioning that our product was chosen for use by such folks for their award-winning projects nor any suggestion that we lack the proper perspective on this.


It was more or less how it was worded, if you had said "LW helped companies and individuals achieve X awards" that would have been different to me.



Apologies, I don't see how any of my comments could be construed in such a way. I did not speak in any specifics to our development plans in that regard and have no information that the LIghtWave team is ready to release on the topic. There's no basis on which to make such an assumption.


I was making a projection to see if you'd bite. I'll take it as a "no comment"




Better is on the way. Vector Blur has already been addressed, and motion blur and a good many of the rest of the issues you mention are in planning to address once the main core changes for v9.0 are in place, as they will be very dependent on those changes; the list of features for v9.0 is preliminary, subject to change without notice, and in particular not exhaustive as regards the potential features that may ship in the product. Then, of course, there will be the other releases in the 9.x series.


that's really all I wanted to hear, that Newtek values similar improvements, recognize their importance, and has plans to make it so sometime in the future.


As for the fellow you are talking to, if there's a way he can get a look at how some of the current mixed pipelines are managed and evaluate LightWave's renderer directly to see if it can produce the look or looks he wants for their projects, that's really the best way to find out if it will fit his needs. Hands on with what's here now is always the best way to make such a decision for current and immediate future needs.

They value flexibility in custom shaders to achieve a lot of the looks they need. And they've made a large investment into their current software, and employee's, getting these types of companies to budge takes a large incentive. Even if advantages are seen in the future, the cost of switching their pipeline among constant tight deadlines is risky at least. It's a hard decision for someone to make. Entertaining a few licenses can be a foot in the door. It's difficult to come to a conclusion, someone see's the advantage, but isn't sure how it's going to meld itself into the pipline, if the extra time spent now on getting up to speed will be worth it in the long run. He noticed Fprime above all else, thats what sparked his interest.

Hands on, what's here now, may not be suitable for their needs, but I don't want him to completely dismiss future LW builds based on what's out now.

funfun
08-21-2005, 12:57 AM
Jarrede, I can understand and I agree your point about LW, but not all.
But, why not wait few more months until LW 9? I think LW9 is a very important update for both NEWTEK and users, otherwise, I am sure many ppl will find another way this time. I would not say LW9 must be great, until I really have it on hand, but what they showed us so far is quite promising and impressive. Time will tell.

Lyr
08-21-2005, 01:07 AM
As someone who just bought Maya7, i'm really disappointed, no Viper for FX,
no Fprime for instant everything, and new stuff like the ToonShader....
Will not work with the Hardware, Vector, or Mental Ray Renderer.....



Alot of this statement is just false. The new toon shader uses paintfx as a base which then can be tesselated to nurbs or polygons for rendering in any software that recognizes nurbs or polygons, which covers them all I believe. Who knows when LW9 comes out there will probably be pipelines that are doing toon modeling/animation in Maya (since you can see the whole toon effect in the openGL viewport while you work) then point ovening over to LW for rendering. It's a very flexible and powerful system, try not to get so disapointed so easily.

Zarathustra
08-21-2005, 01:10 AM
That does not make it inappropriate to discuss the quality on its own
Did I say anything you posted earlier (http://forums.cgsociety.org/showpost.php?p=2562602&postcount=57) was "inappropriate"? I don't understand what you mean by using that word.

If LightWave's renderer couldn't produce the required quality, no number of free render nodes would have gotten it to where it has gotten to in the industry.
Well of course. If it sucked, it really wouldn't matter how many nodes you had. I didn't say it couldn't produce, and it's a testament to it that it's held up for so long but there are other renders out there. Better renderers out there. If they become affordable, then they would be a real challenge.

The combination of rendering quality and free render nodes has indeed been a foundation of the product's success.
Exactly. That was my point. We can bicker over which played a bigger role, but the # of nodes is a big factor and it appeared to me you were saying quality alone was the selling factor. That's all.

Jarrede
08-21-2005, 01:18 AM
Jarrede, I can understand and I agree your point about LW, but not all.
But, why not wait few more months until LW 9? I think LW9 is a very important update for both NEWTEK and users, otherwise, I am sure many ppl will find another way this time. I would not say LW9 must be great, until I really have it on hand, but what they showed us so far is quite promising and impressive. Time will tell.

I have stated I was waiting for the release of LW9 to come to my own personal conclusions about it's place in my pipeline. :)

The main point I was trying to make regarding the VFX president is, if I could have told him he could use their current render solution that they love, it would have left him with little reasons not to consider using LW. It's a difficult decision for companies, and if they know it can be integrated along side other tools their using, it poses less of a risk to them, and it makes the decision a bit easier.

Therefore, IMO, it should be within any 3d software developer's best interest to try and make these changes as easy as possible for them, by doing so, they increase the chances of tapping into new clients, boosting sales, and reputation. I was simply trying to express a different side to the value of mainstream 3rd party renderer support, other than my own values as an artist. Not, "look LW's render sucks cause it doesn't work for him" :)

PaZ
08-21-2005, 02:00 AM
I work on Lw and other tools from many, many years.

Lately i've been very critic towards NT; after LW 9 press release, i should admit they've worked in the right direction and my hopes are still high.
Lets come to LW's renderer now. Using it daily in every VIZ area, i find it not only aged, but very poor when it comes to control. I'm gonna give my opinion on current engine, and it's also my positive advice to NT.
LW engine can do the magic just adding an area light, but you cannot control that magic unless you use tons of workarounds. Control is power; current lighting toolset is very, very poor. Every kind of shadow should be accessible form every kind of light, and the same for falloff options (which could be improved with a graphic editor or a gradient). Give every light a cone on influence and falloff, and introduce customshape Arealights even without GI. Shadow maps must have bias control, baking tools and much more. Again, there are other shadowing technologies which have never been implemented in LW jet. Shadow Designer - like tools sould be implemented in some form/measure.
Inclusion/Exclusion must me much more selective: i/e form light, shadow, diffuse, specular, radiosity, reflections at least.
Every light should be able to project images and procedurals.
I could go further on lights but its just to show what i mean when i say "we have too few controls".
GI is the weakest area: there are tons of newer techs and engines, the one we get is really too simple and poor, it was poor even when we got it in 6.0. Put an "advanced tab" for pro users, and introduce newer algorithms (or implementr Kray). Current interpolated GI has problems with transparent surfaces, sample control is paleolitic at best.
Crucially important, introduce a good GI storage method which allows an effective caching for animation/walkthrough. Kray is still a brilliant example, i really hope NT will tightly collaborate with this guy.
Antialias is still too slow and multipass; introduce other kind of AA methods which can free users from multipassing, which has good points but also lots of drawbacks. Most of all, AA should implement some smart geometry detection to get proficient on complex textures and moire-like effects. This is also a texture filtering issue, another area where LW has still a lot to catch.
Fortunately we'll get Vodka in 9 so another major limit has gone, camera management tools.
Rey recursion limit should be defined for reflections and transp. separately.
Reflections requires a reworking for additive issues and more control.
Motion blur is by no mean a closed chapter, Vector Blur cannot still be a substitute for classic MB which is still too slow and not suited for some tasks.
Netrender is a strong point for Lw, but still no frame splitting; rendering large GI images in LW is still a pain.
I could write much more, but i think these are already lot of things.
Given this all, i get beautiful images form LW but they could be better and faster. I'll repeat this, i've high hopes for 9 but an objective comment on current engine cannot be so enthusiast as some here seems to push.
Not bad, not so good, and many, many thing which requires strong improvement (replacement) since they're stuck to mid 90s technologies.

Paolo Zambrini

Freak
08-21-2005, 03:08 AM
Alot of this statement is just false. The new toon shader uses paintfx as a base which then can be tesselated to nurbs or polygons for rendering in any software that recognizes nurbs or polygons. try not to get so disapointed so easily.

I think Maya is a powerful piece of software, no question... And i do like many things....
But i can send samples of the Toon Shader, The Hardware and Vector renderers are identical,
(i.e missing much of the details) and Mental Ray, seems to work sometimes.. :) Some of the demo scenes, do not render in MR..... The eyes and the mouth of the Carrotop.ma for example. (the rest shows up) Often some of the FX, are missing from one renderer or the other....Perhaps with some changes, or converting to poly's or something will help...
There is meant to be a feature to make a global shader network to aid with this,
in 7.0 (showing similar results from the one shader in all renderers)

Maya Fur now previews in MR, without converting to poly's....
However this was only added in version 7.0 (And only in Unlimited)
Maya7 also increased integration of MR with motion blur, Vertex Colouring,
so up until (and including) now , MR and Maya still have some areas
that the MR renderer doesn't talk nicely with... (But it get's better over time)

Hardware OGL Shaded Lines are nice though......
The ToonShader and PaintFX are the highlights so far, however......
As you say maybe there are ways to make them show up in MR....
There are definately some holes in the native support in many places.
(as does LW with shaders in Fprime)

Mental Ray is certainly slower than LW's renderer..... IMO,
Maybe with time, i'd learn to learn it's in's and outs....
But even the IPR is slow, (C4D's IPR is much nicer)
For speedy workflow, LW is way ahead with Viper for HV
and FPrime for Lighting and surfacing......
That's not to say, there ain't plenty in Maya i do like....

Joviex
08-21-2005, 06:48 PM
Supposedly Lightwave 9 is going to have adaptively-tessellated subdivision surfaces.


NOT to be confused with micropoly displacement.

Though I do find it to be a neat feature.

ThE_JacO
08-22-2005, 03:22 AM
NOT to be confused with micropoly displacement.

Though I do find it to be a neat feature.

to be able to tesselate adaptively the limit surface the engine needs to be micropoly capable.
if something like AHSDS will be in the app then so will be micropoly displacement.
if that happens that will be a huge step forward. A full-on rendering engine only company like Mental Images, to implements that a few years back without killing retro-compatibility, had to make MRay into 2 rendering engines to compete with PRMan on the same playfield, and that is what MRay's rapid motion mode is nowadays, and it's why it's nearly as fast as PRMan is at things where only micropoly based engines usually excel (kickass moBlur, adaptive tesselation etc.).

annaleah
08-22-2005, 04:54 AM
Very practical comments and very much what we know to be true - the application has to grow constantly, and the renderer in particular has to advance very quickly at this point. The team is hard at work on that.


When do you think the team will have this ready?

cpan
08-22-2005, 06:35 AM
Mental Ray is certainly slower than LW's renderer..... IMO,

bullshit! :)

please stop talking about things that you even don't know what they are lol :p



take care
cpan

tais
08-22-2005, 08:36 AM
I'm with Paz on this one...

Freak
08-22-2005, 09:35 AM
bullshit! :)

please stop talking about things that you even don't know what they are lol :p

take care cpan

Wow, CGTalk, just seems to keep adding Troll, after Troll these days, ;)

It's so nice to get inteligent, and specific qualification of your opinion,
oh please enlighten us more with your detailed and oh so accurate info.
No contest, LW's renderer is way faster, than the software renderer (just), or MR
in Maya7, (MR was only in Draft mode) tested using various lights, with a .obj.
MR is as slow as shit.. IMO...It's nice, but it's slow....

Oh, wait...... You seem to offer a very childish response to my opinion,
But then i read, you are only a 17 year old child, student from Romania....
And suddenly, you are forgiven. ;) LoL...

Before we start a useless, flaming session..... Let's just say...
Okay, you are right... MR is slower than LW... :)

PetterSundnes
08-22-2005, 09:38 AM
Sniff sniff, this smells like a thread closure coming up...

:-P

cpan
08-22-2005, 09:47 AM
indeed, no render contest at all, but the heck... if you don't know how to setup mray I also don't expect it to be as fast as you expect it to be. Maybe LW's renderer is faster at certain things (such as plain scanline without raytracing or maybe motionblur) but overall mray it's faster, ofcourse, if you know exactly what you'r doing with all those settings it has.
so, Freak (hehe you'r nickname is suggestive :)), keep it down and after you learn both softwares at their full posibilities (aka fakes, speedup tricks, BSP, etc) speak again.

as i said before... stop saying things about other renderers that you even don't know how they work.

oh... and u used draft preset right? lol I can see your knowledge on mray allready ;). mray isn't a software that has the solution/preset for every scene... it has to be tweaked in direct relation with the scene (lights, amount of polygons, etc)
and one more thing... u used the maya shaders right? well try using only mray in the scene (aka mray shaders, lights, etc) to see it's strenght.

take care
cpan

PetterSundnes
08-22-2005, 10:26 AM
My comment was ment in general for this thread and the way it is developing, not at Freak specifically. (just to make it clear)

RobertoOrtiz
08-22-2005, 02:22 PM
Plese keep the personal attacks off the forum and off this thread. So far people have been very cool expressing their VIEWPOINTS. Lets keep it that way.

And the forum is doing just fine, Thank you.


-R

PetterSundnes
08-22-2005, 02:42 PM
:thumbsup:

As pointed out in the thread about fish eye lense effect, there is a way of doing that without having it as a toggle feature under the camera properties:
http://forums.cgsociety.org/showthread.php?t=269300

Thats what I have always liked about 3D, is to take real world knowledge into LW, test it out, and see it working. No need to set up a specific type of material, for a specific type of camera for a specific type of renderer.

Freak
08-22-2005, 09:47 PM
i Maybe LW's renderer is faster at certain thingssuch as plain scanline without raytracing or maybe motionblur) cpan

Yes it is faster, well done!!! (Now you are agreeing with me) it seems you are getting smarter with age. PS LW's render engine is based on a Polygon renderer with selective raytracer.......

It's doesn't do scanline, renders at all as you mention.... :)
Perhaps you should learn LW's renderer, to it's full extent.....
Before offering your incorrect assumptions...... (No offence!) LoL.... :)
Unlike Maya and Max's scanline' renderer.....

So i think you should maybe to a little more reasearch, on telling me what i do or don't know,
about a render engine, iv'e been using since before you were 10 years old.... ;)

lightblitter22
08-22-2005, 11:06 PM
oh... and u used draft preset right? lol I can see your knowledge on mray allready ;). mray isn't a software that has the solution/preset for every scene... it has to be tweaked in direct relation with the scene (lights, amount of polygons, etc)
and one more thing... u used the maya shaders right? well try using only mray in the scene (aka mray shaders, lights, etc) to see it's strenght.

Come over into the C4D forum, young padawan, and bring MentalClog with you. Our very own XSI troll has been feeling lonely lately, and we always like to roast another MRay fanboy on a spit. :thumbsup: I think we'll try garlic sauce this week...

bt3d
08-23-2005, 08:18 AM
It's funny to see how often a thread becomes a personal flamewar. That's the reson why I first though, come on, don't post in this thread, because it will become a boring 'my renderer is better than your ones'. Luckily it takes some posts of more or less objective views before it swapped. I would assume that some people get paid by the software developers, to make such onesided 'marketing' statements, since I really don't understand why they make some renderer bad without any sense and maybe knowledge.
There are surely situations where the LW renderer is faster than MR, but there are also situations where MR is faster than LW. As usual it depends on your scene and what you want to do, and believe me I know what I talk about because I work with both. In my personal experience things like Ambient Occlusion (or dirtmap), GI, soft reflections, area lights and antialiasing are much faster in MR, but hey, that's no reason for me to say that MR is faster at all.
I've worked with a lot of different render engines in the past and they all have their special areas where they are top. I think the advantages of the LW renderer have been pointed out more than ones in this thread (easy of use together with good useable quality).

IC12
08-23-2005, 09:00 AM
Yes it is faster, well done!!! (Now you are agreeing with me) it seems you are getting smarter with age. PS LW's render engine is based on a Polygon renderer with selective raytracer.......

It's doesn't do scanline, renders at all as you mention.... :)
Perhaps you should learn LW's renderer, to it's full extent.....
Before offering your incorrect assumptions...... (No offence!) LoL.... :)
Unlike Maya and Max's scanline' renderer.....

So i think you should maybe to a little more reasearch, on telling me what i do or don't know,
about a render engine, iv'e been using since before you were 10 years old.... ;)

I have very young children and I think I'll introduce them to these forums.

Their language and temperament is well suited to this kind of pathetic tit for tat.

KevinJM
08-24-2005, 02:33 AM
Come over into the C4D forum, young padawan, and bring MentalClog with you. Our very own XSI troll has been feeling lonely lately, and we always like to roast another MRay fanboy on a spit. :thumbsup: I think we'll try garlic sauce this week...

You just made my day man. Cheeerssss:applause:

Jarrede
08-24-2005, 03:52 AM
bullshit! :)

please stop talking about things that you even don't know what they are lol :p

if you know exactly what you'r doing with all those settings it has

take care
cpan

I've noticed it's slower in some areas, but faster in others. Overall though I got the feeling it was slower.

" if you know exactly what you'r doing with all those settings it has"

ah! that must have been it, my setup knowledge is limited.

One other thing I've noticed, what really brings LW's render engine to a hault, is transparencies, ...I'm currently working on something that has a bottle in it, and if it gets close to the camera render times go through the roof. Can't use shadowmaps for lights because they won't see through, so I gotta have raytraced lights in the scene...

One thing that I thought of that would help this is being able to set seperate ray recusion limits for things, like I need to have it set to 6, just to see through all the bottles surfaces, but I don't want reflection and refraction bouncing around that many times...

IC12
08-24-2005, 10:49 AM
Hey Freak

Thanks for the private email you sent me outlining what you intend doing to myself and my children after my previous post.
Interesting reading. So interesting in fact I've forwarded it to the UK police.

There's some scary people out there.

Freak
08-24-2005, 11:32 AM
Hey Freak
There's some scary people out there.

Errr, Do you mean me? Are you a troll, MR fanboy or something?
I think you have the wrong person... I didn't send you a PM!
But feel free to forward my name to anyone you feel fit too!
I can send you a Photo by PM if you like? :)

They may be interested in my drug lab, and gambling den/brothel too! :)
If you have something to say about my language, that's fine...
But please try to keep posts related to threads on topic, nobody
likes wet blankets who complain, and offer nothing new.
And try not to make it personal, just because your clients keep coming to me.
in droves, and remark how much better my work is, has nothing to do with
Lightwaves rendering options... :)

But you don't need to make stuff up. We still love you down at the pub! :) Come and have a pint sometime... We only talk behind your back sometimes. :)

IC12
08-24-2005, 11:58 AM
Ok - my last word on the subject:

Somebody sent me a threatening email and it referred to my post where I quoted you. I forwarded the email to the police and gmail (google), not the name of the person who sent it. I did this as the contents were disgusting.

RobertoOrtiz
08-24-2005, 02:01 PM
People for crying out loud we are talking about a rendering engine,
not comparative fundamentalist religions.

ENOUGH.

The topic has been beaten to death and it has been hijacked by a few.
SO ENOUGH

IC12 send the a copy of the e-mal.
And to all this sort of behavior will not tolerated.

I will personally BAN the person who send it.


CLOSED THREAD


-R