What 3d/graphics software would you rec...

[Brian D. Goff]

Anybody running these on a Mac with bootcamp with any success?
XSI in particular. It looks like ZBrush runs native on Mac just not with the latest update???

I think XSI 6 requires a spezific grafics card. Download all the demos you can get and find the perfict 3d package - don't forget to look a Cinema 4d too.

 

[Gavin Greenwalt]

I've seen XSI 6 running on a macbook. Don't know what their specific setup was.

 

[Jaime Vallés]

Blender is free...

 

[Mark L. Pederson]

Houdini

http://www.sidefx.com/index.php?option=com_content&task=view&id=584&Itemid=219

You can build a house in Maya and and you can build a house in Houdini. It's just that in Maya, you need to engineer the hammer first.

Luxology is a year or two away from being the KILLER 3D app. I said it before, and I will say it again, Adobe is gonna buy them while Apple builds another iPhone model.

 

[Priyesh P.]

Houdini

http://www.sidefx.com/index.php?option=com_content&task=view&id=584&Itemid=219

You can build a house in Maya and and you can build a house in Houdini. It's just that in Maya, you need to engineer the hammer first.

Luxology is a year or two away from being the KILLER 3D app. I said it before, and I will say it again, Adobe is gonna buy them while Apple builds another iPhone model.

I can only agree. Modo rocks, but unfortunately it still doesn't have character animation. That was a big dissapointment for me this year. The other must-features missing in modo are: construction history and node based texturing, the "shader tree" is crap to me, too abstract.

 

[Rocket]

I'm going to have to check out Houdini, downloading the trial version now.

 

[Jim Hoffman]

This is clearly one of those times when a website helps sell your product. I'm downloading what demos I can - I just use Mac - but would consider buying a PC when I figure this out...

BUT - since this is a product that will take a very long time to master I admit I am influenced by the gallery work on the websites. I've been around long enought to know that those decks can be rigged - but its hard not to judge. These companies that are taking advantage of the design work of their sites are smart.

Still - wow is it impressive what a little money and LOTS of talent can do.

 

[Joe Carney]

>>I'm most familiar with Lightwave, but I would not recommend it to people who are just starting out or who are buying into a new software solution. The reason why is that NewTek's direction with Lightwave doesn't appear very promising and development on it has become rather stagnant. Too many political issues within the company these days too and many suggestions from users, beta testers and even inside personnel fall on deaf ears. Perhaps I know too much...
<<
Jeff I just got Lightwave 9.2 and it rocks. Plus if you buy the boxed version you get about 400.00 worth of training materials (books, dvds) for free.
One of those books is "Essential Lightwave 9" from wordpress. A must have for newbes like myself. I come from Animation Master and need to relearn a few things.

Lightwave has unlimited render nodes included in the price. Plus there is pmgMessiah for animation and Fprime for fast High quality rendering if needed and Worley upgraded his license to unlimited render nodes.

I think your info might be a litte out of date. You are right, there was a period when it looked like Lightwave was on life support, but no more I think.
One of the other reasons I puchased is the large 3rd party plug-ins and a large supportive community (cgtalk, spinquad, newtek...)

But..Blender is getting very good and they have world class sculpter and fluid tools. Wings3d is also free and a gem. There are several Renderman options available too. Blender has a built in compositor as well.

If you have 3.5K USD or so to spend, for all around Broadcast oreinted graphics and animation, Cinema4D is great and they can output their renders directly to After Effects projects. I think the VRay interface is coming along as well.

Joe C.

 

[Gavin Greenwalt]

Haha well if you're looking for "gallery marketing" I was at the Seattle Brazil r/s User's Group last night and one of the brazil devs mentioned that the Transformer models and rigs were made in 3D Studio Max by a bunch of ex-Blur artists from the digi-matte department.

 

[Jeff Kilgroe]

Jeff I just got Lightwave 9.2 and it rocks. Plus if you buy the boxed version you get about 400.00 worth of training materials (books, dvds) for free.

Yep, that's a good deal. And LW9.2 is arguably one of their best releases ever in terms of it actually working as it should.

Lightwave has unlimited render nodes included in the price. Plus there is pmgMessiah for animation and Fprime for fast High quality rendering if needed and Worley upgraded his license to unlimited render nodes.

Yes, and unlimited render nodes is one of the primary reasons I standardized much of my work on LW and continue to use LW as my primary 3D software. This is a huge deal when you actually need to deliver animations on a regular basis and with deadlines. I run a render farm here with 24 systems at the moment... It would cost me thousands per year to maintain support for that if I were using Mental Ray with XSI or Maya. FWIW, MAX also has unlimited render nodes, but I don't think their renderer is as nice as LW and FPRIME. MAX really needs to be used with third party renderers like Mental Ray or VRAY.

I was actually trying not to dwell on the subject or turn this thread into a discussion of why one software is better or worse than another. And looking back at what I wrote, I don't think I came across the way I intended. I really shouldn't have said that I wouldn't recommend Lightwave, because I actually would recommend it if someone needs a complete end-to-end solution and they have the need to drop a software application right into existing production workflows. Especially if they're a small operation or a one-man show.

I know Lightwave inside and out and know a few people inside NewTek that have a good feel for the direction LW is headed. Or more accurately, I should say where Lightwave is not headed. Most of my disgruntled feelings toward Lightwave have to do with ongoing issues and shortcomings that are repeatedly ignored from one revision to the next. All too often, the inclusion of new features is foremost on the development roster while addressing various issues like a shoddy network render implementation, impotent multithreading and poor SDK access have received little or no attention for nearly a decade.

To expand (here comes the ranting)... LW does have some serious shortcomings with network rendering. It's a pain to set-up the first couple times until you figure it out because the implementation is just downright stupid. You essentially have to use a third-party render manager software for any significant installations because LW's ScreamerNet II interface is a pathetic joke that has no management capabilities at all. A few years back when I actually had some employees and I was running lots of render jobs, I wrote my own render manager software. It was crude and console-based, but very powerful. IMO, it was more powerful than than most render managers for LW are now. I wanted to put a GUI on it and clean up a few things and sell it, but I never got around to doing that. Just one of those things that I probably should have done. But I actually migrated my workflow to Station-X's Spider and have customized it and re-written many parts of it to fit my needs. Unfortunately, I can't release any of that as there's no rights to do so. I was too slow in my attempts to acquire Spider when Station-X went belly up and it was sold to a software developer in India. This developer has yet to do anything with the source code, but they are unwilling to deal on release or development of the product. Rather frustrating. It is distributable in it's last released form and source code is available for private use. anyway, that's going off on a tangent...

Lightwave also has a brain-dead thread management system. As do many software applications, but LW is pretty bad. It leaves it to the user to manually choose how many threads they want the renderer to use and then it just mindlessly splits the frame into a segment for each thread. In most situations, each thread runs to completion on that slice of the frame and then dies. But in some cases it will circle around and begin another task, maybe, but this is hardly the best way to divide up a render for multithreaded processing. Additionally, LW's plug-in and post-process interfaces were single-threaded and so were the plug-ins included with LW. So often the render would be half done and then lock itself to a single thread as a plug-in runs or other task completes. Multithreading has been mostly broken in LWSN (the screamernet render node software) for years and has only with the release of LW8 actually properly acknowledged the thread setting in the LWSN config file. But it still is often more efficient to run individual single-threaded instances of LWSN per CPU or core instead of a single instance running multiple threads to account for each CPU or core in a system.

If you're getting into Lightwave and will be giving network rendering a try and running on Windows, I have a small utility that restricts CPU affinity and priority on my site. Very handy for LWSN setups.

I think your info might be a litte out of date. You are right, there was a period when it looked like Lightwave was on life support, but no more I think.

I don't think Lightwave ever looked as if it was on life support. It's always been a major player, even though it doesn't seem to get the same respect as Maya and MAX. Most of my frustration with LW is ongoing as I've had a serious dislike for their multithreading model, network rendering implementation and SDK.

One of the other reasons I puchased is the large 3rd party plug-ins and a large supportive community (cgtalk, spinquad, newtek...)

The Lightwave community is indeed awesome. I don't participate much in those forums and whatnot much these days. Probably because I haven't been doing as much animation over the past two years. But I'm somewhat of an old name in the LW community (DarkScience), primarily on the not so much used anymore usenet group. At any rate, I've actually been shifting back into some 3D work lately and I'm currently working on a short for a client as a holiday piece titled "Clockwork Christmas". Entirely modeled in Modo. Rigged, surfaced, animated and rendered in LW9.2. I also have pMG Messiah Studio, but almost never use it. I can do everything I need in LW and have found it to offer little to nothing that LW can't do. Actually my Messiah Studio license is for sale if anyone is interested.

Lightwave is also only 75% complete if you don't have some of Worley's tools. FPRIME is a must if you're doing any kind of fast turn-around render work and most projects in general. Unfortunately, it doesn't work with Sasquatch (their hair renderer for those who don't know), hopefully soon. I actually much preferred Shave and a Haircut as a hair solution. For those who don't know, that was the hair/fur renderer software created by Joe Alter (VFX Software Developer on various '90s films like Godzilla). Shave was originally produce for Lightwave and then Maya (and I believe C4D), but these platforms were dropped when he was bought out by SoftImage and Shave has been integrated into XSI as their solution and is no longer developed or supported for other platforms. Anyway, I'm rambling...

I think a good approach for a true beginner in the 3D world would be to establish a budget. Try demo software and the free ones like Blender and see what appeals to them. There are lots of options and none of the major software applications are truly better than another. Well, I take that back, XSI:Extreme is where it's at, but that's hardly a beginner option or even a realistic consideration for most small studios, let alone an individual.

I know it's not the cheapest option, but I would recommend starting with Modo. My next recommendation would be XSI:Essentials. OTOH, Blender still amazes me for all it offers in a free package and only a few years ago, many of us were paying big money for applications that didn't do everything Blender does now.

 

[Priyesh P.]

I've bought lightwave in 1997 and still use it from time to time, until I totally master XSI.
But Lightwave still sucks, even with 8.0 (that I'm using, but I guess most issues stayed the same with 9.2) (sorry). Why? Read this:
-it derives 90% of it's functionailty by too many little plugins, nearly each one with their own, stupid way of setup
-non-integrated animation/modeler parts (it makes me nuts to do minor changes on models, not to say that it is impossible to animate modelling operations at all)
-character animation sucks
-no construction history
-render tweaks are a pain without Fprime

 

[ColinSmith]

Ah, Lightwave.... we love to hate it... ;-)

It's the one I know by far the best, so for the work I do there is no point to changing. If I was starting totally fresh I MAY be tempted to go to XSI, but that's another thing....

Multithreading... well I don't know much about the programming details, but LW9.2 is the only app I run that will max out all 4 cores, so.... I'm not sure..... are you saying they are all working, but not producing efficient output?

fPrime..... yes it's 3rd party, but LW + fPrime is still a cheap way to get those kinds of capabilities..... even if you just use it as a previewer during setup. With9.2 LW's native render engine had a big big tweak and it a lot faster in general, and now has pretty good radiosity built in too.

For me, I don't mind modeler being seperate at all..... if it gets integrated and it works I don't mind that either, but as 80% of my modeling is in Rhino anyhow, I am used to modeling in one app and moving it to another. If the Rhino to LW path was smoother that would make a lot bigger difference to me.
Now I do find modeler itself to be in need of a big clean up of the tools... to many multiple options that do vaguely the same thing, differently....... but I guess if I knew it better I'd get past that.

I do think that for a studio of 1,2,3 people LW is really worth a look though, it still offers a lot for the money and it's realatively easy to produce quality renders, but of course you should check out the other options and see what fits best.

Another vote for Rhino for any "hard" modeling though. If you have any kind of CAD experience then you will be amazed how much like AutoCAD Rhino is, while at the same time how much BETTER it is for 3D work. I moan non-stop whenever I am forced back to AutoCAD for just about anything these days.

 

[Gavin Greenwalt]

Yes, and unlimited render nodes is one of the primary reasons I standardized much of my work on LW and continue to use LW as my primary 3D software. This is a huge deal when you actually need to deliver animations on a regular basis and with deadlines. I run a render farm here with 24 systems at the moment... It would cost me thousands per year to maintain support for that if I were using Mental Ray with XSI or Maya. FWIW, MAX also has unlimited render nodes, but I don't think their renderer is as nice as LW and FPRIME. MAX really needs to be used with third party renderers like Mental Ray or VRAY.

Very important point. It can't be emphasized enough what significance a render farm is if you intend to provide any sort of 3D Work.

It's simply not cost effective to sit around waiting for a render in order to continue working and the cost maintaining a renderfarm can sink you really quickly if you aren't ready for it. As far as pipelines go I've only really built from the ground up a Maya and a Max pipeline. My maya pipeline took 3 weeks to get up to speed (with a lot of personal code and duck tape holding it together) in the end I found its sattelite system lacking to say the least and found I had to trick into doing what I wanted in order to render faster than a single system.

Max took 1 day and includes backburner which can queue/prioritize/split and stitch single frames/schedule and a lot more. The difference was night and day. I can't speak for LW's solution since I've never used it, but if it's a good one, it can make all the difference if you're actually doing a lot of work and your time is valuable to you.

btw Max's unlimited render nodes does extend to Mental Ray so you're not limited to its internal scanline. I actually really like the scanline because it's fast, does everything except GI and is a "known quantity" for me since I've used it for so long. And did I mention it was fast? Oh well Brazil is my weapon of choice now a days.

 

[Priyesh P.]

My conclusion and adivce for starters: get XSI OR Maya, Max, C4D. Or if you can wait another year or two, get modo.
But please forget Lightwave, it's time is over...
(My opinion)

 

[Mark L. Pederson]

Houdini

http://www.sidefx.com/index.php?option=com_content&task=view&id=584&Itemid=219

You can build a house in Maya and and you can build a house in Houdini. It's just that in Maya, you need to engineer the hammer first.

Luxology is a year or two away from being the KILLER 3D app. I said it before, and I will say it again, Adobe is gonna buy them while Apple builds another iPhone model.

Update on my previous post - check this out -

http://www.studiodaily.com/main/news/feed.rss/8426.html

 

[pat@hpnc.com]

Take a look at http://forums.cgsociety.org/ . They have a huge amount of members of the forum and sections for each of the major 3d and 2d software packages.

 

[Jim Hoffman]

That Sony/Side Effescts thing sounds interesting.

I've decided to start with Modo to get my brain wrapped around this whole thing. I'll move on to something else when I get the flavor of this one...

Thanks for the input - any thoughts or ideas - keep em coming.

As a side note - I'm out at USC this week and stopped in the bookstore to see what they are selling to their students. Maya and 3DS Max seem to dominate the shelves. This is such a cool campus...

 

[Jeff Kilgroe]

Wheee... Cell board to work with Houdini. Sounds expensive. Curious choice of partners, I wonder why Sony didn't go after SoftImage? Or maybe they did and it didn't work out? Or what about Mental Images (MentalRay) or Pixar (Renderman)? Somewhat unclear as to what the Cell board is actually going to do here.

Hoffman, Modo is a great place to start. The best thing to do is get 3D workspaces, tools and common practices firmly set in your brain. A good foundation in 3D will enable you to shift to other software pretty easily. Or at least with a minimal amount of pain. Houdini is more specialized and not as mainstream as Maya, MAX, XSI, etc.. There are some things it does very well (better than anyone else), but as a full-featured, all-in-one solution, I would still choose something more common.

 

[Peter Karlsson]

If your a one-man shop, I would look into a XSI/LW solution.. You constantly run into bumps with Maya, and need a custom Mel script for all the quirks. I would only recommend Maya for houses with more than five 3D artists.

LW has gotten a tremendous facelift regarding shaders/rendering, the last couple of years, and it gets better all the time. The AA/DOF/and Motionblur is beautiful. XSI has really strong animation capabilities, and easy J-/VBScripting that lets you do about anything. Using pointoven forth and back between the apps works really good.

One application seldom has all the features you want all the time, so I suggest trying to look into atleast a combo that works well together! Sofar I feel that XSI works well with all the rest.

 

[Jeff Kilgroe]

Hmmm.... I haven't used Maya i a couple years, but I didn't find MEL scripting to be an issue back then. Much more convenient than LW's LScript. But hey, I use LW all the time. And I find that every software will throw you a curve ball now and again that can only be dealt with by writing a custom script.

Maya can be very powerful in the hands of a single user -- I've seen some remarkable results with it from 1 & 2 man operations. I think it comes down to each person deciding what they need for their application. Maya is a very good choice, even for a single artist, if a lot of scripting and data viz is part of the regiment. If you're not comfortable with scripting or don't really have much programming in your background, then Maya probably isn't for you.

XSI's greatest strength is also it's biggest weakness and that is their encapsulated business model. They provide all the tools you need... There's hardly any third-party anything to go along with XSI. Also, if the full set of tools become necessary - hair/fur, fluids, cloth, rigid-body FX, etc.. Then XSI becomes very costly to maintain and to a small 1 or 2 man operation this can be an issue if they're not using it on a very regular basis.

The best advice for someone new to all these software packages is to try as many as they can. Learn about all their relative strengths and weaknesses and then weigh that against the type of work they will be doing.

One other note about LW... While they did do a massive overhaul of the shader system -- nodal workflow, yay!!! The majority of LW is still stuck in the dark ages. The software has improved little over the past decade and many of us long-time users have quit screaming over most of our issues because NewTek obviously has no intention of fixing them. But for a glimpse of that, you can read my prior rant over LW in this thread. But LW is still my primary 3D software I use. And that's because it still fits my needs and budget the best. I don't use it to model anymore though, not since Modo 201. I have a love/hate relationship with LW and I will continue using it until my other tools (like Modo) phase it out completely, which will happen if NewTek doesn't start doing something serious to get back in tune with what users want. The other problem with LW is that you're locked into Worley's plug-ins for much of the advanced functionality and enhanced abilities. Unfortunately, Worley is extremely slow at developing compared to others in the market these days. The shortcomings of LW's SDK and slow development of FPrime and Sasquatch along with their limitations over at Worley has shifted me over to a Modo/XSI workflow for most complicated projects involving hair.