PDA

View Full Version : 3D Max Nitrous to Core Parallels



Andyjaggy
05-14-2010, 07:56 AM
http://www.maxunderground.com/archives/011007_nitrous_putting_together_the_pieces_of_a_ma x_core_rewrite.html/comment-page-1#comment-1074

A breif history of some events that happened in the 3D Max world, that strike a startling similarity to Lightwave and Core.

crashnburn
05-14-2010, 06:04 PM
Yeah, the part about NDA stood out the most.

probiner
05-14-2010, 06:29 PM
Auch!

(reads a bit more)

OOOuuuuhhh!


Good Luck Newtek!!!

Ňgrťn
05-15-2010, 02:33 AM
I'd love to see that Nitrous, lol. Whata hogwash... Already seen the Invention of Lying. Already seen chronic liars in high positions too. All bow to ArthurDesk if the excalibur comes out from the old dead rock ! :D

cresshead
05-15-2010, 09:55 AM
"my conclusion was (when I was asked by somebody of Autodesk on the project) was that at this stage I would prefer them to fix the issues in 3dsMax instead. Mainly because it felt like Nitrous was no longer on track of becoming a complete 3d package. And to be honest, thatís all I wanted it to be. I didnít need another application dropped in the pipeline that we didnít really ask for in the first place."

yep its pretty wierd hey

indeed, there's some VERY similar things that appear to be going on with the lightwave suite.

Andyjaggy
05-15-2010, 01:03 PM
Yeah I was frankly startled at the similarities. Eventually they just killed it because everyone just wanted the Max they knew and loved, but improved, not a new application that wouldn't be Max. Sound familiar? Ha. NT is certainly in a tough spot, I hope it all works out for them.

hrgiger
05-15-2010, 08:28 PM
There's certainly some similarities in the situation. Newtek has to strike a balance between delivering a next generation application like CORE and keeping current users of Lightwave happy with new improvements and additions to the Lightwave that they're familar with. Which is pretty much inline with the recent announcement about a change in Lightwave development. I think they know they just can't have Lightwave languish while CORE is being brought to full application status, especially if it's going to take more time then they anticipated.

As long as CORE hits it's targets set in the features document put out last November for it's first release, they'll be in good shape for the next version. The features outlined in the release hit on all the major points of a solid 3D application- modeling, UV mapping, dynamics,shading, lighting, rendering, modifier stack,nodal workflow, scripting, SDK, animation and compositing. And while some of the features are just initial implementations, I would think in the second version you'll see a lot of refinement in all the key areas.

It's interesting to note that what they tried to accomplish with Nitrous, they are now doing again with XBR. Only now instead of a complete rewrite of the software, they are replacing major portions of the current Max code. So if you count when they started working on a rewrite with Nitrous, they've been wanting to do a rewrite now for over 10 years and XBR still has at least a few to several more years before it's completed. Maybe if they had kept going with Nitrous, they would have had their new application by now.

Whether or not the situation is similar, I still believe a rewrite of Lightwave is the way to go. So many inherent flaws in the 2 application worklfow and the oudated architecture. Even if they went the XBR route and just replaced code, they'd probably end up redoing most of it anyway. I think it's been mentioned that they realized it would be less work to start from scratch. All of the work that went into the 9.x rendering engine required replacement of a lot of the rendering/raytracing code and how long did that take?

Dexter2999
05-15-2010, 09:22 PM
Yeah I was frankly startled at the similarities. Eventually they just killed it because everyone just wanted the Max they knew and loved, but improved, not a new application that wouldn't be Max. Sound familiar? Ha. NT is certainly in a tough spot, I hope it all works out for them.

That is a part of the key difference with the LW/CORE. The LW community has been divided for years over split app. vs. unified app. So, at least some portion of the LW users will be made happy with this transition.

Also, people who can afford MAX could just as easily jump ship to XSI or MAYA as they are in comparable market range as far as cost goes.

There are many LW users who can't afford to make that jump up to another app. so easily. So they will be prone to being more patient.

I am not saying that they will retain 100% of their user base. But I think there will be a considerable portion of market retention and while they add some things to make the app. friendlier to Maya users perhaps an influx of new users to the program (though I don't think I would expect them in the 1.0 release phase.)

Although the article has MANY similarities. The circumstances are not identical because of the demographics of the LW user base compared to the MAX user base. I think Newtek has a little edge (not a huge one) that gives them just enough leeway to make this move.

It is still a precarious journey any way you look at it though. And there will be some attrition in the userbase because you can't make everyone happy.

cresshead
05-16-2010, 03:55 AM
i can see some valid points of view in that if the new app is "totally new" with no link to the past app then your in a position that learning this new app or ANY new app is the same starting point...so what's so special about Nitrous up against maya, xsi or houdini back then...

so long term going XBR internal upgrade/rewrite keeps your loyal users onboard the app as it grows from the inside out.

the hardest part of buying a app is deciding "which" one to invest in when you start out...so keeping users is a priority if the app basically remains the same structure they bought into in the first place or have to drop it for some totally new one cos the developer says it's the end of the road for that app...

the toughest part about core is getting it a "NEWTEK feel", not maybe a layout or a modeler feel as it's a unified app so that's always going to be a tough bridge to cross on how to handle stuff without looking too maya/modo/max like.

Newtek have put quite abit of dev time into getting a lightwave like navigation system up and running for example in CORE.

Titus
05-16-2010, 08:43 AM
Interesting article. But Nitrous' story is more Duke Nukem Forever than Core. I think Core is closer to Modo in this regard.

archijam
05-17-2010, 01:26 AM
Because it's big, and grey, don't make it an elephant.

Such cautionary tales are useful as just that.

One of the key aspects of core is skinability. It is not hard to imagine a lw skin and workflow that can allow a new engine that can look and feel like the old one, but perform at a new level.

.. Or look and feel like max! A big part of winning new users in this marketplace will be PERFORMING differently, but feeling familiar. This is where the time spent in maximising the qt benefits can really pay off.

erikals
05-17-2010, 05:26 PM
to me it looks like NT has done a a great job with Core,
you have to remember that it's tons of code to write.

C++, QT, Python, Powerful SDK

NT did the jump at the right time.
however, that's a damn difficult jump for any of the big 3 3D apps to make.
(only Si has done it so far, but didn't quite make it, and ended up being bought by AD)

probiner
05-17-2010, 05:46 PM
.. Or look and feel like max! A big part of winning new users in this marketplace will be PERFORMING differently, but feeling familiar.

archijam i can't stop laughing at this part, it's silly i know, but i can't...
it's like you were talking of a man that cheats on his wife with a specific woman, cause.. she performs differently, but you know?... there's something in there that feels familiar... so... it's ok :D


Erikals, some characteristics that you are pointing have, precisly, parallels with the article.

Anyway, Newtek is Newtek.

Cheers

Lightwolf
05-17-2010, 05:56 PM
however, that's a damn difficult jump for any of the big 3 3D apps to make.
(only Si has done it so far, but didn't quite make it, and ended up being bought by AD)
Well, Max and Maya made that jump quite some time ago. And LW 5.6->6.0 was a mere hop compared to that.
C4D seems to somehow manage to have some continuity in development though.

Cheers,
Mike

erikals
05-17-2010, 08:29 PM
didn't know, must have been a loong time ago?..

cresshead
05-18-2010, 01:27 AM
to me it looks like NT has done a a great job with Core,
you have to remember that it's tons of code to write.

C++, QT, Python, Powerful SDK

NT did the jump at the right time.
however, that's a damn difficult jump for any of the big 3 3D apps to make.
(only Si has done it so far, but didn't quite make it, and ended up being bought by AD)

well sort of....

3d studio made the jump to 3d studio max
alias power animator and wavefront's app made the jump to maya.

colkai
05-18-2010, 02:08 AM
That is a part of the key difference with the LW/CORE. The LW community has been divided for years over split app. vs. unified app. So, at least some portion of the LW users will be made happy with this transition.

Well yes, but then again no, as the divided app has now sub-divided, so it is a 3-app suite. Once Core finally becomes a single app, we can talk unified, but Core to LWX modeller / layout is no more unified as such as Layout to modeller via the hub.

This for me will be the interesting aspect of how Newtek eventually try to market LW with Core, pursuading people that a 3-app suite is better than a 2-app suite over something like simply moving to XSI.

I'm still enough of a LW fan to hope they do pull it off, but I can see it being a tricky thing to do.
Not so much for the LW faithful, they will stick with LW anyway, but more for attracting folks back who previously moved away from LW and new markets.

cresshead
05-18-2010, 01:28 PM
personally i think that making lightwave a 3 app thing when it had enough trouble selling as a 2 app thing compared to unified apps like softimage, maya, blender and cinema4d..splitting a project to 3 3d apps of which none are sculpting apps...so 4 apps if you need some sculpting is getting to be a pretty silly prospect for marketing it as a move forward.

i can see quite a few people just deciding to "move on"...keeping their current lightwave going as they learn their new 3d app such as cinema4d, houdini, blender, modo or wot not.

my question has to be... was whoever mis informed when they said lightwave was at a dead end nd it was time to create a new app...seems that some people think there's still roads to be traveled down in lightwave and it's not a dead app afterall..

was core a wrong turn?...or just a turn taken too early..??...or not given enough manpower to bring it online fast enough.

i'm also REALLY tired of the vague answers from some departments on such things...i'd prefer either a straight answer or we can't tell you...compared to the fog/mist we get usually.

i suppose we'll have t wait n see what shakes out...though nothing ever happens quickly..!!

erikals
05-18-2010, 02:03 PM
a programmer for Modeling Tools,
how long has that position been available?

think of all the Modeling Tools that could have been available in Core if they just
had hired a Core Modeling Tools Programmer 1 year ago.

hrgiger
05-18-2010, 02:22 PM
Well yes, but then again no, as the divided app has now sub-divided, so it is a 3-app suite.

Well, yes, but then again, no.:) CORE was always meant to rely on Lightwave and modeler for the first version or two, this new 'change in development' hasn't changed anything as far as that goes. The suite thing, which I understand isn't the final nomenclature, is just putting positive spin on what is unfortunately, the way CORE has to be presented until it is a single application. The only change as far as I'm aware of, is that they will continue to add more functionality to modeler and Layout until CORE can stand on its own. Which is good news for current LW users who do not yet see the appeal of CORE and need something that's production ready today. I didn't like the idea at first of not putting all the focus on CORE, but I suppose I'm coming around to the idea that Lightwave simply can't languish for 2, 3, 4 years until CORE replaces all LW functionality.



This for me will be the interesting aspect of how Newtek eventually try to market LW with Core, pursuading people that a 3-app suite is better than a 2-app suite over something like simply moving to XSI.



Well, if you're speaking about current Lightwave users, I suppose, in general, that they will stick around if they see that the Lightwave they already know receives some new features and improvements that makes upgrading worthwhile. CORE should be seen as a sweetner for this first release and will hopefully provide some of its useful additons to a LW pipeline. We'll just have to see how useful the Bullet physics, UV mapping and other CORE tools are and how easy they are to combine with current modeler and Layout.

Dexter2999
05-18-2010, 02:23 PM
CORE was not a wrong turn.
CORE is where LW has to end up to stay competitive in the 3D market.
They have to get away from the old code and try things to attract a larger user base.

What went wrong?
Over optimistic on the timeline in which CORE would a viable standalone application.
IMO CORE should have been designed from the get go as a modeler replacement. This would have slowed down developement because in addion to new features they would have spent alot of time making it backwards compatible.

When CORE hits the market in 1.0. It is going to be in its infancy and probably be about as productive as infant as well. Meaning, eat, sleep, poop. Yes babies are neat little machines but not renowned for their productivity or usefulness. What they are is potential and that is what CORE most likely will be.

And to COLKAI:
LW has not subdivided, CORE is not a requirement to continue using LW. LW will stand alone without it. CORE will be available for those that want to dip their toes into where LW is going to be in three or four years...

Newtek doesn't have to convince anyone not move XSI or MAYA. I pointed that out in my first post. All they have to do for the majority of their user base is provide basic functionality at a price point that lets people with little money get their 3D fix.

I know people will argue Blender. BUT I find that most people have never heard of Lightwave and almost no one I talk to outside of 3D artists have heard of Blender. So, that doesn't really present itself as a viable professional alternative if people just getting into 3D haven't even heard of it...much less find an established user base for the software.

All Newtek has to do...is nothing. They can do absolutely nothing and as long as their pricing is where it is people who can't afford to spend more will buy it. And that is independent shops and hobbiests.

Money is perhaps the second most important factor when people are making a purchase.
FIRST: What will it do?/ Does it do what I need to do?
SECOND: How much is it?

CRESS: Getting a bit carried away there. You have to buy sculpting app for MAYA and XSI as well. MUDBOX doesn't come included. Even if it was "included" doesn't make it built in. So that is a moot point.

I understand user frustrations. But really, most users here don't seem to get business. It isn't entirely predicated on the features users want. There have to be trade offs against need vs. want. There needs to be a balance between feature value and product pricing (this is where the LW users mostly get lost), and there needs to be an influx of income to keep developement on track.
If it makes you guys feel any better...you are much better artists than I am.

erikals
05-18-2010, 02:38 PM
...that they will continue to add more functionality to modeler and Layout until CORE can stand on its own. Which is good news for current LW users who do not yet see the appeal of CORE and need something that's production ready today...
not sure, depends on what they add...

hrgiger
05-18-2010, 02:59 PM
not sure, depends on what they add...

Well, and I agree. I wanted to see them focus on getting CORE up to speed rather then divert time away from that to add more features to modeler/layout. But I do see the need to keep making useful additions to the Lightwave that exists now until CORE is that great new app.

cresshead
05-18-2010, 07:44 PM
1,2,3,4.

1:blender...does everything to a degree including sculpting

2:max+mudbox
2:xsi+zbrush
2.cinema4d+3dcoat
2.maya+modo

3.layout+modeler+blender
3.layout+modeler+zbrush

4:layout+modeler+core+3dcoat
4:layout+modeler+core+blender

Blender's rewrite seems to have gone pretty smooth so far..yeah the workflow/shortcut's arn't all in yet but it's looking pretty useful already.

Dexter2999
05-18-2010, 08:27 PM
Yeah, I'm betting most LW users don't need hair or fluid or even CA.
Because people who were looking for those things chose another package to begin with.
I'm not saying they don't want those things. They want as much as they can get for as little as they can get away with paying for...that's natural.

I actually think the average LW user would get more milage out of a robust multi pass rendering being built in than any of the above mentioned things.


There are things missing from LW that are why it was so inexpensive to begin with.
No hair
No fluids
No sculpting
No multipass rendering
No autorigger

I have been using LW professionaly for almost 7 years and never used any of those things. And I'm glad I haven't paid extra for things I don't use.But now Newtek is trying to get back in the high end (film) market and they need to come up with these things to get there. And the retail price will go up accordingly.

I know I just do spinny logos, but I swear I could about as well as some of the crap I see on TV (VegiTales, Mickeys Clubhouse, Nasonex Bee, General Car Insurance, Scrubbing Bubbles) without any of the advanced features that are being planned for CORE.



NEED vs. WANT.

colkai
05-19-2010, 02:27 AM
to COLKAI:
LW has not subdivided, CORE is not a requirement to continue using LW. LW will stand alone without it. CORE will be available for those that want to dip their toes into where LW is going to be in three or four years...

But then, what's the drive to move to CORE? Core is where all the new action is, sure there will be added features to LWHC, but they are not going to put too much effort in there if the end aim is to remove LWHC completely. that is also a business thing, why put months and months of work into a program you intend to kill in the forseeable future?

As I said, the LW faithful will adhere and wait as long as it takes and yes, Newtek do not have to do a darned thing. They will release what they do when they do, with the features they decide they want in the software, if that isn't what people were expecting or hoping for, that is life.

Personally, I bailed so I can wait and see if what is released is anything like I'd hoped for and need, so far, the vagueness still leaves me (and obviously others) uneasy.

I've always held, Core 1.0 is a big challenge and if those who are passionate about it expect it to be a product very much in its infancy when it is release, I really do think it will be interesting how one markets such a product. Somehow, I doubt, "trust me, it'll be good in a couple of releases" isn't a marketing slogan for the masses.

Yes, I understand, eventually it will an uber-powerful replacement for LW and no-one questions that the re-write MUST be done, but over optimistic is a very polite way of stating how it's fell so far.

I don't feel the need to dip my toe into an upgrade which is almost a proof of concept of the new LW technology, if it isn't usable and stable in a day to day environment, it's not a purchase worth paying for.

Dexter2999
05-19-2010, 08:43 AM
Colkai:

I don't think you are going to be all that taken with CORE 1.0.
What is the motivation to use it?
Well, when CORE 2.0 rolls out there won't be a choice because it will most likely replace Modeler. (I don't know that for a fact just conjecture.) So it may be worth at least playing with to learn what's what.

The good news (to me) is that LW is going to be getting a little love to make it compatible with CORE and it is getting the benefit of some of the R and D done for CORE. I think you will like that part at least.

I don't forsee too many converts with CORE 1.0. But I think it will get people's attention. It will have many people waiting on the edge of their seats for 2.0 (or 3.0).

Matt
05-19-2010, 09:03 AM
NewTek are not Autodesk (thankfully).

;)

Lightwolf
05-19-2010, 09:08 AM
NewTek are not Autodesk (thankfully).

;)

No, but it also means they can't really afford their Nitrous either.

Cheers,
Mike

robertoortiz
05-19-2010, 09:24 AM
Colkai:

I don't think you are going to be all that taken with CORE 1.0.
What is the motivation to use it?


From the point of view of a developer there is lots of things CORE offers. And fro mteh point of view of a small studio (the kinds that had a small dev department) the opne architecture of CORE,and it full support of python might prove attractive.


NewTek are not Autodesk (thankfully).

;)

And this is another reason why people will be curious about CORE. Nt would be wise to push in SIGGRAPH its open architecture as a key selling point.

colkai
05-19-2010, 09:26 AM
I don't forsee too many converts with CORE 1.0. But I think it will get people's attention. It will have many people waiting on the edge of their seats for 2.0 (or 3.0).

Yep, I agree, I feel Core 1.0 isn't going to draw the crowds as it were. My worry then is, by the time Core 2.0 or 3.0 finally comes around, will there be the market to draw back?
I know it's all supposition really but I really do wonder if more work should of been done before we even got a hint about Core, right now, it's been 2 years in design / build and over a year in beta and still a long way off being a product to push as production ready.
It's tough I'll grant with limited resources and one can only hope that Newtek pull it all together in the end. :)

Roberto,
To be honest, yes, as a dev, there may be things that look sweet in CORE, but unless it's a viable usable product, interest may just be that, curiosity, with a "Hmmm, that looks like it could be good eventually", not necessarily a "must buy that".

The open architecture is not much use if once you plug into it, the product doesn't do what you need it to. Again, all vagueness and as always, I guess we will see what we see, when we see it, whenever that may be.

robertoortiz
05-19-2010, 09:44 AM
Newtek has one BIG advantage over Autodesk.
The ornet nest it stirs up with the re-write is considerably smaller than the one that Autodesk will be dealing with.

With the CORE re-write you were dealing with a significant smaller user base, and that was (and is a pain) for the developers.

NOW imagine a user base that is way bigger, and having to satisfy the needs of 3 differents apps (XSI, Maya and Max).

I dont envy them at all.

Nemoid
05-19-2010, 09:56 AM
IMO : the first and more important features for an app suited to stay in the market now, are openness, programmability and nice handling of large scenes/ amount of textures.
Openness and programmability are what mainly sold Maya into big studios, because they use it as the base of their pipeline. Plus, it has some nice animation capabilities and handling of large scenes. n facts, Maya, as it is out of the box, has a slow workflow if not customized. It starts shining when you add MELS and pyton scripts to it and to host/connect with proprietary tools.

BTW this is what the big/medium studios are used to use now, plus other commercial packages helping them to complete tasks faster, like for example Lw was for modelling and rendering, Modo, and other apps.

what can be needed in the market is something with the openness and programmability of Maya and great ease of use without having to script/program. And hopefully not being too pricey.
Something like XSI is actually very near to this definition (unfortunately it's sold by AD, but it works nicely for sure) along with Modo, which is incomplete tho.
I also read in other forums that even XSI is not as open as Maya is.
So very probably, the main point is the more open the software is, the better.

Nemoid
05-19-2010, 10:07 AM
Newtek has one BIG advantage over Autodesk.
The ornet nest it stirs up with the re-write is considerably smaller than the one that Autodesk will be dealing with.

With the CORE re-write you were dealing with a significant smaller user base, and that was (and is a pain) for the developers.

NOW imagine a user base that is way bigger, and having to satisfy the needs of 3 differents apps (XSI, Maya and Max).

I dont envy them at all.

LOL me neither. On one side, the advantage of this is covering differet fields: Maya = movies, MAX= games, XSI = TV maybe, but basically all the 3 apps can do several things in every area so they intersect a bit each other, even feature wise. With this method , tho, you sell 3 apps instead of one and so you have a larger userbase both established and potential.

BTW you even have to satisfy them, and features couldn't be coded into a coordinated way, so that the 3 apps will be a bit of a different world one for the other.

Newtek is into a differen position. They mainly produce and are leaders in the field of hardware and software for TV production (and its the main Lw market as well), then they Make Lw which demonstrated to be more capable than expected, in good hands, even because of the rendering engine. So, i think Nt is not as big as AD, but could develop CORE quite well.

probiner
05-19-2010, 11:23 AM
Well, when CORE 2.0 rolls out there won't be a choice because it will most likely replace Modeler.


One thing it was not stated yet, it think, it's precisely about this, what will CORE replace or add to the LW pipeline, for now. Sure in the future it will replace LW all around, but what for the next year?

"Replacing modeler", doesn't sound right. Because that's the old way of thinking (and it will sure be there as long as LW is with CORE tech and not alone).
I think it would be easier (yeah sure, i know about things...) to replace tasks and not applications. So, from what i saw of CORE until now, it has VPR, good handling of Subdivion surfaces and the sweet render engine we all like. Conclusion; it's a good scene handler.
So, personally i would go with rendering task replacement first. (not this rigid but here it goes) Model in modeler, animate in Layout, Preview and render in CORE.
If you think about it, that's probably what many ppl do with LW, they choose it for a specific task.

So what would be nice to hear from NewTek whats the task replacement order schedule.

Anyway, i don't know if all this LW + CORE integration doesn't give them more headaches than speed to make soemthing come out. It's not that they are making an old and new architectures have some compability features, they're working on making them rely on each other. That must some load of work.

Cheers

hrgiger
05-19-2010, 11:44 AM
So what would be nice to hear from NewTek whats the task replacement order schedule.



I'm sure we'll all have a better idea of how LW will be replaced once this first version comes out. It's easier to see what's left to do when we know what has already been done.

probiner
05-19-2010, 05:11 PM
Ok.
I'm just raising a point here of wich task could be replaced/improved by CORE, taking into account what features would make ppl to look at CORE as something that could be used right away in their pipeline. So even if the app is growing, it grows side by side with them.

CORE could start as a
Scene Management and Rendering app: the area so far where CORE seems to be more mature.
It seems to handle heavy geometry well; it has VPR, Instancing, the usual good render. So, replace Layout render task (if it comes wih Multipass, Progressive Refinement, etc., better)

Because Modeling (even with Parametrics, Snapping, Construction Plane, LWCAD confirmed) and Animation are the two big areas where most plugins, workarounds and tricks have been created and it's not easy to replace them right away, plus giving them the needed improvements.

ignorant, out! :D

colkai
05-20-2010, 02:50 AM
True enough, until we know what Core 1.0 is and what it actually does in the LW suite pipeline vs layout / modeller, it's very hard to come to any conclusions.

All I know is, at first, (as with the 9.x upgrade), there was much talk of modeller overhaul, then we saw physics, then we saw VPR, then we saw other things and what we didn't see was any significant movement on the modelling front. Heck, in some cases, there was movement but backwards.

Now the talk seems to have very much moved away from Core 1.0 being focussed on modelling and this again was a reason for stepping off the bus. I am sure Core 1.0 will have some cool features for some people, but I feel more and more that a Modeller "replacement" it won't be, (misguided as that thought may or may not be).

Andyjaggy
05-20-2010, 08:33 AM
I heard core makes a pretty wicked cylindar.

wfstecko
05-20-2010, 10:23 AM
I heard core makes a pretty wicked cylindar.

Thank you. Now I have an excuse to replace my soda-soaked keyboard.

W