Page 3 of 15 FirstFirst 1234513 ... LastLast
Results 31 to 45 of 217

Thread: Lightwave 12 and Lightwave's future

  1. #31
    Goes bump in the night RebelHill's Avatar
    Join Date
    Nov 2003
    Location
    jersey
    Posts
    5,770
    Quote Originally Posted by 3D Kiwi View Post
    They would have to do something pretty special to compete esp in the highend market where i guess the money would be. And it would still have to deal with the reputation that the name lightwave would bring.
    The "special" is the unlimited render node offering. You keep that, and its a full on broadside against those other renderers.
    LSR Surface and Rendering Tuts.
    RHiggit Rigging and Animation Tools
    RHA Animation Tutorials
    RHR Rigging Tutorials
    RHN Nodal Tutorials
    YT Vids Tuts for all

  2. #32
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,507
    Quote Originally Posted by saranine View Post
    Newtek leaves Lightwave 12 pretty much the same as Lightwave 11. Autodesk will be waiting to pounce. Within a year Autodesk will buy up Lightwave.
    I agree with you that LW12 will make or break LW3DG, but I disagree with the fail endgame. What is it that you imagine Lightwave offers that Autodesk wants/needs?
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  3. #33
    Super Member H_Molla's Avatar
    Join Date
    Feb 2003
    Location
    Cairo, Egypt
    Posts
    349
    Quote Originally Posted by jwiede View Post
    I agree with you that LW12 will make or break LW3DG, but I disagree with the fail endgame. What is it that you imagine Lightwave offers that Autodesk wants/needs?

    In my opinion, what lightwave offer that autodesk might be interested with is the fast qality of the built in rendering and the way lightwave displacement work.
    Also the way node's work with each part and how can non technical people start do work with it as1 2 3.

  4. #34
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,507
    Quote Originally Posted by RebelHill View Post
    LW has a fantastic renderer, it has a fantastic sufacing and material system. Such good quality, so easy to learn and to use, its the one thing that sets LW apart from the competition more than anything else. So with that, Id turn LW into nothing more than a plugin renderer for maya, max, houdini, you name it. Get it working with as many other content creation apps as possible. Make it the most hardy, mean production render you can. Pour all future effort and development resources into that, gaining some added acceleration and advancement along the way, and when it was done... continue to sell it for around a grand or so, WITH the current offering of unlimited render nodes.
    I just can't see LW selling as an independent rendering plugin without first adding buckets. It's a feature ALL of its competitors already offer, and the inability to split work across buckets across render nodes (something, again, most of its competitors offer today) would put it at a serious competitive and performant disadvantage.

    I doubt most potential customers will believe LW's surfacing is "fantastic" given just the existing content and materials available (for good reason). All evidence seems to suggest that providing documentation and other supporting content is something LW3DG do poorly (to be clear: when they do it they can do it well, but they refuse to do enough, so net result is inadequate==poor). LW3DG would also need to put serious effort into providing a catalog of production-quality modern materials, useful compound nodes, and so forth -- I'm a bit doubtful they have that kind of effort in them, even if their survival depended on it (and in part because they don't ever seem to believe it does).

    That said, I DO agree with you that the scenario you propose is about the only realistic one left. Based on the productivity shown over last few years, and what we've received as customers, their chances of adding enough fast enough in order to regain a frontal position in the market is exceptionally unlikely. I hoped otherwise for a long time, but when I realized the years of productivity shortfall manifested Chronosculpt, instead of some immense LW retrofitting effort behind the scenes, my hopes mostly died, alas.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  5. #35
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,507
    Quote Originally Posted by 3D Kiwi View Post
    They would have to do something pretty special to compete esp in the highend market where i guess the money would be. And it would still have to deal with the reputation that the name lightwave would bring.
    Even in the rendering market, they're technically well behind the pack: Buckets, working caustics, micropoly displacement, more efficient/performant GI solutions (LC, etc), importance mapping, OpenSubDiv (Maxwell, Vray, and others have added or are now adding in-renderer support) and so forth. Just adding bucket rendering alone (a "must have", IMO, since it gates so many other aspects, like work-splitting and efficient memory usage) would be a major internal rework based on what the SDK indicates the internals look like today. And that's not even counting the necessary backing catalogs of production-level materials, documentation, etc. that all the others already offer, and which would need to be available for LW to actually compete.
    Last edited by jwiede; 06-23-2014 at 10:46 AM.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  6. #36
    I like Rebel Hill's take on going to a renderer but i dont see NT going for that one. It might be interesting for them to do a render engine for other apps in addition to continuing to develop LightWave...I just dont if it would be realistic for them to do so with their resources.

  7. #37
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,507
    Quote Originally Posted by RebelHill View Post
    The "special" is the unlimited render node offering. You keep that, and its a full on broadside against those other renderers.
    What kind of pricing are you imagining for LW-as-renderer such that LW3DG can continue to give away free unlimited render nodes? That approach doesn't seem very sustainable as a business model (esp. given the current trends in cloud-based and commercial farming).

    After all, most current third-party renderers already give 3-5 free render nodes per full license seat, that's more than enough free nodes to cover most indie and small to medium business situations. They only really care about paid render nodes in large installation situations.
    Last edited by jwiede; 06-23-2014 at 10:38 AM.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  8. #38
    Axes grinder- Dongle #99
    Join Date
    Jul 2003
    Location
    Seattle
    Posts
    14,732
    Craig, you no doubt use.. what are they called? "include files"? ...to ease your way in programming. That's all I'm suggesting with my "kindergarten/tinkertoy/Mechanno" compound nodes scheme: training wheels for nodal users.

    Now, you may or may not be correct in the stance that networks are too particular to specific productions to make this worthwhile, but it's worth a conversation. Certainly the few networks I've saved have come in convenient over and over: if I watch another tutorial with someone creating Yet Another Rainbow Gradient I may scream. And limitations can be listed in the Node Comment field.

    And in no way did I suggest the removal of the nuts and bolts, nitty-gritty nodes that would make up these tinker toy blocks: in fact, I'm counting on them to be 'nodal gateway drugs' that people would customize to better fit their specific production needs.

    We can have it both ways: all the current nodes, AND the shiny tinker toy easy conglomerations. We don't expect electronics workers to work with AND, NAND and NOR gates any more.
    They only call it 'class warfare' when we fight back.
    Praise to Buddha! #resist
    Chard's Credo-"Documentation is PART of the Interface"
    Film the cops. Always FILM THE COPS. Use this app.

  9. #39
    Eh, what?
    Join Date
    Mar 2003
    Location
    Right here, right now
    Posts
    604
    Quote Originally Posted by RebelHill View Post
    Well... here's my take...

    ...The softimage story is the one to take to heart here... It fell behind, and despite roaring back with an AWESOME offering... it was unable to make ground, and died... Whattaya think LWs chances are by compare??
    But wasn't XSI, especially in its latest iterations, aimed at a rather niche market? I mean - Max and C4D are both "generalist" apps that evolved to cover what, 90% of the architectural rendering /motion graphics markets, and that's an awful lot of people. Hell, "traditional" graphic designers here consider Cinema as the _only_ possible solution - and apart from the software's innate strenghts, that's also because of the enormous amount of readily available assets. Which, incidentally, has been one of LW's major shortcomings for the last, say, 10 years or so.

    What Id keep... would be the one thing that has been LWs shining strength, both in the past, and today... its renderer.
    I... Don't know about that. In its current state, LW's renderer wouldn't be very palatable, I fear: other solutions offer more features, the same speed if not more and, often enough, less hassle (plus the aforementioned assets). You are probably biased by your own hyper-competence, but when it comes to surfacing why would the average, one-man-band 3D artist want to learn how to connect 20 or so nodes in order to get a nice-looking brushed aluminum, when in a different engine it's just three clicks away?
    Of course this might - let's say will - turn into a non-issue when they release the new LW engine and it turns out to be three times better than Vray and Octane put together - yay! But at that point, ditching the app altogether could alienate the relatively few diehard users (it sure would alienate me) with no guarantee of gaining new ones through the "plug-in" scheme... Rock and a hard place indeed.

    ADP.

  10. #40
    Goes bump in the night RebelHill's Avatar
    Join Date
    Nov 2003
    Location
    jersey
    Posts
    5,770
    Quote Originally Posted by jwiede View Post
    What kind of pricing are you imagining for LW-as-renderer such that LW3DG can continue to give away free unlimited render nodes? That approach doesn't seem very sustainable as a business model (esp. given the current trends in cloud-based and commercial farming)..
    More or less around current pricing... If they can survive presently selling LW around the grand-ish mark, with a few hundred for upgrades... then they can survive doing the same if they sold a stand alone renderer that increased their market share... even more so given the lower overheads which would be involved in developing and upkeeping such a "smaller" product.

    As for the "LWs render doesnt have do this presently"... arguments... I DID specify that some months of hardcore (no pun intended) development to make it a true production beast would be needed before such a release.
    Last edited by RebelHill; 06-23-2014 at 10:46 AM.
    LSR Surface and Rendering Tuts.
    RHiggit Rigging and Animation Tools
    RHA Animation Tutorials
    RHR Rigging Tutorials
    RHN Nodal Tutorials
    YT Vids Tuts for all

  11. #41
    Goes bump in the night RebelHill's Avatar
    Join Date
    Nov 2003
    Location
    jersey
    Posts
    5,770
    Quote Originally Posted by jeric_synergy View Post
    Craig, you no doubt use.. what are they called? "include files"? ...to ease your way in programming. That's all I'm suggesting with my "kindergarten/tinkertoy/Mechanno" compound nodes scheme: training wheels for nodal users.

    Now, you may or may not be correct in the stance that networks are too particular to specific productions to make this worthwhile, but it's worth a conversation. Certainly the few networks I've saved have come in convenient over and over: if I watch another tutorial with someone creating Yet Another Rainbow Gradient I may scream.
    I presume you mean "libs", but, actually... no. In python I write EVRYTHING from the ground up (development from first principles you might say)... and in LS Ive written my own wrapper for the existing sdk that fits the things Im creating and makes the "higher level" coding more concise, flexible and take up fewer lines. (as a refernce for this, RHiggit V1 used near 200,000 lines of code, V2, which is vastly superior, is less than 10,000). Now obv... thats just me, and it may not be the most recommendable way for others to go working... but creating your own adaptations that work specifically for the things you're creating WILL give you efficiency savings.

    And so sure... nothing wrong with compounds, presets, whatever you like... but again, the number of possible, useful things you COULD potentially ask for is near infinite. How do you begin to approach that?? Obviously, you need to start by prioritising... but how do you do that??

    I submit that the best way is to allow users to do it for themselves, and to SHARE these outcomes (which ofc you can do via .nod files). Sure, if certain ones are more "common" or useful, by all means, native-ise them (which is what material nodes are, which is what some of the math operation nodes are, etc)... but you CANNOT expect it to be all things to all people... and what you CERTAINLY cannot do, is to go renaming things, and overriding established conventions to try and match the lowest common denomeator (yes, that's a math pun, sorry).

    The dot product, whilst delivering a comparison between vectors... is only ONE kind of comparison, others are possible with other methods... Do you not think that having nodes called, "vec compare type 1, vec compare type2" etc would be even MORE confusing?

    Once again... you HAVE to try and work within standard known and established conventions, not try and reinvent them.
    LSR Surface and Rendering Tuts.
    RHiggit Rigging and Animation Tools
    RHA Animation Tutorials
    RHR Rigging Tutorials
    RHN Nodal Tutorials
    YT Vids Tuts for all

  12. #42
    the only competitive "edge" lightwave can really pull out is pricing.

    If it remains the lowest cost "full" 3d app then there's always a market for it depending on how close the competition is pricing their own apps.

    however there's another card to lean on...it's not owned by a huge corporation...you're less likely to get done over in some licence change eula on an update.
    render node licence capabilities is another good card compared to arnold and vray but the renderer in lightwave needs a large overhaul now too though.

    lightwave really could do with adding GPU rendering....hint flippin hint....and a bucket renderer...micropoly displacement....i'm sure many more people can chime in.
    Last edited by cresshead; 06-23-2014 at 11:20 AM.
    stee+cat
    real name: steve gilbert
    http://www.cresshead.com/

    Q - How many polys?
    A - All of them!

  13. #43
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,507
    Meh, never mind, I've talked myself out of this idea's viability, because on top of the technical issues already mentioned, LW3DG would also need to provide tightly-integrated plugins for all the major packages to support Lightwave as a third-party external renderer. That's a lot of work, esp. as most require material integration, translation of various geometry types that LW doesn't support native, and so forth. Even if they just took Max or Maya, as the most rich packages, they'd individually still be a huge amount of integration work.

    In any case, weren't a few big studios already more or less using LW in precisely such a "render-only" capacity, but have since stopped doing so? If their stated reasons (which seemed to focus on render tech features and overall quality) were legit, there's little reason for them to misrepresent them, so that seems to more or less confirm that LW's "immediate" viability as a commercial third-party render engine in its current state is questionable. Which brings us back to chasing the pack w.r.t. technical features, and the likely losing battle therein.

    I believe there was a time when repackaging LW as a render engine would have been a viable plan. Unfortunately, I strongly suspect those studios ceasing to use LW marked the end of that period -- that was the "tipping point" where staying with LW-as-renderer became competitively prohibitive compared to other external renderers' offerings.

    P.S. Forgot render proxies in the technical "missing features" list, that's a huge one for arch-viz and certain other genres.
    Last edited by jwiede; 06-23-2014 at 11:30 AM.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  14. #44
    Goes bump in the night RebelHill's Avatar
    Join Date
    Nov 2003
    Location
    jersey
    Posts
    5,770
    Quote Originally Posted by jwiede View Post
    ...weren't a few big studios already more or less using LW in precisely such a "render-only" capacity, but have since stopped doing so? If their stated reasons (which seemed to focus on render tech features and overall quality) were legit, there's little reason for them to misrepresent them, so that seems to more or less confirm that LW's "immediate" viability as a commercial third-party render engine in its current state is questionable.
    A large part of the issue has also been the workflow of having to exchange data between the apps, and the fact that you had to re-do certain things exclusively in LW... If you have it as a plugin, which is integrted, you dont have that issue. The lights, the material system, etc, are all in the host app, just like the other plugin renderers.

    And again... Im not suggesting doing it with the renderer in its current state as of today... youd need to "heighten" it, take it forward, and pour the development resources into doing that which would enable you to get to the necessary point much much faster.
    LSR Surface and Rendering Tuts.
    RHiggit Rigging and Animation Tools
    RHA Animation Tutorials
    RHR Rigging Tutorials
    RHN Nodal Tutorials
    YT Vids Tuts for all

  15. #45
    Quote Originally Posted by RebelHill View Post
    A
    And again... Im not suggesting doing it with the renderer in its current state as of today... youd need to "heighten" it, take it forward, and pour the development resources into doing that which would enable you to get to the necessary point much much faster.
    OK, let's pretend for a moment that this scenario (LW s rende ronly) is viable (although i highly doubt) how do you see LW current DEVS all suddenly start workign on Rendering code when they are more or less specialized for diferent aspects (modeling, surfacing, animating, Gui, I/O stuff, Languages (phyton,Lscript..) etc. etc.). Should LWG3D fire 90% of it's current DEVS and keep just ones who are specialized in coding render engine for last decade or more or what ?
    RAM-Studio
    WS - Dual Xeon E5-2698v4/128GB/Win10x64/x TitanX(M) + 2x GTX 1080Ti + GTX 1080
    My LWM Video Car Modeling Tutorial

Page 3 of 15 FirstFirst 1234513 ... LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •