PDA

View Full Version : Poll: How many users would like an embedded molecular modeller for LW?



inkpen3d
10-10-2012, 09:57 AM
The other day I came across a free embedded Python-based molecular viewer called ePMV (http://epmv.scripps.edu/) that has been developed by the Scripps Research Institute. In their words: "We have developed an open-source plug-in, embedded Python Molecular Viewer (ePMV), that runs molecular modelling software directly inside of professional 3D animation applications to provide simultaneous access to the capabilities of all of the systems".

ePMV, which is on a par with the excellent Molecular Maya plugin, enables users to load Protein Data Base file directly and then view/animate/render them within the given 3D application.

Currently, ePMV is only available for C4D, Maya and Blender. However, I have contacted one of the developers of ePMV, Graham Johnson at the Scripps Research Institute, and he reckons it would be possible to extend it to include LightWave.

Below are a few extracts from email conversations I've had with Graham Johnson on this subject:

"If Lightwave has a thorough and robust Python API, then it's possible that it can be wrapped with uPy which would enable ePMV and other plugins to function."
"Currently the plan is to [also] support 3DStudioMax, Houdini, Modo, then possibly Lightwave."
"ePMV and uPy are opensource, so if you know any lighwave users who are also programmers, we can guide them through what it would take to make it work sooner rather than later!"
"this is no task for Python hobbyists and especially not for novices - possibly for either - and would be a long slow process, whereas a pro-coder could get it up and running in a matter of a few days (a few hours if it's well wrapped in Python.)".

So, this poll is simply to ascertain how many LW users would be interested in having ePMV made available for LW.

PLEASE REMEMBER TO ENTER YOUR CHOICE IN THE POLL OPTIONS AT THE TOP OF THIS POST!

BTW, if there are any of you who are expert Python programmers and who are also experts in using the LW SDK and who would, under the guidance of the ePMV developers, like to take on the task of modifying ePMV to run under LW, please let me know and I will put you in contact with Graham Johnson.

Thanks in advance.

Peter

v1u1ant
10-10-2012, 10:01 AM
+1 for embedded molecular modeller for LW?

Cryonic
10-10-2012, 11:11 AM
I currently use PyMOL -> Blender -> Lightwave for getting molecules from PDB into Lightwave. One example of it is:


http://www.youtube.com/watch?v=IYYPPEtaBhs

If I could take them straight into LW then it would be nice. There used to be a plugin that did that, but it doesn't work with newer versions of LW :(

kopperdrake
10-10-2012, 12:38 PM
Now that's an interesting coincidence as I believe I will be asked to get some data out of a molecular model next week, and I *think* that may be PyMOL data. Is the work flow quite simple? I was going to see if Polytrans could do it, but I'm all ears to a better workflow!

inkpen3d
10-10-2012, 12:56 PM
FYI, PDB_Reader_v2.85.ls still works in LW Modeler v11.0.3 - it's okay for importing PDB files as point-clouds for HVs, ball-and-stick, or CPK type models.

It doesn't seem to be available on the internet any more, so I've attached it here for anyone who doesn't already have it.

Regards,
Peter

108478

kopperdrake
10-11-2012, 03:35 AM
Peter you dude - thanks! Saved and filed :thumbsup:

inkpen3d
10-11-2012, 04:54 AM
No problem! ;)

BTW, like your showreel.

johnliebler
10-11-2012, 11:11 AM
I would definitely be interested in seeing this in Lightwave! I have used Pymol and PBD reader, and more recently Chimera, but it is always a multi-step process with limitations.

inkpen3d
10-11-2012, 02:17 PM
OMG! John "Inner Life of a Cell" Liebler - you're my hero!

Really love your creations - I aspire to match your quality of work!

Enough said before I go and embarrass myself! ;)

johnliebler
10-14-2012, 11:29 AM
:) Thanks, Inkpen. Now I'm embarrassed. :)

Seriously, I'm surprised more people haven't responded to this poll. Lightwave used to be the best software for molecular animations, and a plug-in like this would go a long way to making that so again.

jeric_synergy
10-14-2012, 02:08 PM
Seriously, I'm surprised more people haven't responded to this poll. Lightwave used to be the best software for molecular animations, and a plug-in like this would go a long way to making that so again.
Seriously? How much worldwide demand can there be for accurate models, versus something that just looks accurate? Or even 'acceptable'.

kopperdrake
10-14-2012, 04:39 PM
I have to admit that whilst I would, at this moment in time, find it a useful plugin, this is the first time I've been asked to produce an actual molecular structure accurately from data. I don't know how many people produce this type of work on a regular basis?

Cryonic
10-14-2012, 07:04 PM
I can think of one group that needs accurate models... Researchers doing papers.

jeric_synergy
10-14-2012, 08:01 PM
I can think of one group that needs accurate models... Researchers doing papers.
Meh. Unless they need Hollywood-calibre animation or surfacing, their current tools probably, make that surely exceed LightWave's 'accuracy'.

inkpen3d
10-15-2012, 05:10 AM
I can think of one group that needs accurate models... Researchers doing papers.

Precisely, my clients increasingly want accurate representations of molecules and their interactions, so there definitely is a demand for this type of thing. See some examples here of the kind of thing they're after: http://www.molecularmovies.com/showcase/

jeric_synergy
10-15-2012, 02:56 PM
... so there definitely is a demand for this type of thing.
"A demand" isn't the same as "enough demand to warrant diverting resources to it when there's already an existing pipeline."

A better Illustrator import facility would have a thousand times as many users.

inkpen3d
10-15-2012, 03:57 PM
"A demand" isn't the same as "enough demand to warrant diverting resources to it when there's already an existing pipeline."

A better Illustrator import facility would have a thousand times as many users.

There are several existing pipelines, all of which are cumbersome, each with their own limitations, and all are very inefficient.

I fail to see how one, or even two, Python experts interested enough to take this small project forward would really be a major diversion of "resources" - what exactly would it be diverting them away from? It seems you are simply setting up a straw man for the sake of continuing to argue your position. If you are completely uninterested in an embedded molecular modeller, there's really little point in you continuing to post on this thread as you have made your argument quite eloquently. :)

It remains to be seen if enough people are interested (i.e. to create the demand) to warrant taking this further, which is after all the whole point of this poll. :D It may in the end turn out that there's not enough demand, in which case we're stuck with the existing cobbled together pipelines, or saving up enough money to purchase a copy of Maya! ;)

Regards,
Peter

inkpen3d
10-24-2012, 07:05 AM
Although the response to this poll has not been overwhelming, it was sufficient for the guys at the Scripps Research Institute to consider assisting with extending ePMV to run under LightWave.

Here's a quote from the email I recently received from Graham Johnson on this subject:

"22 raises an eyebrow, of course triple digits would be much more enticing. I guess you can shift the pole or ask the community another question... "are there any python programming experts who are willing to work under Ludovic's guidance to make this happen, or would the community prefer to wait (what could be up to a year or two, or possibly never, to see if Ludovic gets interested enough to do it himself)?"

Ludo is the developer of the uPy software that allows all of this to work (by all of this, I mean ePMV and 3 other uPy plugins he's enabled/developed).

He's incredibly busy for at least the next 3 weeks, but after that, he may have time to post some general API requirements on the uPy website and then your LightWave Python programmers can look through LightWave's Python API to make sure LightWavePython meets all of those uPy requirements, e.g., Certain mesh, material, instancing, etc. manipulation capabilities must exist for the system to work.

Ludo has found that that some Python APIs for 3D Animation packages are very superficial and don't provide enough access to the core functionalities of the program to enable ePMV to run efficiently or at all. If those capabilities don't yet exist, you guys would have to get together to badger NewTech to make them happen, or get together to build them yourselves (with a 3rd party Python wrapper). The Python API needs to be thorough, efficient (low-level), and robust for this to make sense.

If LightWave appears to be ready, then Ludo can provide the programmers with template files to enable them to transpose one of the existing wrappers to LightWave and he can help them troubleshoot if they run into problems. Theoretically, if Lightwave is well prepared for Python and you have one or two extremely talented programmers working, they can get uPy/ePMV/autoPack up and running in a matter of hours.... of course in practice its never that clean and usually takes days and sometimes weeks to debug... most often due to problems in the Application's Python API and/or bizarre dependencies or quirks within the software itself (i.e., differences between it and the other packages already wrapped)."

So, the question is, are any of you (or do you know of) an expert Python programmer with a good working knowledge of the LW SDK and willing to volunteer some free time to undertake this project?

Thanks in advance.

Regards,
Peter

digitaldoc
10-24-2012, 10:43 AM
This is an excellent and free pipeline for bringing Proteins and other chemical structures into Lightwave in several formats.
Written for LW 7.5 but works with LW 11.


From the Lightwave Plugin Database

Author: James Redman

Availability: Free

Added to DB: December 7 2009

Details:
A utility that enables molecular models generated with Accelrys Discovery Studio Visualizer to be recreated in LightWave Modeler. It supports protein and nucleic acid ribbons, as well as ball and stick molecules.

http://www.kombyonyx.com/4767/11646.html

AND

Accelrys Discovery Studio Visualizer (free version)

http://accelrys.com/products/discovery-studio/visualization-download.php

108746

inkpen3d
10-24-2012, 11:37 AM
This is an excellent and free pipeline for bringing Proteins and other chemical structures into Lightwave in several formats.
Written for LW 7.5 but works with LW 11.

From the Lightwave Plugin Database
Author: James Redman
Availability: Free
Added to DB: December 7 2009
Details:
A utility that enables molecular models generated with Accelrys Discovery Studio Visualizer to be recreated in LightWave Modeler. It supports protein and nucleic acid ribbons, as well as ball and stick molecules.
http://www.kombyonyx.com/4767/11646.html
AND
Accelrys Discovery Studio Visualizer (free version)
http://accelrys.com/products/discovery-studio/visualization-download.php


Thanks for posting - that's one of the pipelines that I currently use. :thumbsup:

However, it has several limitations:

For large protein complexes or certain styles of molecular representation within Accelrys the exported POV file is translated into a HUGE LScript file by dspovtols. When this LS is run within Modeler (to generate the geometry) brings even a reasonably high spec system such as mine to its knees!
The geometry produced for some molecular representations by the pipeline is very inefficient - e.g. the simple "line" representation of molecules in Accelyrs when exported to Modeler ends up as a set of tubes, each terminated by a sphere (to produced the rounded end of the line as seen in Accelyrs). And, as mentioned above, it also takes an age to process these LS files produced by dspovtols, or, for very large protein structures, completely cripples the system.
Accelyrs has its own limitations - e.g. although you can display "biological units" in Accelrys, you cannot export them (i.e. some PDB files contain information about how the protein sub-units are assembled into a whole biological structure, such as a virus protein coat). Currently Accelyrs can only export a single sub-unit at a time, and even if it could export a whole biological unit, the resulting huge POV/LS file would be unusable in Modeler!
The pipeline is obviously not a very interactive process - you have to keep re-running the pipeline every time you want to , for example, change the type of representation of a given molecule.


An embedded viewer in LW such as ePMV would circumvent most, if not all, of the above limitations.

Regards,
Peter

digitaldoc
10-24-2012, 11:56 AM
I voted YES as well!:thumbsup:

inkpen3d
10-24-2012, 12:48 PM
Many thanks Steve! :thumbsup:

Hieron
02-08-2013, 05:22 AM
"A demand" isn't the same as "enough demand to warrant diverting resources to it when there's already an existing pipeline."

A better Illustrator import facility would have a thousand times as many users.

I think you'd be surprised about the sci viz industry and just randomly making up proteins is often not an option. This tool is an important asset to any modern 3D package, period. Al other suggested routes for LW are outdated and alot less powerful.

Whether or not illustrator importing needs to be in, is another matter.

zardoz
02-08-2013, 05:36 AM
well I have a degree in biochemistry but life does some tricks on you and here I am doing 3d in a fx production house...but I would love to see a more advanced molecular importer in lightwave.

PS: boo for the eleven who voted no! ;P

daforum
02-08-2013, 06:11 AM
FYI, PDB_Reader_v2.85.ls still works in LW Modeler v11.0.3 - it's okay for importing PDB files as point-clouds for HVs, ball-and-stick, or CPK type models.

It doesn't seem to be available on the internet any more, so I've attached it here for anyone who doesn't already have it.

Regards,
Peter

108478

Cool, thanks for the zip Peter :thumbsup:

I did find it online again here too: http://www.chromecow.com/

daforum
02-12-2013, 03:22 AM
This is an excellent and free pipeline for bringing Proteins and other chemical structures into Lightwave in several formats.
Written for LW 7.5 but works with LW 11.


From the Lightwave Plugin Database

Author: James Redman

Availability: Free

Added to DB: December 7 2009

Details:
A utility that enables molecular models generated with Accelrys Discovery Studio Visualizer to be recreated in LightWave Modeler. It supports protein and nucleic acid ribbons, as well as ball and stick molecules.

http://www.kombyonyx.com/4767/11646.html

AND

Accelrys Discovery Studio Visualizer (free version)

http://accelrys.com/products/discovery-studio/visualization-download.php

108746

Does the LW Molecule Importer work on Mac too?

EDIT: Just had a look at the LW Plugin database and it says Windows only.

Nanobiz
02-12-2013, 06:32 AM
Well, i would be very interested in such a plug-in... heck, I would even pay for it...!

grahamj
02-19-2013, 10:13 PM
I think you'd be surprised about the sci viz industry and just randomly making up proteins is often not an option. This tool is an important asset to any modern 3D package, period. Al other suggested routes for LW are outdated and alot less powerful.
Nice to meet you:
Hi Everyone,
InkPad3D has worked hard to get our attention, and now that we are aware that the LW community has some interest in our plugins, we will try to support LW. Its exciting to see through these discussion that some very famous molecular animation pieces have used a LW pipeline! This effort, however, may take time and will likely require a LW programmer to tackle. I'm writing here to try to recruit increased interest (to motivate us) and support (to help us tackle this challenge).

ePMV and autoPACK are uPy apps that make molecular modeling and animation easy:
The embedded Python Molecular Viewer (ePMV) (http://epmv.scripps.edu) app discussed here was the first of a slowly growing number of scientific tools (including autoPACK (http://www.autopack.org/cellpackchallenge2012)) written in our ubiquitous Python API uPy (http://upy.scripps.edu) wrapper. If we, or someone from the LW community can wrap LW's Python API with uPy, then ePMV, autoPACK, and others will become instantly available along with any future apps that are written in pure uPy. For LW plugin programmers: if you write future LW plugins in pure uPy, you can expand your potential user-base because they will also function in Maya, 3dsMax, Cinema 4D, and Blender (you don't need to write 5 separate versions).

The problem:
Ludovic (the developer of uPy) browsed through LW's Python API last night and said that it looks complex, convoluted, and sparsely documented. He is very busy with other projects right now and has never used LW directly (perhaps he wasn't even looking at the latest version), so the point is that we will need your help if you want it to happen soon. The ideal situation would be for us to guide a LW Python programming expert to create the uPy wrappers. For a programming expert, this involves "simply" pasting the corresponding LW-Python functions into two of our template uPy files. With that task completed (and debugged), ePMV and autoPACK will become instantly useable in LW.

A caveat?:
uPy, ePMV, and autoPACk are open-source, and currently LGPL-openSource, so no iteration is sellable... i.e., could you guys find/encourage someone to take on such a project for free or perhaps with little more than a [Donate] button on their own website to encourage their efforts- knowing they would have to release the code open-source themselves?
¿Perhaps NewTek themselves would be interested in doing it (should take them just a couple days)... you could rally to argue that this might keep molecular and cell-biology animators from jumping ship as the industry increasingly demands the level of accuracy and flexible options that apps like ePMV (http://epmv.scripps.edu), mMaya (http://www.molecularmovies.com/toolkit/), and bioBlender (http://bioblender.eu) provide?

Stopgap:
In the mean time, there are solutions to import molecular and other scientific representations from other molecular viewers. You can even create animations directly inside of dedicated molecular viewers like PMV (http://www.youtube.com/watch?v=Hyuv1KvT3hE) and Chimera (https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnx1Y3Nmam9obnNvbnxneDo0ZTIy NThiYWU4ODY2NDYz).
To work in your favorite 3D host, however:
In addition to the little plugin I see being passed around on this thread, the traditional path we've used since the mid 1990's is to generate the representation you want in a molecular viewer and to export that as a mesh or collection of cylinder/spheres to a universal format for import into your software... most often VRML. Different molecular viewers have different styles of ribbons for example that certain clients demand over other styles, so if they need sweet beads on your alpha helices, you can export a vrml from PyMOL (http://www.pymol.org). If you want a coarse surface or an electron-micrograph surface from volumetric data, you're best off using Chimera (http://www.cgl.ucsf.edu/chimera/), etc.

For simple projects like a turntable animation of a single molecule for a pharmaceutical add, these techniques are still quite useful. For more complex animations or if precise chemical colors are needed, then these techniques are limiting. Because you are dealing with "dead" (de-coupled from the data that was used to make them) colored-per-vertex geometry, you can't get chemically colored meshes, make changes to the representation or do things like initiate chemically relevant simulations.

Do everything in your host (hopefully LightWave will become a host to join Cinema 4D, Maya, Blender, and 3dsMax):
Rather than building a molecular graphics engine from scratch in a host like LW, ePMV "simply" provides a GUI connection to a powerful molecular graphics engine that works behind the scenes to do all of the "importing, exporting, visualization, simulation, etc." for you. You can import tens of thousands of molecules from public databanks as instructed by your clients or do things like make "near"-accurate DNA models from scratch simply by typing a sequence of As, Ts, Gs, and Cs into one of ePMV's extension widgets. Power users can do things (http://epmv.scripps.edu/documentation/tutorials/written-tutorials/general-epmv-tutorials)like initiate Molecular Dynamics simulation directly in the host (hopefully LW) viewport, or have scientifically valid physics engines interoperate with the host's physics, etc.

A Niche? Yes, but growing:
Indeed, scientific visualization, molecular modeling and animation, and cell biology subject jobs still make a relatively small portion of the overall 3D job market. However, as others have pointed out here, it is a growing market need both in scientific communication and pop-culture. At the same time, our general population is slowly starting to recognize and become visually educated (very slowly i realize) about what DNA should look like or the shape of an antibody, and they are joining the scientific market and slowly demanding more accurate portrayals of these once purely abstract shapes. I strongly encourage artistic license and abstraction to convey your message and there will still plenty of jobs for years to come that don't require (nor should they) accuracy, but with a tool like ePMV, you have the option of easily creating an accurate representation in seconds that can be used directly or as a reference for any direction a 3D artists wants to push.

TODO:
If he can find time, Ludovic will try to make our most basic test-script work (this would popup a little GUI in LW with a button to build a sphere into the scene and a slider to adjust the radius in realtime). He will test for a couple other basic needs, and then we will likely have to wait for a few months to finish some other projects or until we see a much increased interest in the voting poll or better yet, a hardcore, gifted LW Python expert who can tackle the challenge.

To inspire more interest–autoPACK Visualization Challenge:
You would have to generate the model in a different software package and then import it into LightWave (or team up (http://forums.cgsociety.org/showthread.php?t=1087203) with someone who could provide an export of one of the models), but we've presented an HIV Visualization Challenge (http://autopack.cgsociety.org) on CGSociety and it would be great to see some LW users participate. You can win lots of money and hardware (http://autopack.cgsociety.org/autopack/prizes) ($2,500 is just part of the video category 1st place prize)! Autodesk has graciously sponsored this contest, but there are no limits on the software you use, as long as you use autoPACK at one point in the pipeline and roughly describe your protocol. Less than one month left!

lwanmtr
02-19-2013, 10:32 PM
The more tools we have to use, the better. Even though it's not on the top of my list, something like this would just add to what I'm able to offer.

inkpen3d
02-20-2013, 07:05 AM
@GrahamJ

Many thanks for your invaluable contribution to this thread.

I've sent a PM to Rob Powers bringing this to his attention. Hopefully Rob will be able to assist, or at least advise, on a course of action that will enable us to bring ePMV and the many other very powerful uPy-based tools that are available, which will greatly enhance LightWave's scientific visualisation capabilities such that it can compete with the likes of Cinema4D, 3dsMax and Maya.

Cheers,
Peter

devin
02-20-2013, 08:25 AM
I work in the legal industry and we find ourselves in need of accurate molecular models all along. We do patent litigation for pharma quite often and that seems to be picking up more as of late. Another "yes" vote.

dnch
02-20-2013, 08:50 AM
i was recently asked for molecular animations for educational purposes, this would help greatly

johnliebler
02-20-2013, 09:13 AM
Thanks for chiming in Graham! And thanks to Inkpen for continuing to push this forward.

It would be very exciting to see ePMV and autoPACK in Lightwave. I'm really just an artist, so I'm no help on the programming side, but I'll see if I can drum up any interest in my (limited) circle of influence. I have a project coming up soon that could really benefit from some accurate and easy to use tools.

Cheers!

John

flyboy
02-20-2013, 10:58 AM
Can LW produce STL files to work with 3D printer plastic modeler machines ? A friend of mine has recently bought one and intends to use Autocad but asked me if LW has a file extension that will work with the Airwolf 3D printer.

digitaldoc
02-20-2013, 11:09 AM
@GrahamJ

Thanks for the input and interest.
As a retired physician and freelance medical/scientific animator your pipeline is intriguing and would be most welcome here.

Thanks to Inkpen for pushing this forward and sticking with it!

jeric_synergy
02-20-2013, 11:47 AM
I think you'd be surprised about the sci viz industry and just randomly making up proteins is often not an option. This tool is an important asset to any modern 3D package, period. Al other suggested routes for LW are outdated and alot less powerful.
I'm unconvinced, and a forum poll is not going to convince me either. This is a perfect example of something that is far better served by a third party developer rather than having the already extremely busy LightWave devs dragged off into some niche culdesac.

You know what WOULD convince me? MONEY. Start a kick-starter-like escrow auction, start accumulating cold hard cash, and the first independent dev to produce a specified product gets the guaranteed funds PLUS rights to profit from subsequent sales.

It's pretty damn easy to say "Hell yes I'd like a free cool tool!", but your real needs are revealed when you open up your wallet.


You want a likely dev who already has an insight into a highly technical market? Ask that guy who created Volu-medic: he's a genius AND he knows the market.

Then show him the $$$ you already have and say "This is yours when the plugin works to our satisfaction."

dnch
02-22-2013, 08:38 AM
I'm unconvinced, and a forum poll is ....


oh i thought we were talking about external devs, not newtek devs, they should focus on more universal lw tools ofcourse

devin
02-22-2013, 09:14 AM
I'm unconvinced, and a forum poll is not going to convince me either. This is a perfect example of something that is far better served by a third party developer rather than having the already extremely busy LightWave devs dragged off into some niche culdesac.

You know what WOULD convince me? MONEY. Start a kick-starter-like escrow auction, start accumulating cold hard cash, and the first independent dev to produce a specified product gets the guaranteed funds PLUS rights to profit from subsequent sales.

The way that I read this thread is that the sentiment of the majority is to get outside developers involved. I don't think that anyone is above paying for a solution either, provided that it is a solid one that will not get abandoned.

aurora
02-22-2013, 09:17 AM
I think and constantly poke the fire for LW to support a slew of different scientific visualization needs. I have worked for a great many years with molecular biology and biochemistry, they were my first two degrees, and while I now work in most any all all facets of high performance scientific computation and visualization, molecular dynamics is still a top favorite subject of mine. I would be more then willing and desirous to help develop, test, support, or what ever may be needed for a tool such as this in LW. But I will preface this with I do have my own suite of molecular viz tools for LW that tie in with VMD and will continue to work with any all VMD tie ins I can but this is a computational need and a render need. I do all my final renders in LW.


What I kind of question though is, 'molecular dynamics or docking energy calculations on the fly'. Define this fly, as in my work such a fly is equivalent to combining all the flies in Africa into a uber-mega Godzilla fly requiring hundreds to several thousand cpu and/or gpu hours.

Hieron
02-22-2013, 10:07 AM
Nice to meet you:....

Thanks for your insightful input. Hopefully someone capable with experience in the LW python SDK can be found and persuaded to part with his spare time.
These are great tools and LW would be much improved by supporting them.

Nice HIV Visualization challenge!

jeric_synergy
02-22-2013, 12:11 PM
Ahhh, if the consensus is "outside devs", heck yeah.

I'll continue to flack for my "start with SOME money in hand" program: I think it's a bit much to ask indie devs to just start writing in the vague hope of some remuneration.

Surely with all the indiegogos and kickstarters out there there's some escrow service for such a thing. ??? :(

johnliebler
04-06-2013, 01:17 PM
Bump!
Thought I'd poke this crocodile, see if there's any new interest?
:)

inkpen3d
04-07-2013, 05:34 AM
Bump!
Thought I'd poke this crocodile, see if there's any new interest?
:)

Alas, I've been working flat-out on 3 different projects for various clients that, ironically, have all involved molecular modelling! Consequently, I've not been able to devote much time to this recently.

The last thing I did was to send that PM to Rob Powers a few weeks ago asking him if there was a chance that the dev team could possibly help us out in any way and in so doing make a useful tool accessible to LW - something that is already available to several of its competing 3D applications - for very little effort/cost (i.e. one of their python experts could achieve in a couple of days what would possibly take us novices weeks). Rob responded by saying that this topic would be discussed at their next dev team meeting. However, so far I've not had any feedback from Rob regarding the outcome of that meeting - I suspect little, if anything, will be forthcoming. Even if the odds are slim of getting the dev team's help, it was worth a try - you don't get anything if you don't ask!

Cheers,
Peter

jeric_synergy
04-07-2013, 08:30 AM
I'm saying again: if you guys REALLY want thi$, you'd put some money in escrow and give it to the first plugin/script that met your specs.

inkpen3d
04-07-2013, 10:38 AM
I'm saying again: if you guys REALLY want thi$, you'd put some money in escrow and give it to the first plugin/script that met your specs.

If all else fails, we may have to do as you have been suggesting. Either way, we still need to locate someone with the required skill-set, and that could be present us with a problem.

The following observation that Graham Johnson made in his post is cause for concern...



Ludovic (the developer of uPy) browsed through LW's Python API last night and said that it looks complex, convoluted, and sparsely documented.


Which doesn't exactly inspire any confidence that the current shambolic state of LW's Python API makes it up to the job!

An additional concern of mine is that the inherent schizophrenic nature of LW (i.e. that it is still largely split into two applications, Modeler and Layout, glued together by the flaky Hub) might turn out to be a show-stopper if it turns out that ePMV requires an integrated environment to interface to - something that all the other 3D applications ePMV works with on have in common!


All the best,
Peter

jeric_synergy
04-07-2013, 03:48 PM
Perhaps there's a vendor who already deals in scientific and medical LW plugins that would be interested.

Now, who would that be? ;)

chikega
05-26-2013, 08:05 AM
I recently saw this CG Challenge using autoPack, cellPack and ePMV.

http://www.cgsociety.org/index.php/CGSFeatures/CGSFeatureSpecial/autopack_challenge_winners

It seems that Softimage, Maya, Cinema4d and Blender are supported. Scientific/Pharmaceutical Visualization is a multi-billion dollar industry. In fact, many of my medical illustrator colleagues are doing extremely well in this field. And given the current state of VFX in the U.S., this is an opportunity for Lightwave to branch into this lucrative field.

http://www.autopack.org/home

aurora
05-26-2013, 05:30 PM
Autopack was cute put of little help for me. For my work I do my runs in NAMD, Amber, Gromacs, VASP(ugh hate it) do initial recon in VMD and then send it out to LW for rendering. For membrane lipids I often cheat and just use Instancer, as long as I'm not doing dynamics where the membrane is part of the model. The only thing I ever find wanting in LW is the need for high vertex/poly counts.

chikega
05-26-2013, 07:54 PM
Sounds deep, Tim :)