PDA

View Full Version : Lightwave Bullet Dyamics "Convex Pieces" is Broken in layout 2015



jboudreau
02-07-2016, 09:13 AM
Hi Guys

Looks like Dynamics is broken in 2015.3 especially when it comes to convex pieces. Collision is not working well with convex pieces in lightwave 2015, 2015.1, 2015.2, 2015.3 the pieces keep going through the floor no matter what I change also the visible representation of convex pieces is completely different from 11.6.3 to 2015.3.

I opened the same scene in layout 11.6.3 and it works perfectly, there are no pieces that go through the floor. It works as it should and the visible display is correct as well. In 2015.3 the display looks like a bunch of dots what's changed and why am I getting these dots instead of how it use to look in 11.6.3?

Here are some screen grabs showing you the difference

132252

132253


I'm going to keep on checking what else might be broken in 2015.3 regarding dynamics, Ill keep you all posted.

Could anybody also please explain what's changed and how to fix these issues if possible?

Thanks,
Jason

prometheus
02-07-2016, 09:18 AM
Hi Guys

Looks like Dynamics is broken in 2015.3 especially when it comes to convex pieces. Collision is not working well with convex pieces in lightwave 2015, 2015.1, 2015.2, 2015.3 the pieces keep going through the floor no matter what I change also the visible representation of convex pieces is completely different from 11.6.3 to 2015.3.

I opened the same scene in layout 11.6.3 and it works perfectly, there are no pieces that go through the floor. It works as it should and the visible display is correct as well. In 2015.3 the display looks like a bunch of dots what's changed and why am I getting these dots instead of how it use to look in 11.6.3?

Here are some screen grabs showing you the difference

132252

132253


I'm going to keep on checking what else might be broken in 2015.3 regarding dynamics, Ill keep you all posted.

Could anybody also please explain what's changed and how to fix these issues if possible?

Thanks,
Jason


yes..I recall having some issues with that too...need to verify it and check again though.

jboudreau
02-07-2016, 10:02 AM
Any luck Prometheus?

prometheus
02-08-2016, 07:02 AM
Any luck Prometheus?

To busy right now..

prometheus
02-08-2016, 07:19 AM
found a couple of minutes to check it...and ergo, somethings not right.

what I did..

added a cube in and a ground plane in layout directly with modeler tools, in 11.6.2
fractured the cube in 20 pieces, made it parts body, convex pieces.. and ground plane static ..thatīs it, pieces hits the flor and do not fall through the groundplane.

loaded that scene in lw 2015.3 32 bit, hit play and all is just the same..it works.
now...I disabled the cube , created a new cube with modeler tools directly in layout 2015, fractured it the same as I did with 11.6.2 and made it parts body...then hit play, and when hitting the ground, some pieces are falling through the ground..all the settings is the same as the first cube..so I suspect something may have changed with the creation of the cube item, or the fracture tool?

I can see that the polyamount isnīt the same for the cubes, so my thought is that something may have changed on fracture, and not actually convex pieces?

jboudreau
02-08-2016, 07:23 AM
found a couple of minutes to check it...and ergo, somethings not right.

what I did..

added a cube in and a ground plane in layout directly with modeler tools, in 11.6.2
fractured the cube in 20 pieces, made it parts body, convex pieces.. and ground plane static ..thatīs it, pieces hits the flor and do not fall through the groundplane.

loaded that scene in lw 2015.3 32 bit, hit play and all is just the same..it works.
now...I disabled the cube , created a new cube with modeler tools directly in layout 2015, fractured it the same as I did with 11.6.2 and made it parts body...then hit play, and when hitting the ground, some pieces are falling through the ground..all the settings is the same as the first cube..so I suspect something may have changed with the creation of the cube item, or the fracture tool?

Hi thanks for testing this

When you loaded the same scene into 2015.3 did you recalculate the dynamics, If not it's using the Dynacache file that was created from 11.6.3 so it would be the same no pieces going through the floor. In 2015.3 select your parts object and hit the reset button. Now recalculate the dynamics the pieces should go through the floor.

Thanks,
Jason

prometheus
02-08-2016, 07:39 AM
Hi thanks for testing this

When you loaded the same scene into 2015.3 did you recalculate the dynamics, If not it's using the Dynacache file that was created from 11.6.3 so it would be the same no pieces going through the floor. In 2015.3 select your parts object and hit the reset button. Now recalculate the dynamics the pieces should go through the floor.

Thanks,
Jason

forgot to reset...just tested by reseting it, but only the cube fractured in 2015 falls through the ground, so the cube that was fractured in 11.6.2 works as it should.
I also tested to run the free crackit plugin, which has a different fracture algotrithm, which I think also is nicer...it doesnīt fall through the ground either, so this leads me to think it is the fracture tool in 2015 that has some issues, and not the actual convex pieces mode in bullet.

prometheus
02-08-2016, 07:42 AM
Edit...I was wrong...just moved the 11.6.2 cube..and it now also falls through the ground.

jboudreau
02-08-2016, 07:43 AM
forgot to reset...just tested by reseting it, but only the cube fractured in 2015 falls through the ground, so the cube that was fractured in 11.6.2 works as it should.
I also tested to run the free crackit plugin, which has a different fracture algotrithm, which I think also is nicer...it doesnīt fall through the ground either, so this leads me to think it is the fracture tool in 2015 that has some issues, and not the actual convex pieces mode in bullet.

Hi, You are probably somewhat right but if you model a box and array on xyz 20,20,20 for example and set it to parts, The boxes go through the floor aswell, The only shape you can set so it doesn't go through the floor is Sphere and Mesh. All other go through the floor in 2015.3 but not in 11.6.3

Thanks,
Jason

prometheus
02-08-2016, 07:45 AM
Hi, You are probably somewhat right but if you model a box and array on xyz 20,20,20 for example and set it to parts, The boxes go through the floor aswell, The only shape you can set so it doesn't go through the floor is Sphere and Mesh. All other go through the floor in 2015.3 but not in 11.6.3

Thanks,
Jason


You could change the mass distribution to surface or vertices, I think that would help avoid the pieces falling through, but this seems like a bug that shouldnīt be there.

prometheus
02-08-2016, 07:50 AM
You could change the mass distribution to surface or vertices, I think that would help avoid the pieces falling through, but this seems like a bug that shouldnīt be there.

Tested with a cube fractured again with crackit fracture and not the native ones, also falls through...
Tested surface mass distribution...still some pieces falling through...
Tested vertices mass distribution...a lot better, donīt see anything falling through, yet..who knows later on.

jboudreau
02-08-2016, 07:52 AM
You could change the mass distribution to surface or vertices, I think that would help avoid the pieces falling through, but this seems like a bug that shouldnīt be there.

Unfortunately I already tried that and it didn't work. If you set the mass distribution to vertices then the pieces will not roll or rotate like they should.

Thanks,
Jason

prometheus
02-08-2016, 07:58 AM
Unfortunately I already tried that and it didn't work. If you set the mass distribution to vertices then the pieces will not roll or rotate like they should.

Thanks,
Jason

yes..but it works, how natural and realistic is another thing :) but of course...it shouldnīt be like this, one would like it to behave as before...with proper roll and rotation as you said.

think we need to report this, I really would like a fix on this, how would we else be able to efficiently work with the new bullet dynamics bones etc and work with convex pieces for faster simulations..it means one would have to step back to 11.6 and then you can not use the bullet bone stuff, ragdolls etc.

jboudreau
02-08-2016, 08:10 AM
yes..but it works, how natural and realistic is another thing :) but of course...it shouldnīt be like this, one would like it to behave as before...with proper roll and rotation as you said.

think we need to report this, I really would like a fix on this, how would we else be able to efficiently work with the new bullet dynamics bones etc and work with convex pieces for faster simulations..it means one would have to step back to 11.6 and then you can not use the bullet bone stuff, ragdolls etc.

Unfortunately it's not just this that is broken in 2015.3, I have reported the going through the floor issue

This is what I found that is broken in 2015.3 but not in 11.6.3

- Parts go through floor no matter what settings you change for collision margin etc the only collision shape that works so the parts don't go through the floor is "sphere" and "mesh". (11.6.3 does not have any problems works great with all collision shapes)

- Dynamic simulations are 7x slower calculating in lightwave 2015.3 (11.6.3 is 7x faster and does not have any problems works great)

- Convex Pieces is not rendered or displayed the same as it was in 11.6.3. (2015.3 shows a bunch of dots - 11.6.3 shows a representation of the actual objects which is what it should do)

- dynamics recalculate after removing from the dynamics panel or deleting objects in your scene that are in the dynamics panel but are not activated (This is broken in both 11.6.3 and 2015.3)

There are also more issues that DrStrik9 found aswell you can view the post here http://forums.newtek.com/showthread.php?149687-More-dynamics-blues-%85/page2

I'll update the list as I go

Thanks,
Jason

prometheus
02-08-2016, 08:32 AM
yeah...2015 did unfortunatly not start out well for me either, a lot of crashes with bullet in general..that was thoug fixed quite fast, then crashes with texture falloff, and I am now witnessing a huge amount of crashes on other areas that I just had to put on a seperate thread.

So..unfortunatly the 2015 experience..hasnīt been that smooth, one would of course like to utilize the new stuff in there, but at the same time a bit frustrating when it disrupts your workflow in som many areas.

I mainly bought in to it..not for the sake of the new features as a major part, but more of a good price deal, and to set me up for further lightwave versions which I got attracted to with the new blog stuff, other than that I might
leave 2015 alone for a while and go back to 11.6.2 or 11.6.3 which has been quite stable compared to 2015.

Modeler 2015 has been a bit more stable, and some new stuff with the texture options that I like, then again it seem to have broke how it chamfers edges and keeps or drops the selection after chamfer, which is completly different..I need it to behave as 11.6 did..so that is a minor issue too.

Surrealist.
02-08-2016, 10:48 AM
But you say fracture does work if it is set to mesh though right? I don't have it to test right now as I am rendering. Just wanted to clarify that point.

And hope you get some movement on these bugs. :)

DrStrik9
02-08-2016, 11:59 AM
Mesh makes calc times in 2015.3 MANY times slower.

This brings up the problem with moving to an annual product update paradigm without the proper resources to make it happen seamlessly (which I am sure is good financially in the short term). However, if you make your loyal customers into paid beta-testers, you will eventually lose loyal customers. No one wants that to happen. It would be tragic to see fewer and fewer people buying into the next rev, because the last rev still doesn't work properly.

Surrealist.
02-09-2016, 04:32 AM
Are the calc times on Mesh drastically slower in 2015?

jboudreau
02-09-2016, 04:50 AM
Are the calc times on Mesh drastically slower in 2015?

Yes there is a huge difference 7x slower in 2015. DrStrik9's simulation was taking 14 hours to complete until I told him to try it in 11.6.3 which only took about 2 hours for the same simulation. On top of that convex pieces which is even faster calculating is broken in 2015 pieces go through the floor.

Thanks
Jason

Surrealist.
02-09-2016, 05:19 AM
OK. Thanks for the heads up. I'll do some testing on my end and see what I come up with. Unfortunately I don't have a 11.* version.

jboudreau
02-09-2016, 05:21 AM
No problem. You can download it from your account.

Thanks
Jason

lino.grandi
02-09-2016, 05:27 AM
You may need to increase the Dynamics Frame Rate.

jboudreau
02-09-2016, 05:51 AM
You may need to increase the Dynamics Frame Rate.

Hi Lino

Doesn't that just make the dynamics play back faster, I gave it a try and it's still 7x slower than in 11.6.3

Please see this post http://forums.newtek.com/showthread.php?149687-More-dynamics-blues-%85 #15 It has a scene that you can test both in 11.6.3 and 2015.3 you will see what we mean

Also the following things seem to be broken in 2015.3

- Parts go through floor no matter what settings you change for collision margin etc the only collision shape that works so the parts don't go through the floor is "sphere" and "mesh". (11.6.3 does not have any problems works great with all collision shapes)

- Dynamic simulations are 7x slower calculating in lightwave 2015.3 (11.6.3 is 7x faster and does not have any problems works great)

- Convex Pieces is not rendered or displayed the same as it was in 11.6.3. (2015.3 shows a bunch of dots - 11.6.3 shows a representation of the actual objects which is what it should do)

- dynamics recalculate after removing from the dynamics panel or deleting objects in your scene that are in the dynamics panel but are not activated (This is broken in both 11.6.3 and 2015.3)

Hopefully you can verify or let us know what the problem might be

Thanks,
Jason

jboudreau
02-09-2016, 06:36 AM
You may need to increase the Dynamics Frame Rate.

This helps some with the piceces going through the floor but man do you ever take a hit on calculation time. It's extremely slower calculating increasing the dynamic frame rate. Also 11.6.3 does not need it's dynamic frame rate increased at all and the simulation is almost real time. I'm sorry to say (I'm not trying to be rude here) but this is not a solution, changing settings only to make it 10x slower when previous versions of lightwave worked perfectly without having to change anything and calculated the simulation 7 - 10x faster. Can Someone anyone Lino etc please verify this and let us know what the problem is here. There is a huge issue with bullet dynamics in 2015.3

Thanks,
Jason

lino.grandi
02-09-2016, 07:14 AM
Hi Lino

Doesn't that just make the dynamics play back faster, I gave it a try and it's still 7x slower than in 11.6.3

No, Dynamics Framerate controls the precision of the simulation. I've set it to 360 and change the Static object shape to Box...the scene works fine now.

jboudreau
02-09-2016, 07:22 AM
No, Dynamics Framerate controls the precision of the simulation. I've set it to 360 and change the Static object shape to Box...the scene works fine now.

Yes but why do we have to do that, In 11.6.3 you can leave it at it's default settings 180 and nothing goes through the floor and it's 7 - 10x faster, I'm sorry but your solution is extremely slow calculating compared to 11.6.3 and this is on a very simple simulation. Try this on a complex simulation and instead of 2 hours to run in 11.6.3 it will take 14 hours. That's a lot of time wasted.

So please answer these very important questions

1. Why is it so much slower than 11.6.3
2. Why do we have to adjust the default dynamic framerate
3. Why does convex shapes show up as little dots instead of how it use to show up in 11.6.3 which gave a great representation of what the pieces were.
4. Why even though I change dyanmic frame rate to 360 (whhich now takes way longer to calculate) do pieces still go through the floor

I'm sorry but I get a bit frustrated when I get a response back that the scene is fine now when honestly you and me both know it's not. Seriously test the same scene on 11.6.3 and then tell me it's fine when it takes 7x the time to calculate especially when it's set to mesh, convex pieces is slower too in 2015.3.

In 11.6.3 you don't have to start messing around with dynamic frame rates and pieces don't go through the floor, I made my floor as thick as possible and still pieces were going through HOW is this possible? again no issues in 11.6.3

I think I understand now why these small bugs are getting missed during the testing phases. As long as it works by changing all kinds of things that shouldn't need to be changed, that's all that matters even though it worked way more efficient, faster and better in previous versions. I'm sorry but this really frustrates the hell out of me. I thought we were making software for the better.

Thanks,
Jason

lino.grandi
02-09-2016, 07:30 AM
About the other issues you mentioned, it would be great if you could report them officially.

jboudreau
02-09-2016, 07:35 AM
About the other issues you mentioned, it would be great if you could report them officially.

I reported them, but never heard back anything

Thanks,
Jason

lino.grandi
02-09-2016, 08:08 AM
I reported them, but never heard back anything

Thanks,
Jason

We'll look for the reports.

souzou
02-09-2016, 08:20 AM
No, Dynamics Framerate controls the precision of the simulation. I've set it to 360 and change the Static object shape to Box...the scene works fine now.

I still get a few pieces going through the floor with these settings (in 2015.3).

Jason - you can turn off the little dots by switching 'Draw Bodies' to none in the world tab.

I get same results as you in 11.6.3 and 2015.3 (though my times are slower - my PC is getting old)

jboudreau
02-09-2016, 08:24 AM
I still get a few pieces going through the floor with these settings (in 2015.3).

Jason - you can turn off the little dots by switching 'Draw Bodies' to none in the world tab.

I get same results as you in 11.6.3 and 2015.3 (though my times are slower - my PC is getting old)

- I get pieces going through the floor too with those settings.

- Oh yeah I know that, I'm wondering why it was changed to show dots instead of how it was in 11.6.3 which was in my opinion a much better representation of what the parts should look like

Thanks for your feedback man, Hope all is well
Jason

Ma3rk
02-09-2016, 03:08 PM
FWIW, I tested your scene in 11.62 as well as 2015.3 and getting the same results. Wow.

2015.3 has been out for HOW many months now? And this is just NOW coming to light?

I'm just gonna bite my tongue some more for now & hope there's a bug fix upgrade coming soon.

jboudreau
02-09-2016, 03:21 PM
FWIW, I tested your scene in 11.62 as well as 2015.3 and getting the same results. Wow.

2015.3 has been out for HOW many months now? And this is just NOW coming to light?

I'm just gonna bite my tongue some more for now & hope there's a bug fix upgrade coming soon.

Thanks man for doing the test.

Yes and the craziest thing is these bugs were in 2015.0 so that means they got through the testing process in 2015, 2015.1, 2015.2 and are still in 2015.3.

haha Yeah I should bite my tongue too. But I bet this won't be fixed until Lightwave next, I hope I'm wrong

DrStrik9
02-10-2016, 11:17 PM
Jason,

Since fracture is broken here, I don't typically use convex shapes. So I can't test them without a huge amount of labor to correct fracture's many errors. Have you tried upping the dynamics frame rate to say, 600, just to see if it helps in 2015.3?

Also, since I swapped out png's for psd's in my former bullet nightmare, I am finding that 2015.3 calculates much faster. Of course, upping the frame rate to 600 will take longer, but I think it "may" solve the falling-through-the-floor problem ... ? -- Plus, then you can up the frame rate of the scene, and render frames to play at a normal frame rate, giving a super-cool slo-mo effect. :+)

jwiede
03-14-2016, 01:09 PM
jboudreau, did you ever hear anything back from LW3DG regarding the other dynamics bugs mentioned?

jeric_synergy
03-14-2016, 05:12 PM
jboudreau, did you ever hear anything back from LW3DG regarding the other dynamics bugs mentioned?
I'll jump in here and ask what status you (jboudreau) see when you look at the SHOW ALL REPORTS page on your account?
132900

jboudreau
04-08-2016, 08:40 PM
Hi Guys

Sorry I got really busy and didn't get a change to update you guys

Here are the bugs I reported and the responses I got from them

(LWB-2039) dynamics recalculate Currently Closed

dynamics recalculate after removing objects from the dynamics panel or deleting
objects in your scene that are in the dynamics panel but are not activated
(This is broken in both 11.6.3 and 2015.3)

Response:

it does yes - but bear in mind we're at the mercy of the bullet library - we only implement it, we don't design it

Do you know of another software that implemented the bullet library that this doesn't happen?

Maybe you guys can help me here because I don't use any other software that has bullet dynamics so I can let them know if it's an issue with other software

(LWB-2038) Convex Pieces is not rendered or displayed the same as it was in 11.6.3. Currently closed

Response:

This is addressed internally for the next version of LightWave. After it's release, we can revisit possible patches for 2015.x

(LWB-2037) Dynamic Simulations are extremely slow to calculate in 2015.3 compared to 11.6.3 on the same scene Currently Closed

Response:

We have implemented the updated Bullet library for the next version of LightWave, and this does help our tests internally. More of it is now multithreaded, and this includes some of the collision detection.

(LWB-2036) LW Bullet Dynamics Dynacache size limitation (Casuse scenes to re-calculate)

Response:

Yes, it's a limitation based on the cache files being 32bit.

This issue has been solved internally for the next revision of LightWave.

(LB-1197) Lightwave Bullet Dyamics "Convex Pieces" is Broken in layout 2015 Currently Closed

Response

Fixed and tweaked things to be more stable and less prone to items being
pushed through others.

So it looks like they addressed all the issues, I guess we will have to wait until LightWave next is released to validate this.

Thanks,
Jason

CaptainMarlowe
04-09-2016, 12:40 AM
Which are great news if it is indeed fixed.

jwiede
04-12-2016, 01:47 PM
Which are great news if it is indeed fixed.

If they get released as a fix update for 2015, that is.

It's been eight months since the last update was released for 2015, and while working on the next version is important, getting fixes for the current version out to customers in a timely manner is as, if not more, important. There's nothing the least bit "timely" about customers waiting eight months or more for fixes, esp. when it's just to find out many months later that obtaining the long-awaited fix will require further payment as well. Such situations put customers over a barrel, and that's something very few customers want to experience, esp. when their efficiency is already compromised by LW defects.

If LW3DG cannot produce update releases more often than what we've seen to date, there is something seriously dysfunctional in LW product management and/or LW release methodology. Customers need fixes to LW bugs released in a much more timely manner than is occurring.

Snosrap
04-12-2016, 02:07 PM
Hi Guys

Sorry I got really busy and didn't get a change to update you guys

Here are the bugs I reported and the responses I got from them

(LWB-2039) dynamics recalculate Currently Closed

dynamics recalculate after removing objects from the dynamics panel or deleting
objects in your scene that are in the dynamics panel but are not activated
(This is broken in both 11.6.3 and 2015.3)

Response:

it does yes - but bear in mind we're at the mercy of the bullet library - we only implement it, we don't design it

Do you know of another software that implemented the bullet library that this doesn't happen?

Maybe you guys can help me here because I don't use any other software that has bullet dynamics so I can let them know if it's an issue with other software

(LWB-2038) Convex Pieces is not rendered or displayed the same as it was in 11.6.3. Currently closed

Response:

This is addressed internally for the next version of LightWave. After it's release, we can revisit possible patches for 2015.x

(LWB-2037) Dynamic Simulations are extremely slow to calculate in 2015.3 compared to 11.6.3 on the same scene Currently Closed

Response:

We have implemented the updated Bullet library for the next version of LightWave, and this does help our tests internally. More of it is now multithreaded, and this includes some of the collision detection.

(LWB-2036) LW Bullet Dynamics Dynacache size limitation (Casuse scenes to re-calculate)

Response:

Yes, it's a limitation based on the cache files being 32bit.

This issue has been solved internally for the next revision of LightWave.

(LB-1197) Lightwave Bullet Dyamics "Convex Pieces" is Broken in layout 2015 Currently Closed

Response

Fixed and tweaked things to be more stable and less prone to items being
pushed through others.

So it looks like they addressed all the issues, I guess we will have to wait until LightWave next is released to validate this.

Thanks,
Jason

Thanks - that was almost as good as some of the LW Next blogs! :)

Snosrap
04-12-2016, 02:12 PM
If they get released as a fix update for 2015, that is.

It's been eight months since the last update was released for 2015, and while working on the next version is important, getting fixes for the current version out to customers in a timely manner is as, if not more, important. There's nothing the least bit "timely" about customers waiting eight months or more for fixes, esp. when it's just to find out many months later that obtaining the long-awaited fix will require further payment as well. Such situations put customers over a barrel, and that's something very few customers want to experience, esp. when their efficiency is already compromised by LW defects.

If LW3DG cannot produce update releases more often than what we've seen to date, there is something seriously dysfunctional in LW product management and/or LW release methodology. Customers need fixes to LW bugs released in a much more timely manner than is occurring.

Yes - but maybe it's best that they work through the issues with the new Bullet libraries in LW Next and get it fully beta tested and out the door and then drop the new Bullet stuff in a 2015.4 update. That makes more sense to me. Yes I understand your point, but with such as small team they have to use their resources wisely and get the best return out of their efforts.

hrgiger
04-12-2016, 03:30 PM
If LW3DG cannot produce update releases more often than what we've seen to date, there is something seriously dysfunctional in LW product management and/or LW release methodology. Customers need fixes to LW bugs released in a much more timely manner than is occurring.

Sometimes things get pushed to the next version. I can't imagine its realistic to expect everything to be fixed, don't know any other software company that addresses all issues for each release so I don't think this is exclusive to the LW3DG. We certainly see this happening with Modo currently.

Personally I'm just glad to see these kinds of things being addressed in the next version, sounds like very positive improvements.

Snosrap
04-12-2016, 06:00 PM
Personally I'm just glad to see these kinds of things being addressed in the next version, sounds like very positive improvements.

Agreed!

jwiede
04-12-2016, 07:27 PM
I can't imagine its realistic to expect everything to be fixed, don't know any other software company that addresses all issues for each release so I don't think this is exclusive to the LW3DG.

Nobody argued "all issues should be fixed within their version cycle" in the first place. What relevance do such statements have to a discussion about LW3DG sitting on fixes for many months after announced resolution? Deferring release of bug fixes to work on "new version efforts" is explicitly prioritizing new efforts ahead of bug fixes.

hrgiger
04-12-2016, 07:43 PM
sorry if I missed it, I'm late to the discussion but they announced resolution to these particular issues?

But beyond that anyway.. I guess my point in saying referring to all issues is that its too simple of a response to simply say that LW3DG is prioritizing new efforts over bug fixes. Perhaps it is the case that they're implementing new efforts for the very reason that the old methods were doing a poor job of addressing said issues, maybe that's why they suggested they would revisit 2015.x after the new release was done. But in any event, I even see Shane over on the Modo boards, in response to criticisms from users about not tackling issues that were supposed to be addressed for 901, talking about releases sometimes causing more issues then they have the resources to manage yet I haven't seen you criticize The Foundry for such things. And that's the Foundry which has infinitely more resources then LW3DG has. In any event, I'm glad that such issues were noted and reported by Jason and I'll be glad to see them handled in this release or next or possibly both.

souzou
04-13-2016, 05:26 AM
Nobody argued "all issues should be fixed within their version cycle" in the first place. What relevance do such statements have to a discussion about LW3DG sitting on fixes for many months after announced resolution? Deferring release of bug fixes to work on "new version efforts" is explicitly prioritizing new efforts ahead of bug fixes.

Except these bugs were only reported 2 months ago, with LW3DG already well into the LWnext development cycle, and they only responded/resolved them in the last few weeks. So they've hardly been sitting on fixes for many months.

DrStrik9
04-14-2016, 09:12 AM
The bullet and other physics-related problems in LW 2015.3, and backward, drove me to endure learning the horribly-confusing Blender interface, to be able to use a MUCH more fully implemented set of Bullet functions than LW's. Along the way, I have confirmed once again that no software is perfect, including Blender. But when calculating sims, Blender is maybe 7-10 times faster than LW, and doesn't corrupt its own sim and scene files because of an arcane 32-bit limitation. Blender does, however, have similar sim SIZE limitations as LW, but for different reasons. There's always one more bug, and most of them seem to be related to strange memory leaks. On the upper end, Blender can bring an entire system to its knees, requiring full-system reboot. However, if kept to a reasonable size, Blender's hard-body, soft-body, cloth, particles, smoke, and fluids integrate well with one another, allowing the creation of some pretty amazing visual effects. And Blender's fracture actually works as well. The worst problem is that learning Blender is a bit like learning a foreign language, like maybe Mandarin, Cantonese, or Russian. And don't get me started on the Blender windows interface ... Oh my, what a mess.

50one
04-14-2016, 09:29 AM
Blender interface is different and of you don't appreciate you deserve to be beaten up with a stick.

:D