View Full Version : Crash with model in 2015.3 Modeler

09-05-2018, 08:14 AM
I've come across a problem with regard to 2015.3 Modeler on a Windows 8.1 PC.

I have two models. When using the Layers panel to select their layers to display them both together in Texture mode I get an instant crash. This happens only on one machine and not with another one I have tried with very similar specs.

I need to get a recommendation as to what to do about this, as it doesn't seem right that one machine would be causing a crash while another does not on the same exact files.

I thank anyone in advance that can help me with this.

Below is a copy/paste from the error event log that I got from Windows Event Viewer:

Faulting application name: Modeler.exe, version:, time stamp: 0x55c53160
Faulting module name: hwrender.dll, version:, time stamp: 0x55c530fd
Exception code: 0xc0000005
Fault offset: 0x0000000000051a52
Faulting process id: 0x22ec
Faulting application start time: 0x01d4446b2a5b65ce
Faulting application path: C:\Program Files\NewTek\LightWave_2015.3\bin\Modeler.exe
Faulting module path: C:\Program Files\NewTek\LightWave_2015.3\bin\hwrender.dll
Report Id: a0cd51c2-b05e-11e8-83f7-fcaa142d04ad
Faulting package full name:
Faulting package-relative application ID:


- System

- Provider

[ Name] Application Error

- EventID 1000

[ Qualifiers] 0

Level 2

Task 100

Keywords 0x80000000000000

- TimeCreated

[ SystemTime] 2018-09-04T16:21:51.000000000Z

EventRecordID 325190

Channel Application

Computer Jims_Lightwave


- EventData

C:\Program Files\NewTek\LightWave_2015.3\bin\Modeler.exe
C:\Program Files\NewTek\LightWave_2015.3\bin\hwrender.dll

09-05-2018, 09:16 AM
hwrender.dll suggests the variable between the working/non-working systems is the GPU and/or its driver. Can you set the driver version on the faulty computer to match the one that works (usually by updating, sometimes by rolling back)?

09-05-2018, 09:34 AM

So you mean to see what version of NVidia video driver is working on the good machine and to see if I can roll it back or update on the other one that is crashing?

I know I did an update recently on the machine that is crashing.

The other machine has not been updated in awhile.

09-05-2018, 09:45 AM
First try updating the crashing machine...assuming a recent bad update has just been corrected with another update. If there are still issues, try rolling back to the driver version that works on the other system.

This, of course, is a guess. You could post your system specs and driver versions to see if anyone else is experiencing the same problem...or someone might see the specs and tell you that they should be fine, and to look to something else.

09-05-2018, 10:17 AM
The only real difference between the two machines is that the working one has a 770 Nvidia card, while the non-working/crashing machine has a 970.

I have tried other slightly simpler models and had no crash on the problem LW PC.

Memory issue?

09-07-2018, 09:38 PM
I tried a Nvidia driver version rollback and still am having the issue with the model.

Really befuddled with this!

Any other suggestions?

09-09-2018, 08:36 PM
In case anyone checks back on this. It's fixed!

I tried loading the models in 2018 and it worked fine on the same machine I had trouble with in 2015.3. Said to me that it couldn't be memory, or the display driver.

I redid the Modeler configs and it fixed the problem. Not sure how it got corrupted though.

09-10-2018, 05:58 AM
Are you using custom polygon handlers? e.g. LWCAD, TrueArt's Bezier, EasySpline, etc. etc.

09-17-2018, 06:12 PM
As I said, I got it working properly the other day. It was a corrupted Modeler config file that was causing the crash.

Sensei, No, none of those. No clue. Just glad I solved it.

09-21-2018, 12:13 PM
As I said, I got it working properly the other day. It was a corrupted Modeler config file that was causing the crash.

So going forward, instead of DELETING the old configs, folks need to move them aside / rename them / whatever to "disengage but keep" the old configs. When the newly-generated config file solves the problem (as it did here), they need to take the old config files, package them up, and file a bug report with Newtek saying "This config crashes LW during {X}!".

Given enough motivation perhaps Newtek will finally make LW's config parsers a bit less fragile, and the config file handling a bit less prone to corrupting the users' configs.