Decimator crashes DAZ
Klling Cop
Posts: 0
I bought the Decimator Plugin to reduce the number of polygons.
Actually it works well.
But the great problem is that it really often crashes my DAZ :( Especially items with a high mesh resolution cause very often a crash when I click on 'Prepare to Decimate' or 'Set Weights'.
My workaround is to edit only one item after the other and to save as many times as possible.
But even then it crashes quite a lot of times.
Does anyone have the same trouble with this plugin or knows what's the problem?
Comments
It might be a memory problem. What OS are you using? What version of DS exactly (Help > About DAZ Studio)?
What are your system specs, RAM, processor etc.?
Mac OS X 10.7.5 and DAZ Version 4.5.1.6 Pro Edition (64 bit)
system specs are
processor: 3.06 GHz Intel i3
RAM: 16 GB (right now 4,5 GB available)
graphic board: ATI Radeon HD 4670 256 MB
you think the problem might be not having enough (free) RAM?
I don't know anything much about Macs, but you seem to be using a lot of your available memory on other processes. I would think that Decimator will want as much memory as you can give it since it has to do a lot of work there.
Can you release some more of your available memory, and give that a try?
A Mac user may come along with a better suggestion.
I have the exact same problem.
High res meshes, you just have to decimate like 2-3% at a time and save until you get it right.
I've been monitoring the crashes with the decimator and they all seem to be an Access Viaolation at some point. For those that don't understand how memory is managed in an OS, this basically means that it is trying to access memory in a way which corrupts the heap (the memory space of the application) so when the application tries to access a memory address which is store in the heap it might have been overwriten with a bad address which causes the crash. On some of the crashes I can see that the decimator is trying to access memory directly and this can definetly cause these types of issue.
From what I read in the decimator's description it does not look like DAZ was the only one working on it so I don't know if they can fix it or not. I just which they did, cause working like this really sucks.
In the crash log below you can see where they are trying to access memory directly however this may or may not be the reason for the crash. Without source code and instruments attached to the process to turn on heap bounds I can not tell.
Process: DAZStudio [61249]
Path: /Applications/DAZ 3D/*/DAZStudio.app/Contents/MacOS/DAZStudio
Identifier: com.DAZ.DAZStudio
Version: 4.8.0.59 (4.8.0.59)
Code Type: X86-64 (Native)
Parent Process: ??? [1]
Responsible: DAZStudio [61249]
User ID: 501
Date/Time: 2016-01-21 13:11:56.341 -0500
OS Version: Mac OS X 10.11.1 (15B42)
Report Version: 11
Anonymous UUID: AB91162C-E727-9793-162A-FC57F2213F75
Sleep/Wake UUID: B330257F-7DB6-4D08-8CA1-3855F9FFEE7B
Time Awake Since Boot: 260000 seconds
Time Since Wake: 7600 seconds
System Integrity Protection: enabled
Crashed Thread: 0 Dispatch queue: com.apple.main-thread
Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000000148fe4000
Exception Note: EXC_CORPSE_NOTIFY
VM Regions Near 0x148fe4000:
CoreAnimation 0000000148fe2000-0000000148fe3000 [ 4K] rw-/rwx SM=ZER
-->
CoreAnimation 0000000148fe6000-0000000148fe7000 [ 4K] rw-/rwx SM=ZER
Application Specific Information:
Performing @selector(qtDispatcherToQAction:) from sender NSMenuItem 0x1343effa0
Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0 libsystem_platform.dylib 0x00007fff840e2c49 _platform_bzero$VARIANT$Haswell + 41
1 QtCore 0x00000001041f67d8 qMemSet(void*, int, unsigned long) + 24
2 libdzcore.dylib 0x0000000100c84391 DzSkinBinding::applyScale(DzVertexMesh&, DzNode&) + 661
3 libdzcore.dylib 0x0000000100c8bf2d DzSkinBinding::apply(DzVertexMesh&, DzNode&) + 101
4 libdzcore.dylib 0x0000000100944276 DzObject::buildWSGeometry(DzNode&, bool, DzVertexMesh*) + 282
5 libdzcore.dylib 0x000000010094210e DzObject::update(DzNode&, bool) + 784
6 libdzcore.dylib 0x0000000100b11cee DzScene::update() + 50
7 libdzcore.dylib 0x0000000100dcacc2 DzUndoList::undo() + 722
8 libdzcore.dylib 0x0000000100dcb148 DzUndoStack::undo() + 230
9 libdzcore.dylib 0x00000001001f7c19 DzAction::qt_static_metacall(QObject*, QMetaObject::Call, int, void**) + 219
10 QtCore 0x000000010430b381 QMetaObject::activate(QObject*, QMetaObject const*, int, void**) + 2001
11 QtGui 0x000000010459afa2 QAction::activate(QAction::ActionEvent) + 194
12 QtGui 0x0000000104551ac5 -[QCocoaMenuLoader qtDispatcherToQAction:] + 69
13 libsystem_trace.dylib 0x00007fff8b873082 _os_activity_initiate + 75
14 com.apple.AppKit 0x00007fff8662d811 -[NSApplication sendAction:to:from:] + 460
15 com.apple.AppKit 0x00007fff8662d5ab -[NSMenuItem _corePerformAction] + 336
16 com.apple.AppKit 0x00007fff8662d30b -[NSCarbonMenuImpl performActionWithHighlightingForItemAtIndex:] + 114
17 libsystem_trace.dylib 0x00007fff8b873082 _os_activity_initiate + 75
18 com.apple.AppKit 0x00007fff8662c1f9 -[NSMenu performKeyEquivalent:] + 357
19 com.apple.AppKit 0x00007fff8662b3a0 -[NSApplication _handleKeyEquivalent:] + 920
20 com.apple.AppKit 0x00007fff865511ed -[NSApplication sendEvent:] + 4274
21 QtGui 0x00000001045525f0 -[QNSApplication sendEvent:] + 80
22 com.apple.AppKit 0x00007fff863b7d9a -[NSApplication run] + 796
23 QtGui 0x000000010455af0f QEventDispatcherMac::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) + 775
24 QtCore 0x00000001042f2db8 QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) + 394
25 QtCore 0x00000001042f6183 QCoreApplication::exec() + 175
26 libdzcore.dylib 0x0000000100242c54 DzApp::doExec() + 100
27 libdzcore.dylib 0x000000010024e5cb DzApp::run(DzApp::GraphicsMode) + 355
28 com.DAZ.DAZStudio 0x00000001000015f6 main + 496
29 com.DAZ.DAZStudio 0x00000001000013e4 start + 52
Hello,
I cannot come with any good tips. I've reported this crash on OS X and the workaround I've been using when I've been using it (to create low poly background figures) has been to decimate the same way you walk on eggs, careful, careful.
I stopped using it, but maybe it's worth bug reporting again.