Noel Borthwick

That's how Russian roulette works with plugins that corrupt the heap it may work today and fail tomorrow

Has nothing to do with newer CPU's its how parallel processing of audio workloads is handled. But as I said I cannot comment until we have your project. If your project file plays ok on a different PC then you would need to troubleshoot at the PC level.

Where do you have these instances? In different tracks, stacked up or in buses. Best thing is to post the cwp file so others can see the layout of the plugins and the exact routing.

Its impossible to offer suggestions without knowing the setup of the project. Load is not distributed equally because of serial dependencies for audio processing. If you have one plugin in a track that takes a long time to process, it will become a bottleneck even though other tracks have low load, because the engine has to wait for the track to finish before it can be mixed. If you post your project file it may become more clear where the bottlenecks lie in your case.

It had nothing to do with Cakewalk not wanting to support it. Roland stopped supporting their hardware. If they still supported it we would have continued. We put a lot of effort into the VS700 and even extended our control surface SDK to do all sorts of editing stuff. This is all open source on Github if someone wants to utilize it for a different product.

Some drivers don’t terminate and hold the app captive. I’ve not seen this in many years but the some of the old Roland Drivers did this.

It only causes the behavior in cbb because it's corrupting the right chunk of memory to trigger the crash The app vanishing without a crash is almost surely caused by the plugin corrupting the heap. Heap corruption is like Russian roulette. It may or may not occur even from one version to another of the app because it is random. heap corruption issues are best diagnosed in a debugger using appverifier but even there it can be tricky to find.

You could say that you have an affinity for hard cores

How do you set this?

Can you zip and upload a version of the project that has the issue? You can pm me the link.

Which build are you running? This issue was fixed in the last update.

@ChristopherM Thanks I looked at the dump. Took awhile to uncompress a 32 GB file This is an unambiguous plugin issue. Cakewalk is simply doing an autosave and the plugin hangs when we try and retrieve the plugin state to save it. Cakewalk calls the getState function on the VST3 plugin and the plugin never returns from the call. This is something only the vendor can solve. ntdll.dll!RtlEnterCriticalSection() Unknown Non-user code. Symbols loaded. Chromaphone 3 Shared.bin!000000006b665367() Unknown Non-user code. Binary was not built with debug information. Chromaphone 3 Shared.bin!000000006b647f1a() Unknown Non-user code. Binary was not built with debug information. Chromaphone 3 Shared.bin!000000006b5ddfe9() Unknown Non-user code. Binary was not built with debug information. Chromaphone 3 Shared.bin!000000006b611ee7() Unknown Non-user code. Binary was not built with debug information. Chromaphone 3.vst3!00007ffb182656a7() Unknown Non-user code. Binary was not built with debug information. > Cakewalk.exe!CVST3PlugWrapper::GetChunk(bool isPreset, unsigned char * * pData=0x000000000014e690) Line 1974 C++ Symbols loaded. The plugin developers should be able to load up the dump file and find out why the plugin is deadlocking. Were you playing when autosave kicked in?

Hi @ChristopherM, Unfortunately your dump link is not available anymore. Can you resend it please.

Hi @Scott Wheeler, Thanks for the message. I'm not familiar with Sitala so I should check it out. I assume that the plugin is a multi-out VST instrument? Both VST2 and VST3 have multi-out support. What is the purpose of the custom API? Feel free to PM me and we can discuss further.

I have forwarded the issue with Air Riser to the InMusic team. They solved a problem with the MPC plugin recently so perhaps they may have some insight.

@Fabio I had not seen your message. Please mention (@) if you need to draw a specific persons attention to a message. I looked at the dump file and the crash is not Waves related so it will be useless to them. The crash is in air riser so you should send it to the vendor for that plugin. Version info from the dump file is below. Unhandled exception at 0x0000000037DD975D (theRiser_x64.dll) in Untitled_12052020_110141.dmp: 0xC0000005: Access violation reading location 0xFFFFFFFE39480214. theRiser_x64.dll theRiser_x64.dll *c:\program files\Cakewalk\vstplugins\AIR Riser\theRiser_x64.dll 1.00.7.19000 10/28/2015 6:11 AM 0000000037D70000-0000000039A6C000 Untitled_12052020_110141.dmp This is not a cakewalk crash so only the vendor can analyze it. If you have a different dump file with a crash in OVox you can attach that.

@Darcy

We've been close to Microsoft for over 20 years

There is no processor architecture specific coding in Cbb for multiprocessing and I would wager most other apps as well. It would be dangerous to do that since performance would vary across different CPU models. To Cbb all threads are equal. My guess is that with threads on virtual cores you are at the mercy of the OS implementation, system load and the hyperthreading implementation, since these threads are not truly running on a dedicated core. This old post kind of sums up this behavior. My guess is that when the primary core is really busy due to a heavy load, the virtual core doesnt get a lot of time to run. In a mixing workload with multiple tracks the full cycle cannot complete until all channels are complete so if some cores are starved I can see it having a detrimental effect to simply running without hyperthreading. I could try and add an experimental mode where we filter out processing workloads on threads assigned to virtual cores. Or perhaps do this dynamically based on the actual CPU workload. i.e. if the workload is lower, then utilize hyperthreaded cores but as it gets closer to 80% or some value dial back to physical cores only. Who knows if that will help

Please read this article. Plug in specific crashes should first be reported to the vendor since they are able to best investigate the problem. If you send us a dump file AND the project file we can forward it to Waves as well.

Ben or anyone else still having issues with iRig HD 2 i let us know if this is ok in the latest build. The folks at IK multimedia have validated the build and it works ok for them so if you have issues please contact IK directly and they can help troubleshoot it.

There are differences between logical and physical cores because the OS handles them differently. Esp in the case of hyperthreading or in the case of AMD simultaneous multithreading (SMT). There are API's to identify whether a core is logical or physical. See GetLogicalProcessorInformation See this benchmark that has more details on how hyperthreaded cores can effect performance.

Yes you can set a preferred core using the SetThreadIdealProcessor function. We in fact have support for this. There are CPU architectures where some cores are efficiency cores and others high performance. While this is very common in the mobile silicon (phones, iPad, etc) its is now becoming mainstream even on other CPU's. AMD's hyperthreading was different from Intel (not sure about their current offerings) and hyperthreading was not as efficient before. Also the new Apple silicon has low powered cores and high powered ones.

Which version of Cakewalk are you running? In case you can't open the app you can check the version in windows via the file properties of the exe file. Make sure you have installed the latest 2020.11 release which is build 99. If you are unsure you can uninstall the program and then re install via bandlab assistant.

@Erlend Holm your crash is different from the OP, its not the same issue. Its failing in the Waves 12 shell. This is something you will have to follow up with Waves on since its their crash. Please contact waves tech support and give them the same dump file AND the track template containing the waves plugins. If you send me the template file I can also forward it to them. Unhandled exception at 0x00007FFA19D6B3C3 (WaveShell2-VST3 12.0_x64.vst3) in Track 1_11282020_015537.dmp: 0xC0000005: Access violation reading location 0x0000000000000019.