Thanks, I have tried the support e-mail you suggest. However, with the lack of response as of yet to the help ‘chat’ function on cantai.app; as well as e-mails directly to R. DeCosta, I’m just not sure I can hold much hope!
That’s my response, not bad at all, really 
:
Hi Benji,
Thank you for the detailed report and specs — that’s exactly what we need.
You’ve identified a real issue: on hybrid Intel CPUs (P-cores + E-cores), Windows can schedule the Cantai renderer on the Efficient Cores rather than the Performance Cores, which limits rendering speed. We’re passing this to our engineering team to address in an upcoming update.
As a temporary workaround, you can manually set the CPU affinity for the Cantai process to use only the P-cores via Windows Task Manager (right-click the Cantai process → Set Affinity). That said, it’s a manual step each time, so we understand if you’d rather wait for the fix.
Thanks again for flagging this with such thorough detail!
Best,
— Cantai Support
Some results after testing. Now with the version Cantai v2.2.3. Cantai Engine connected.
I can’t find bass/tenor Jon in the list of voices. And that is the voice which is automatically chosen when using the Dorico playback template for Cantai.
I use an existing file and by changing playback template try to get the desired output. I have four flows in the file with vocals SATB (mis). However rendering takes forever and always one of the voices will not render and not heard. I then move one note up or down and wait till again a task becomes active. Thereafter I reset the note to the proper position. It is added value but it is not stable to me. Is there something I can do to help?
Update,
I started an empty project and imported the four flows into this new instance.
Cantai worked like a charm. Then I made some changes in the Gloria. The first changes went verry well and quick. But after more changes the rendering took longer and became making mistakes.
It seems memory isn’t released or something like that
Update - Update
I closed the program and started again. This solved some issues. But than I also found out when creating audio, you can only get audio from the first flow. So I have to move the Gloria before the Kyrie in setup and than render again, and than export audio.
I’ve been able to get audio from any flow (Mac version). Now, exporting it is another matter. When exporting audio, I have to move the flow of choice for exporting to be the first flow unless i export an entire work of multiple flows all at once as a single audio file. I’ve interacted with Cantai support team and they know that is an issue. They actually recommended moving the desired flow for audio export to be the first flow before exporting and be sure to let it re-render before exporting. The notifications of Cantai tasks in progress in the status bar is helpful but rendering takes ever so long…
New Update, been able to work better with Cantai now, exporting seperate stems isn’t working and it keeps misty if a line has been rendered or not. Wil try the new Dorico update.
I assume for Cantai it will be better to own a faster computer than my M1 Max
Canti works perfectly fine on both my MBP M1 and my MacStudio M2MAX so you should be fine so long as you have several GB of disk space available to Cantai.
EDIT: “Perfectly fine” describes Cantai doing its best at the moment with the latest release. The overall product still has much to improve upon. Slow processing and some odd pronunciations as well as inability to address divisi on the same staff. The latter may have been addressed through the latest Dorico update allowing Cantai to use Independent Voice Playback.