![garritan aria player trouble only plays one of 16 sounds garritan aria player trouble only plays one of 16 sounds](https://i.ibb.co/wMDsj97/Garritan-Abbey-Road-Studios-CFX-Lite.png)
And it seems to work just fine.Īny thoughts on this theory, or anyone who can test it out and see if they can replicate the results? By setting RAM low, I'm limiting the amount CFX can cache, and so the majority of notes played are forced to be streamed from disk. It may result in higher CPU or I/O load, but I don't care about that. Just trash it, and stream everything as inefficiently as possible. So I've wondered before if CFX could have a setting to NOT cache pre-played notes in RAM. But the key is that CFX can stream straight from disk at low buffer without nearly ANY RAM at all, and problems only appear when RAM is filled. Maybe MacOS sees that the RAM allocation is exceeded and shunts new samples back to SSD swap disk, or something else. And if pops/cracks start to show up only after the RAM buffer is filled, what does that mean? Maybe that some garbage collection or swapping may be introducing latency into the process that is causing buffer misses. If it can play pop-free from fresh bootup until RAM is filled, from where is it loading these samples that are playing pop-free? Presumably it's streaming/loading it real-time from storage, which it can do just fine. This brings my back to my discomfort with CFX's behavior. No pops or cracks AT ALL, even at 64 buffer 92khz (32 buffer did have pops almost immediately). Remembering my experience from before, I tried starving CFX of memory, and set it to 512MB and 16Kb prefetch. The M1 only has 8GB RAM, so no matter the setting, i only had to play 30-60 seconds before pops/cracks appeared. With my M1 Mini, I was getting cracks/pops at every setting, unless I set the buffer up to 512 or 1024 (which was surprisingly playable on the M1, can the buffer latency be shorter than on the old Macbook?). This never sat well with me, you know, because of the implications. Only when the RAM was nearing the limit would cracks/pops start showing up. If I maxed out CFX's RAM allocation to 16GB, I could play for 5-10 minutes before the pops started showing up, and during this time I could see the RAM usage steadily rise in the CFX interface. I think I just discovered something really odd about CFX:īack when I had a 32GB MBP, I was trying to find the best settings for pop/crackle-free playing with low latency. The dealer was like 4 hours from my house so I couldn't just run back and try to troubleshoot it, and I didn't want to try and troubleshoot it after I bought it.
![garritan aria player trouble only plays one of 16 sounds garritan aria player trouble only plays one of 16 sounds](https://usermanuals.garritan.com/ARIAPlayer/Content/resources/images/view_settings.png)
That wasn't the only factor, but it was significant enough that I would have taken a longer look at the NV10 if the noise wasn't present. If I would have had a Roland UM-ONE mk2 cable with me, I might own an NV10 instead of an N1X today. I was using a regular USB A/B cable and stereo cable to connect the computer/piano.
![garritan aria player trouble only plays one of 16 sounds garritan aria player trouble only plays one of 16 sounds](https://s1.manualzz.com/store/data/006872236_1-3d1b916f1151f930012e6dfe1dc85f25.png)
#Garritan aria player trouble only plays one of 16 sounds mac
I had noise introduced when I tried the NV10 with my Mac mini. I'm glad everything is working for you now. It could be because I'm using a Mac Mini instead of a MBP now, so there's no internal trackpad or other components to add to the noise, but whereas I had previously found it an ~80% reduction in noise, with my current setup it's easily an alternative to the DIN-5 MIDI to USB cable. I also tried the iDefender again, and this time it's quite good, more than acceptable. So if you're trying to solve a VST ground loop issue, the Roland MIDI cable gets my vote. Even with Line in Volume set to Max, Line-In attenuator set to +10db, master volume switch at max, and my ear pressed against the NV-10's speaker grille, there's zero noise from the Mac. And it works perfectly! It sends MIDI to the Mac, and I can hear NO noise from the NV-10's speakers whether the line-in is plugged in or not.
![garritan aria player trouble only plays one of 16 sounds garritan aria player trouble only plays one of 16 sounds](https://dt7v1i9vyp3mf.cloudfront.net/styles/news_large/s3/imagelibrary/G/GarritanPersonalOrchestra4_01-NOjH0OK0NeZ30BwKtR6EQhsd5vHAnguD.jpg)
Solution: today, I received a Roland UM-ONE mk2 USB to DIN-5 MIDI cable. I had also tried partially inserting the MIDI plug since I heard that can help. I also tried a cheap Amazon USB-C to DIN-5 MIDI cable, which worked great except it introduced its own kind of ground loop: clicking sounds being sent to the Macbook every time a MIDI event triggered. I tried running the Mac on battery power, changing stereo cables, using a USB-C cable, using a USB DAC, inserting an iDefender on the USB cable, but nothing was 100% successful. This resulted in a ground loop that I could never fully get rid of. So I used the typical setup of USB-MIDI between the NV-10 and the Mac, and a 3.5mm stereo mini cable from the Mac's headphone output to the Line-In on the NV-10. I'm happy to report that my historical ground loop issue appears to be 100% solved!Ĭontext: I want to use a computer (Mac) to play a VST, and output the sound back to the NV-10's speakers to play.