Good morning all those in Eastern Standard Time reading this.
New to GP, my daughter and I are working to convert over setups from Forte 4 to GP and I’m noticing MUCH higher CPU usage in GP that I’m trying to get to the bottom of.
Initial patch load in both systems’ rack spaces are around 10-11% Overhead. It is what it is …
When I play a simple 4 note chord, GP jumps up to 28-33% CPU whereas my Forte sticks at 14% !!! This gets much worse as I play more notes and more complex configurations as you can imagine but my Forte stuff doesn’t grow in usage this way.
Simple test configuration (no Effect, etc.), both systems use same hardware and VSTs:
Dell Laptop i5 3.2 ghz, 12gb RAM
New 3rd Gen Scarlett 18i20 and drivers. 48khz, 128 buffer.
Both Forte and GP contain the simplest setup I can make here with some MIDI splitting to keep the Bass where it needs to be and the 2 other VST’s layered.
Axiom 61 to control 3 VST’s:
Arturia JP-6V, Syntronic 2 ‘Bully’ for Taurus and Arturia DX-V.
So, I am not 100% sure how to approach such a difference in performance becuase I am not expecting this outcome at all.
Where it started coming into play was when I was also playing around setting up more complex Rackspaces with mixers, effects etc and working through the initial trials and errors of ‘more’. That was when the glitching started to return with CPU usage in GP getting into the 50-70% on fairly simple but ‘busy’ playing.
Any suggestions on what I can do to try and figure out where things are falling apart would be appreciated.
Can you upload a rackspace, I could check on my Mac?
What % are you talking about?
The task manager or the cpu usage shown in Gig Performer?
CPU usage reporting in GP / Forte Programs.
Ultimately, I start arriving at Audio Glitching earlier in GP once my CPU in GP starts rising.
My more complex Rackspaces I was working on was where I really started to pay attention to it falling over audio wise with a little crackling and started to get back to basics and work up from there.
Good point though, I can also look at keeping Task Manager up and see what the CPU itself says.
I am away from my gear at the moment but can upload the Rackspace soon.
That doesn’t really answer the question (at least not for those of us not familiar with Forte)
First of the, the CPU usage in GP is solely that of the audio processing. Is that also the case with the Forte one?
Not quite sure what you are doing here but make sure you’re not just trying to emulate Forte with GP - that is not a good idea - the paradigms are quite different. It’s worth reading this blog article written by an ex-Forte user.
Well, which is it — the simplest setup or did you make a more complicated one with GP? Also, did you make sure the same plugins are bypassed (or not) on both systems when comparing?
Also, how did you configure the audio drivers in Gig Performer?
FYI – I just tried the latest version of the JUN-6V on a Mac using internal speakers only, no audio interface.
You’ll note (as I did) that the JUN-6V displays essentially the same amount of CPU cycles as does Gig Performer, meaning that it is the JUN-6 that is sucking up all those cycles.
Very interesting find !
I hadn’t dug into the specifics on what VST’s yet at this point as I was hung up more on the overall difference in what was reporting between my Forte and GP setups.
So, I will compare each Program vs Task Mgr. usage to see how they correlate and also try replacing different VSTi’s to see their effects on things.
Wondering why one would report or handle a VST at a much lower usage than the other on similar setups… not that I’m only hung up on that, just the audio crackle being an issue at one point along the way.
I will also replace some VST’s in my test just to see if this levels the playing field between the two setups.
Thanks ! These were the types of skills I was hoping to expand on during this whole discovery phase.
Seems I missed some of your questions / comments that came in while I was onto other replies…
CPU usage: I will compare CPU monitoring between Task Manager and the 2 programs to see what the differences is.
I have read the Blog you mentioned. No, I am working from ground up with the online tutorials etc. and building as suggested there. The faster I forget the old way, the better overall.
You asked “Which is it ?”. I have built Rackspaces with lots of neat Twiddly Bits, Global Rackspaces and Effects, they have worked very well. Fantastic options here ! Thanks to all the tutorials again on how to start and user input has been fantastic ( especially handling MIDI input messaging blocks for layering, Splitting, etc.
However, I did start to notice the crackling with some more heavy playing and layering / splitting a few VST’s and saw the CPU in GP in the 50-70% range.
This is why I started to investigate.
My first logical choice is to strip down a new Gig file with a simple Rackspace and try to figure out where my issue lies.
Hence my posting.
Do, now that I got some leads to chase, this is great !
The Focusrite drivers are native OS so they set themselves up automatically. The only thing I can do is change the SampleRate and Buffer size. Too bad my Macbook is too old for the Scarlett drivers or I would have tried popping this all into the Mac to give it a whirl too.