• SONAR
  • Sonar x3: Why did my performance drop massively after first run? (p.3)
2014/12/25 11:14:51
guitarblah
mettelus
I may be confused here. The post with pictures of CPU usage with effects off and effects on shows the CPU loading of effects used. Which effects are you using in this project, and how many instances?

8 buses with multiple instances of breverb, some less intensive plugs and some instances of the 64 bit plugs (multi and eq).
Honestly the improvement from turning the power settings to max has given me the confidence to keep working.
If I do run into more problems I would like to have workflow (or optimization) advice handy, but if you can't link me it don't worry about it (I will come back if I need to).
 
Thanks guys.
2014/12/25 12:36:51
microapp
It is likely that a Windows process was NOT running the first time and IS running on successive attempts. 
For Sonar itself, investigate thread optimization and also core parking. Does not look like this is the issue but check it out. ( Just saw you had AMD CPUs, not sure these park cores or what Sonar Thread opt will do for these. )
 
For windows, disable Windows Update (set it to manual) , disable SuperFetch completely.
Double check that your HDD drive is using DMA mode and NOT PIO mode.
If running anti-virus, set it to not scan WAV files on the fly. I would simply disable A/V during sessions.
Try running Latency Mon with the Processes button pressed rather than drivers.
You will prob find some housekeeping chore Windows is doing.    
2014/12/26 16:03:28
mettelus
I do not know of links for work flow specifically, but the big picture to bear in mind is that playing a wav file is simple for any application. As soon as VST(i)s come into play, more processing overhead is required (which can also be compounded by higher sampling rates). Keeping an eye on CPU usage is a good habit as you insert/use these, and you will get a feel for how much loading each consumes as you work with them. A few streamlining things which can be useful are 1) freezing CPU-intensive soft synths and 2) bouncing FX-intensive audio tracks (then archiving (not muting) the original). Both of these can lower CPU usage dramatically, yet allow for return to the original track if further edits are required.
 
I think what I caught in the OP the most was the "as many as I want" comments... everything is finite. When you bypassed FX and saw a dramatic drop in CPU usage, it is indicative of the FX themselves (what they are, what they do, and how they are being used).
 
2014/12/27 01:26:53
Anderton
Also check out this tip (Week 35) from the Tip of the Week thread. For some systems it makes a major improvement, others not so much but there's no downside to trying it. I would highly recommend trying it if you're using an ATI graphics card and drivers.
© 2026 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account