Offline Bounce Session Optimization

[Myocum93]

Ok fellow geeks, here's a technical question regarding Pro Tools Offline Bounce.
I have the Late 2014 iMac (maxed out config) with the 4.0GHz Quad Core Intel Core i7-4930K Processor. Given that this processor has hyperthreading capabilities, this thing should be absolutely blazingly fast when it comes to rendering offline content (not to mention handling pretty decent loads of native plugins). I'm currently running into thunderbolt chassis with an HD Native card and an Omni.

At the moment I'm bouncing eight 30 minute episodes of a really simple session for a cooking doc. Its only about 20 active tracks deep and the plugins are just a ProQ3 on each audio channel (set to zero latency), some heavier processing plugins on my dialogue tracks such as WNS, Fab's De-esser, ProQ3, Pro-C2 (no oversampling), and then AVID's pro-limiters and Multibands on my masters as part of my standard commercial template. Honestly its not a heavy load for a session, yet when I offline bounce I'm barely able to get past x2.8 or x3.0 speed. (btw I am able to cache the entire session to ram only using 6 of my available 32gigs)

At SONY today we were using a single HDX3 Trash Can with the 6 core 3.2GHz config running a MASSIVE session (DX, SFX, MX all in one) with all kinds of crazy native plugins running as inserts (don't ask... we didn't create the template) with plugins like speakerphones, upmixers, ProQ2, De-essers, compressors, gates, verbs, limters, izotope plugins running real time, vocal rider, W43, Neutron 2 - and the list goes on...... This is across nearly 600 tracks of audio all loaded to the max. Its a less than ideal set up for any session to run smoothly. That being said, we hit offline render to bounce a reel and it was hitting x30.0 export speeds!
So my question is why.... Even with 12 vs 8 cores at 3.2 and 4.0GHz respectively, it shouldn't be THAT big of a difference, especially considering the load difference and both sessions were fully cached. Does anyone have any insight on a way that I might not be fully optimizing pro tools or some way its throttling my power? I'm considering writing AVID to see if they have an answer.

Sorry for the long post but wanted to give as much detail as possible!

[BScout]

What’s your drive in the iMac? The trashcan has NVMe drives (which can be 6X faster than a standard SSD.) Cached or not doesn’t matter when a file still needs writing to a drive. That would be a bottleneck for speed.

[BScout]

Also, it only takes one plugin to cock it up. So even though there are more tracks and plugins on the Sony session, something (like Waves DNS) on the iMac could slow it down. Or if one plugin is written to take advantage of a processor feature that doesn’t exist on your older iMac’s chip (like AVX-2)

Need to do a like for like comparison verses different sessions. Or deactivate one plugin at a time and see if one of them is the bottleneck.

[albee1952]

Originally Posted by BScout

Also, it only takes one plugin to cock it up. So even though there are more tracks and plugins on the Sony session, something (like Waves DNS) on the iMac could slow it down. Or if one plugin is written to take advantage of a processor feature that doesn’t exist on your older iMac’s chip (like AVX-2)

Need to do a like for like comparison verses different sessions. Or deactivate one plugin at a time and see if one of them is the bottleneck.

Exactly my thoughts as well. Example: a session with MiniGrand in play for piano will bounce a 3 minute song and 15 seconds(give or take). Swap out MiniGrand for KeyScape and the same 3 minute song takes 2 minutes to bounce