Old bug in DP's Trim plug-in??? [YES, There's a bug]

[FMiguelez]

Am I the only one who really likes Next Gen Pre Gen? Been using DP since v.4, on several different Macs, and I'm delighted to have a system that is more efficient than ever, without using a myriad of workarounds. Yes, there may be a problem with some plugs, but really, what are they? My primary plugs are Kontakt, Aria, SampleModeling, and even Best Service, and I've yet to run into anything that stops me from getting things done. Things are so efficient that I can't think of a reason to globally disable this function.

But that is, or should be, on a case by case basis.
I'm really glad for you that it rocks in your system. But my point is>

WHY should MOTU decide for us how to work?
What's so hard about giving us the option, so anyone can decide for themselves and according to their own workflows, preferences and styles?
We have an awesome set of user Preferences. This would be a great addition to it.

It feels like someone took away my candy out of the blue and arbitrarily.

The thing is, if that is going to be MOTU's new paradigm, then their engine must be perfect and work every time with every plug-in and every Virtual Instrument ...

We are discussing software. They will come as close as they can but perfection is unlikely.

I understand that. I didn't mean perfection literally. That doesn't exist.
But you're making my point for me. PRECISELY because of what you stated, the logical and obvious thing would be to make this optional, so if a user has a plug-in that won't work, he simply unchecks a little option box and he goes back to "Original Mode".

Most users reasonably expect, at the very least, most - the big majority of VIs or plug-ins - to work flawlessly. If this doesn't happen, MOTU will make themselves a constant target from users complaining that x, y or z plug-in or VI is not working. That's why I think they are making things unnecessarily hard for themselves.

So, since they can not assure 100% compatibility with EVERY single plug-in or VI in existence, or if the user simply PREFERS to work the way he's used to, then a little option box in the preferences can be checked/unchecked and voila... everybody happy...

[FMiguelez]

Put your head down and prepare to be swarmed!

Nah! It's all for a good purpose> The continuous future greatness of DP.

Now, where did I leave my machete? Al ataque!!!!

[bayswater]

Am I the only one who really likes Next Gen Pre Gen? Been using DP since v.4, on several different Macs, and I'm delighted to have a system that is more efficient than ever, without using a myriad of workarounds. Yes, there may be a problem with some plugs, but really, what are they? My primary plugs are Kontakt, Aria, SampleModeling, and even Best Service, and I've yet to run into anything that stops me from getting things done. Things are so efficient that I can't think of a reason to globally disable this function.

Can't say I "really like" it, but its not giving me problems now. I DID notice some dropouts and spikes using Yosemite and DP 9.12, but I didn't do much with that combination before updating to Sierra, which I had to do to run the latest Logic update.

I just tried out my "stress" test project on this setup using DP 9.13. It has 64 different test signal soundbites 48K 24 bit dropped a few times each into 192 audio tracks staggered so the number playing at once increases over time to 192. Each track has the MOTU Para EQ insert. There are also soundbites with silence and some with random clicks, applied to all 192 tracks. These output to 24 auxes, each with the MOTU Dynamics plugin. There are 64 MIDI tracks going to 16 instance of Kontakt playing 4 factory patches on each instance. At one point all audio and VIs are playing. MOTU limiter on the master. This might not be as big as some templates described here but it's a lot for my elderly iMac. Buffer is 512.

This plays with no issues, dropouts or audible glitches reported by the DP meters, or Activity Monitor. As noted in another thread, the graphics on the Track window are very sluggish, but things are OK on other windows. This was not the case when I did this test with DP 8.

Activity Monitor User CPU % jumps up and down a few points at each readout, but here's what I got:

Idle 4%
Silence 5%
Clicks 19%
Test Signals 20%
VIs 67 %
VIs and Test Signals 71%
VIs with 1024 buffer 67%
VIs with 2048 buffer 64%
iMac fans ramp up to audible levels

These numbers are comparable, maybe a little better than what I got with DP 8. Latency on VIs is noticeably better, so for the most part I'm getting what MOTU said I'd get.

The real surprise is Logic. With Logic 9, the results were slightly better than for DP 8 but there is a big difference with Logic X.

With Logic 10.3.1, using the basic EQ and Dynamic plugins that come with it:

Idle 10%
Silence 15%
Clicks 15%
Test Signals 15%
VIs 28%
Test Signals and VIs 29%
VIs @ 1024 19%
(Logic won't do a 2048 buffer)
No audible fans.

Graphics are about the same as DP's (excepting the sluggishness of the DP Tracks window noted above)

I think Logic treats VIs differently, so if a specific sample for a specific VI is used for more than one track, it loads one instance into RAM and uses it for all notes. Sometime I'll repeat this using 64 different patches across the 64 MIDI tracks.

I repeated the test in DP 9.12. Same as DP 9.13. Same CPU, no glitches.

Also repeated the test in DP 9.02. It worked OK, no glitches, but the CPU load went to 90% when all instances of the VIs were active. If you consider what MOTU has said about these versions, that is about what you'd expect.

So, what's the common factor? Sierra.

[cuttime]

I just tried out my "stress" test project on this setup using DP 9.13.

Thanks for your input. I did a comparison with MainStage 3, and noticed similar results to your Logic tests. I don't have Logic, though.

One more thing I'd like to point out about the improved latency is, as a keyboard player, the ability to play at a buffer of 64 which something I've never been able to do with DP (but could in MainStage). Maybe the difference is all in my head, but the tactile feel of the keyboard with a near immediate sound renders an experience that is closer to an acoustic instrument than to a VI. I think playing at low buffers has been challenged by some, but yes, I really can feel the difference.

[mikehalloran]

... I DID notice some dropouts and spikes using Yosemite and DP 9.12...

I noticed the same with 9.12 at 512k and Sierra but much better than 9.02 which sounded horrible below 1024k.

When I upgraded to 9.13, I was able to lower my playback buffer quite a bit to 128k and I don't hear any of that audio garbage.

I still have all three installed plus 8.07. I was opening the same test project in all three. I kept comparing just so I could prove to myself that it wasn't my imagination.