I'm through with training wheels, ready to dump Mainstage

[JeffLearman]

Guru's test is the real-world meaningful one. Tools and benchmarks can help us to find the sweet spot faster, but if you have A and B already selected to compare, might as well do the real-world comparison. Just make sure all other things are equal when you do the comparison. It's amazing how things can change without us realizing something is different.

 

For example, turn off your wireless. I recently noticed that I had very good low latency. Then later with all things being equal (I thought) my latency was still good but almost twice as high. Turns out it was TCP; disabling wireless fixed it. I have no idea why TCP was doing something one time and not the other. Perhaps there was a webpage open in a browser somewhere (I might have had Facebook open, for example.)

 

Of course, smart people don't use their music system as their browsing system. Did I ever say I was smart?

[Ashville.Guru]

@Zephonic - we posted at almost the same time...! Again, it's not about latency - it's about audio load handling. The only relevant take-home message from the DawBench studies: interfaces differ *widely* in their ability to handle that load.

 

...but big chords with lots of sustain and glissandos will send it over the edge.

This is what I'm talking about - that's exactly my experience with the headphone outs my i7 Dell, too. But since I got my firewire interface, I have to really, really, try hard (like layer a dozen instances of Pianoteq/Arturia plugins) to push it over the edge.

 

- Guru

[Tusker]

Guru's test is the real-world meaningful one. Tools and benchmarks can help us to find the sweet spot faster, but if you have A and B already selected to compare, might as well do the real-world comparison.

 

I'll certainly take a look at it. I am not a computer geek, but I am curious. As you mentioned, I won't be using the mac's wifi, although there are some attractive techniques to use Ipads as WIFI midi controllers that I'll want to explore separately. Currently, I have a clean boot with the networks off and Mainstage likes it.

 

However, as I said above, these tests are not meaningful, unless the additional capacity is needed by the user. Interesting perhaps? Fun perhaps? We may as well find out which minivans are better at burning rubber....

 

Bad analogy. I might have seen too many myth-busters shows.....

 

[zephonic]

@Zephonic - we posted at almost the same time...! Again, it's not about latency - it's about audio load handling. The only relevant take-home message from the DawBench studies: interfaces differ *widely* in their ability to handle that load.

 

...but big chords with lots of sustain and glissandos will send it over the edge.

This is what I'm talking about - that's exactly my experience with the headphone outs my i7 Dell, too. But since I got my firewire interface, I have to really, really, try hard (like layer a dozen instances of Pianoteq/Arturia plugins) to push it over the edge.

 

I gotcha. And if the OP has a FireWire port available on his MacBook that would make sense.

 

While a FW interface with excellent drivers can make a big difference, like you I doubt cheaper USB models would up the ante much.

I would never recommend the USB ones which rate 5.5 or lower. But even with those, you might find noticeable performance improvement with those over the built-in outs.

 

Apple's CoreAudio makes the onboard sound work amazingly well, especially with their own apps like Mainstage 2.

 

With that in mind, I am unconvinced that a cheap USB audio interface would improve performance, especially considering how USB is an asynchronous protocol.

 

Combine that with the sparse connectivity on modern MacBooks (two USB, one ThB on my 2013 MBA) and I feel it may not be worth the extra investment. You'd have to fork out much more for a proper interface from Mac-centric companies like MOTU or Metric Halo.

 

 

 

[Reezekeys]

Of course, smart people don't use their music system as their browsing system. Did I ever say I was smart?

Sure you are. Why can't you use your browsing system for music? I've been doing it for years. if there were any issues with it I think I'd have found out about them by now. I just make sure to do two things at each gig (this is on a 2007-vintage MacBook Pro):

 

1- cold restart the computer

2- turn off wifi*

 

(3- I disabled widgets years ago, so I don't think about it, but I remember reading that they can take up not insignificant amounts of memory and cpu).

 

* I turn wifi off mostly because it's easy and quick to do, I don't need it, so why not? Actually, I have used wifi occasionally with my music software to control Plogue Bidule with TouchOSC. So I could probably get away with leaving it on.

[PianoMan51]

Before this topic fades into the depths of the NSA's database, I want to thank everyone who spent their time to reply to my dilemma. I spent many, many hours researching the issue before asking, and got better, personal experience-based answers in the first day on KC than I got in my first week on Google. And while some of the answers seemed a little off base to me at first (get rid of the computer, go to Windows, stay with Mainstage, figure out what the real problem is) they're all really point on, from each of your perspectives.

 

So here's what I'm going to do:

1). Back off on the newly installed instruments until my current setup stabilizes.

2). Wait for Plogue Bidule to release the next standalone Mac 3-month demo.

3). Between now and then, try to isolate which piece or pieces in my current setup don't work well with others.

 

I'm guessing that I'm not the only (former) EE on this forum, so you'll understand that Bidule does look like a reasonable course for me.

 

Hey, when you play your next gig, have fun, and make sure the audience does too!

 

Jack

[AnotherScott]

Sounds like a plan! Though I think you might also want to check out the free 15 day demo of Rax, just to see...

[johnchop]

I could not get Rax to get past the AU validation. Every AU failed...?

 

My experiments with MainStage are hitting a wall, mainly due to CPU consumption and stability issues. I'm not sure I want to throw more money at it--even $30--if there are still CPU issues (and reportedly there are). I suppose if I stuck with Logic instruments, I'd be better off. :-/

 

Anyway, I'm back to wrapping my head around Live. The main shortcoming is bank/program changes for external devices. I'm looking at some MIDI VSTs that should help get around that. The idea of having to trigger clips to do such changes is daunting.

 

However, I have learned to like both Instrument Racks (roughly equivalent to an MS channel strip). I also like the ability to route MIDI from one track to another. This allows me to abstract controllers away from specific VSTs. That is, I can create a "Piano Controller" track that receives its MIDI input from my 88-key weighted board. Then I can have my piano VST receive its MIDI input from the "Piano Controller" track. Should I ever change the physical controller, I need only rewire the input to the "Piano Controller" track and not have to touch MIDI controller mappings in any other tracks.

 

And now I'm of the rails of this thread

 

-J

[Tusker]

I'm guessing that I'm not the only (former) EE on this forum, so you'll understand that Bidule does look like a reasonable course for me.

 

I am glad you got something good out of this, Jack. Please keep us posted on your discoveries. You aren't the first person to outgrow Mainstage. Your bidule techniques will help others.

 

My experiments with MainStage are hitting a wall, mainly due to CPU consumption and stability issues. ...

And now I'm of the rails of this thread

 

-J

 

... Wait here's another!

 

John thanks for sharing. It's the experiences I had which are similar to yours that persuaded me that Mainstage works as long as you stay within it's limitations sonically. The logic instruments are good. It's just a terrible host. I didn't "catch" the Ableton workflow much but it was stable, and didn't choke the Macbook Pro by the throat every time I loaded a 3rd party plug-in. (That was MS2) So with MS3 and a fresh computer, I am staying well within the electrified fence. However, like you, I might grow up some day, and need more space.

 

So your experiences with the transition will be great to hear.

[Markay]

Tusker, I had experience with several Windows hosts before I switched to MainStage 2.2.2 . I have not experienced any resource issues with MainStage 2.2.2 on my current machine. In my experience MainStage has been ultra reliable. I have 6 external synth packages, plus VB3 running in my current MainStage concerts.

 

System resources, not MainStage is the issue. With 4 instrument channels in one patch including 3 external synths my CPU usage is 50% and RAM 70%. This is an i5 machine with 4 gig of RAM running at 64 samples and MainStage running in 32 bit mode.

 

I am sure that this patch would choke an older machine with a Core Duo processor with 2 gig of RAM irrespective of whether an external audio interface was used.

 

If your current machine has an i5 or i7 processor and 4 or 8 gig of RAM why not try using your third party VST's, provided of course they are 64 bit.

[Ashville.Guru]

Guru's test is the real-world meaningful one. Tools and benchmarks can help us to find the sweet spot faster, but if you have A and B already selected to compare, might as well do the real-world comparison.

Thanks, Jeff, but I'd like to point out that what I've described is nothing but the DawBench standard...! The only difference is that the DB suite describes 3 standard sets of free plugins, important only if you're communicating the results.

 

Basically gives you a measurement of how much audio load your system as a whole can handle, before choking and sputtering. 'System as a whole' includes the hardware, interface, OS, host - the whole shebang. You can use it to check how much difference changing any component makes - interface, host, or OS. There's nothing Windows or Cubase-centric about the test itself; for e.g. the OP could use it to test whether Mainstage or Bidule can handle more audio load, on his system.

 

However, as I said above, these tests are not meaningful, unless the additional capacity is needed by the user.

The user in this case is the OP, and he most definitely needs the additional capacity...! Which is why I suggested it...

 

Why can't you use your browsing system for music? I've been doing it for years. if there were any issues with it I think I'd have found out about them by now.

Clearly, you haven't worked with Windoze systems...! :grin: We have dual boot Windows and Ubuntu machines at work, with both recieving similar usage, and its fascinating how consistently Windows performance goes down with time, thanks to creepware/malware and other undesirables. I've never seen exceptions to this rule. Ubuntu and OSX don't suffer from this problem, with rare exceptions.

 

My laptop is dual boot also, and I use Windoze only for live music. Ubuntu for everything else.

 

- Guru

[Bobadohshe]

I could not get Rax to get past the AU validation. Every AU failed...?

 

Remember if you're using Logic Pro 9 in 64 bit mode, that you must start the program up at least once in 32 bit mode for newly installed AUs to validate. That may or may not be your problem but it bears repeating.

[b3plyr]

I'm using MainStage 3 with a 2.53 GHz Intel Core 2 Duo, 4 gig of memory. I originally used MainStage 2 but like some of the instruments and features of MainStage 3 better. Controller is a VAX77, and I have no appreciable latency. To date, I have had no issues with system usage. Well, once I had a stuck note using VB3 with MainStage 2. So as they say, your milege may vary. My rig has been reduced from a couple of Nords and a Hammond to worse case the VAX77 and a Mojo, and MainStage. Many times I just use the VAX77 and MS3. Personally I think Apple has done a good job with MS3. Creating patches and concerts quickly and for any occasion is so easy. And to be able to tailor each channel strip is also great. So don't dump MainStage too quickly.

[zephonic]

There's nothing Windows or Cubase-centric about the test itself; for e.g. the OP could use it to test whether Mainstage or Bidule can handle more audio load, on his system.

 

The DAWbench can't be done for Mainstage, as it is really a DAW test. It plays back multiple audio channels that you have to un-mute until the audio starts to distort or drop out.

 

Mainstage is a performance-oriented VI host that can playback a few audio tracks, but it is not a full-fledged DAW. Apple has Logic for that.

 

[Phloid]

...So, bought the new Clav. Oh, it won't run on Kontakt Player 4, it needs 5. Oh, the Kontakt Player 5 won't run on OSX 10.6.8, it needs 10.7 and up....

FWIW, This isn't true. I have run K5 Player on 10.6.8. In many instances NI just changed the system requirements to reflect the lowest common denominator once they started building for features only available on the later OS, even if they did not receive any form of tweak.

[Ashville.Guru]

There's nothing Windows or Cubase-centric about the test itself; for e.g. the OP could use it to test whether Mainstage or Bidule can handle more audio load, on his system.

The DAWbench can't be done for Mainstage, as it is really a DAW test.

I should have been more clearer in my post, sorry. I was referring to the test I described earlier... which is, in essence, the same idea as the DawBench. Of course, the DawBench test suite downloadable from the website can't be used with Mainstage, or any generic host. Which is a bit unfortunate, yes, but the principle is much more useful than the actual suite itself.

 

You could use the same set of plugins used in the DawBench test, and see how many of each your system can run without choking or sputtering. In Mainstage, Forte or Bidule, that would mean enabling more and more 'layers' of plugins, in increments, till you hear pops and crackles. You could use a controller with heavy pedal use, or better still, have an external sequence running through it.

 

That way, you keep the core idea of DawBench test, but without the DAW-centric limitations.

 

In the OP's case, it's simply the most sensible way of determining whether (a) changing the host or (b) adding an interface will let him use the plugins he wants to use.

 

- Guru

[Tusker]

However, as I said above, these tests are not meaningful, unless the additional capacity is needed by the user.

The user in this case is the OP, and he most definitely needs the additional capacity...! Which is why I suggested it...

 

AG, no worries, but you misunderstand me. I was referring to your advice to me:

 

The best way to test it: with the headphone outs, layer more and more resource hungry plugins, one at a time, until you start getting pops and crackles. Then switch to the interface at the same buffer setting, and you should hear clean audio. Or you could just download the DawBench suite of tests itself.

 

Looking forward to your experiences!

 

- Guru

 

Our OP has a plan and is moving forward on it. I think your original advice (which I bravely attempt to summarize below) is certainly most beneficial to the OP:

 

It may be better to spend $500-$700 on an audio interface than to spend $2000-$3000 on a new Mac.

 

No quibbles there, and I am not the one who is knowledgeable enough to quibble.

 

Since our OP has moved on, would it be ok with everyone if I broadened our discussion with a couple of questions:

 

What are the mechanisms by which some interfaces relieve more processing burden from the computer's processors than others?

Additionally, is there still a systemic benefit to firewire (as suggested by the Dawbench graph) over USB? How can we tell if that systemic benefit applies across computer systems, OS's and hosts?

[Tusker]

If your current machine has an i5 or i7 processor and 4 or 8 gig of RAM why not try using your third party VST's, provided of course they are 64 bit.

 

Yes, I will try that eventually. I appreciate the encouragement. The current (air) machine is an I7 with 8 gigs. The MBP which was having difficulty was 4 megs of ram and I forget the processor but it was a 2009 or 2010.

 

One factor which makes me slow to flesh out a computer with the latest and greatest soft-synth presets is ...

 

at least for me ...the binding constraints in live performance sonics are related to room acoustics and speaker/amp quality (and musicians listening to each other) long before a best-of-breed physical model or sample.

 

I'll get there. Thanks.

[Ashville.Guru]

@Tusker: Sorry, my bad...! It's tricky to get the context right in discussions like these. In your case - agreed, the tests are merely academic.

 

I think your original advice (which I bravely attempt to summarize below) is certainly most beneficial to the OP:

 

It may be better to spend $500-$700 on an audio interface than to spend $2000-$3000 on a new Mac.

Happily, much less than that! :grin: By the DAWbench measurements, these are among the best performing audio interfaces out there:

• Presonus Firestudio Mobile: $249.99 (DAWbench rating: 7.09)
  
• Focusrite Saffire Pro : $299.99 (DAWbench rating: 7.26)

And it gets better. If you consider the results of the ADK Pro Audio CPU comparison, the OP might actually get much better performance by adding a $250 interface than from a $2000 Macbook upgrade...!

 

What are the mechanisms by which some interfaces relieve more processing burden from the computer's processors than others?

Additionally, is there still a systemic benefit to firewire (as suggested by the Dawbench graph) over USB? How can we tell if that systemic benefit applies across computer systems, OS's and hosts?

That's relatively well-known (compared to some of still unresolved issues surrounding computer audio!). Wikipedia has a technical comparison between FW and USB:

 

"The FireWire host interface supports DMA and memory-mapped devices, allowing data transfers to happen without loading the host CPU with interrupts and buffer-copy operations."

 

Basically, FW seems to do a better job of to take the load off the CPU while processing audio. Hence my assertion: FW interfaces are more useful for the older machines.

 

- Guru

 

[Reezekeys]

Our OP has a plan and is moving forward on it. I think your [Guru] original advice (which I bravely attempt to summarize below) is certainly most beneficial to the OP:

 

It may be better to spend $500-$700 on an audio interface than to spend $2000-$3000 on a new Mac.

 

No quibbles there, and I am not the one who is knowledgeable enough to quibble.

How about spending $0 (for now) to see if a different software host might solve his problems?

 

Since our OP has moved on, would it be ok with everyone if I broadened our discussion with a couple of questions:

 

What are the mechanisms by which some interfaces relieve more processing burden from the computer's processors than others?

Additionally, is there still a systemic benefit to firewire (as suggested by the Dawbench graph) over USB? How can we tell if that systemic benefit applies across computer systems, OS's and hosts?

I was under the impression that firewire is being abandoned. See the latest MacBook Pros - no firewire. Aside from that, I think Guru's description above nails it FW puts less demand on the OS, leaving more CPU for our plugins & such. The question I would have is, given that processors are getting faster, does this distinction make a real-world difference for musicians anymore? My guess would be maybe it does, if you're Hans Zimmer doing a movie score mockup with 20 instances of VSL. Otherwise, I'm not so sure.

[johnchop]

Yeah, Firewire is done in terms of current Mac hardware, but there's thunderbolt, and thunderbolt-to-firewire as a bridge technology.

 

Meanwhile, I'm quickly realizing that while Live is great for certain things, as purely a keyboard rig "brain" it is cumbersome when it comes to controlling external hardware. Despite having an "external instrument" feature, program and bank changes MUST be put into a clip. While you can store clips in your library--which sorta kinda allows for reuse--you still need to trigger the clip for the program change to occur. It's entirely possible I'm missing something. Maybe. I still need to try a MIDI program change plugin to see if that helps (meaning one more dependency :-/).

 

So, I haven't totally given up MS, nor have I bought into Live fully. Still putting everything through paces.

 

-J

[Ashville.Guru]

The question I would have is, given that processors are getting faster, does this distinction make a real-world difference for musicians anymore? My guess would be maybe it does, if you're Hans Zimmer doing a movie score mockup with 20 instances of VSL.

I'm no Hans Zimmer, but yes, the distinction is making a real-world difference to my live performance! I tend to drive my system really hard, as mentioned above: several instances of plugins of various kinds, background drone sequence, plus my vocals through effects plugins... at 64 samples . All of this would be unthinkable without my FW interface. And the measured difference between FW and USB is not small: FW can handle 3-4 ~1.5 times the load USB can. Knowing that I have that additional headroom is very assuring to me.

 

I've noticed that I'm much more freer and creative in layering sounds from plugins after getting the interface - perhaps because I don't have to worry about my system choking. Just checked - some of my patches have 8-10 plugins used simultaneously . I now realise I was earlier always holding back on patch design, for fear of taxing the system, and I suspect that's the case for many of us.

 

Yup, it's good to have all that headroom to play with...! :grin:

 

- Guru

[Reezekeys]

...the measured difference between FW and USB is not small: FW can handle 3-4 times the load USB can.

Wow that is a big difference. Could that 3-4x figure be due to the specific architecture of your system? Were you comparing USB 2 or 3, and was it to FW400 or 800? I wonder what that figure might be for a current top-of-the line MacBook Pro, comparing its built-in USB3 to Thunderbolt or FW-over-Thunderbolt. I presume we'll see some benchmarks at some point, if they haven't been done already.

[Ashville.Guru]

Wow that is a big difference...I presume we'll see some benchmarks at some point, if they haven't been done already.

I'm sorry, I goofed up on those numbers, which are way off! Mixup with another related comparison . Edited my previous post.

 

The best way to get a reasonable feel for the difference, is to compare FW and USB devices from the same manufacturer on the same system (caveats apply!). A quick glance at the DB results (@64 buffers, RXC only) shows us just how much additional plugins FW can handle:

• RME: ~20%
  
• Presonus: ~40%
  
• Focusrite: ~80%

This is all I had time for at the moment, but it's reasonable to estimate that FW handle 1.5 times more audio load than USB (unless you shell out the extra $1000 or so for RME... :wink:). That's still a significant increase.

 

When I have time, I'll try to arrive at a more robust estimate. Unfortunately, the numbers aren't available for Thunderbolt yet.

 

- Guru

[Reezekeys]

Apropos of all this due to a totally unexpected confluence of fate and luck, I am now the owner of a brand new Retina MacBook Pro. It's actually the lowest of the 15" configuration: 2 Ghz quad-core i7, 8GB ram, and a 256GB SSD. I just got it this morning, and just got back from my first gig using it, in which it performed without incident. Of course I'm expecting a performance boost over my previous laptop but I was already a bit surprised at what I saw tonight though I will not draw any conclusions or bore anyone here more than necessary until I have my regular setup running (all I used tonight was Kontakt and my acoustic piano sample).

[Ashville.Guru]

@Reezekeys - congrats.... Keep us posted.

 

To the OP (for what it's worth):

 

There aren't many live hosts that I haven't tried out on stage. Ableton Live, Bidule, Cantabile, Brainspawn Forte, Live Professor... and I've used Mainstage on a friend's laptop too a long while ago. Hell, for a while I was using Bidule as a plugin within Ableton Live...!

 

I've stayed longest with Bidule (2-3 years, I think), and I've explored its capabilities rather extensively. However, a while ago I switched from Bidule to Forte, and I felt my experiences *may be* relevant, since Forte and Mainstage have the same architecture, and same strengths.

 

The geek in me loves Bidule. It's a programmer's wet dream, actually. In terms of routing flexibility, intuitive interface, nothing else even comes close.

 

The downside? Bidule isn't designed optimally for (i) building sets of several complex patches (ii) recalling those patches quickly through hardware or (iii) rearranging those patches into several songs/setlists. Don't get me wrong, it's not that these cannot be done with Bidule. It's just that the quantum of effort involved grows exponentially with the complexity and number of patches.

 

Say you have just two plugins, Kontakt and VB3.Now consider this scenario, for a given setlist:

Song 1: you need preset A on Kontakt, B on VB3,

Song 2: you need preset C on Kontakt, D on VB3

Song 3: you need preset A on Kontakt, D on VB3

 

Even if your controller is sending PC messages, you'd have to carefully arrange the preset banks within each plugin to form a co-ordinated sequence (A-C-C on Kontakt, B-D-D on VB3), so that a single PC messages calls up the right combination. Now imagine adding a more plugins/effects. Or splits and layers. Or suddenly needing to change the sequence of songs in your setlist.

 

Of course, there are several workarounds, but they're all quite messy and inelegant. You could control presets independently in each plugin through knobs/buttons on your controller. But that means, the more number of plugins involved, the more button pushes/tweaks to arrive at that patch. Mid song, that's not an option. Or you could rely on a very powerful multizone controller which can send channel-specific PC/CC messages on preset recall, etc. But even that isn't ideal - discussed here.

 

If you only use a handful of bread-and-butter sounds, and your needs are more or less static, and you don't need too many patches with complex splits/layers - then Bidule is ideal for you.

 

The live performer in me loves Forte. However complex the patch, it takes only a click to save it, arrange all the patches in a setlist, save multiple setlists, and recall the patch with a single PC message. The quantum of effort is independant of the complexity and number of patches.

 

Nowadays, my patches involve 4-5 zones across 2 controllers, several layers, 8-10 plugins, each zone/layer with independent velocity curves and FX routing. I would have *cringed* to setup more than half a dozen such patches on Bidule. In Forte, I've set up two dozen in the past week itself.

 

Mainstage has a similar architecture to Forte. Scenes, Setlists and Racks in Forte are called Patches, Sets and Concerts in Mainstage (Forte has another tier in the hierarchy, called 'Song', very useful). Plus there are other little things like holding down notes while changing patches, etc.

 

In summary: far from being 'training wheels', Mainstage might just be the most professional host for Macs out there.

 

- Guru

 

P.S. Apologies for the long post...

[Reezekeys]

Guru, it's interesting to read your take on this since I count myself as a total Bidule geek. I agree that managing the song-to-song changes in patches and efx, routing, etc. in Bidule can be a bear if there are lots of plugins to manage. There is no doubt that the immense flexibility Bidule offers comes at a price (complexity), and that not everybody may want to put the effort in. Over the past few years though, I've come up with ways to streamline and simplify the process of creating complex meta-patches that, with one PC change from a controller, change many plugins' patches, along with routing changes, efx, etc. Admittedly, it's still not as simple to do as on an app designed specifically for that job, and it's definitely not as pretty!

 

Forte does sound interesting but a quick check shows it as Windows-only. The OP uses a Mac. Of course I know a Mac can run Windows now, so technically it's possible, but my guess is that most Mac users, unless they are true dual platform people, would probably want to stay in the Mac world for something like this. My original recommendation to the OP to use Bidule wasn't just as a geek standing up for his platform of choice; he was experiencing issues with MainStage running out of cpu juice. My experience with Bidule on my now-former MacBook Pro, running a fairly involved and demanding setup on an almost seven year old computer (at a relatively low 128 buffer) suggested that Bidule might give him the headroom he needs without investing in a new computer.

[Ashville.Guru]

Forte does sound interesting but a quick check shows it as Windows-only. The OP uses a Mac.

Can't blame you for missing it, since it's buried in that long-ass post of mine... but I'm not recommending Forte to the OP! I'm recommending he stick with Mainstage, since it has similar advantages that Forte has. And overcome performance issues with an interface. If he hasn't made up his mind already, of course...

 

My post was a heads-up, that switching the host might just land him with a different set of worries.

 

But from the anecdotal experiences here it appears that Mainstage is a resource hog; and I was just vaguely reminded of a post I made here a couple of years ago. I dug it up, and it's amusing to read it in the context of this thread. And the title of that thread? "Best host for VSTs on a Mac".

 

:grin:

 

- Guru

 

 

[Markay]

But I understand that Mainstage is a resource hog

Well MainStage may indeed be a resource hog, but it does include a CPU and memory meter which is displayed at all times in Edit mode, so it is loud and proud in reporting the resources it is consuming.

 

As I noted earlier MainStage on my machine reports max 50% CPU usage and a pretty steady 50% to 60% memory usage.

 

That doesn't qualify as hog in my book.

 

On another note it is interesting that none of the interface manufacturers have released any USB 3 interfaces. USB 3 ports have been around for a while and offer significant speed improvements.

 

 

[Reezekeys]

Forte does sound interesting but a quick check shows it as Windows-only. The OP uses a Mac.

Can't blame you for missing it, since it's buried in that long-ass post of mine... but I'm not recommending Forte to the OP! I'm recommending he stick with Mainstage, since it has similar advantages that Forte has. And overcome performance issues with an interface. If he hasn't made up his mind already, of course...

Yes of course, I have to read posts after I drink my first morning cup of coffee!

 

I think he's gonna try Bidule first, since the demo will let him spend $0 (once it's available) vs. >$0 for a new interface (maybe a lot more than $0 if we're talking a FW interface, which would presumably give him the cpu breathing room for his MainStage setup).

 

(PS - my new Mac autocorrects MainStage to capitalise the "S." Those crafty Apple engineers...)