Ableton Forum

Be careful,

You have the most powerful mac mini. The 2014 mac mini's are around 40% less powerful than the late 2012.

lapieuvre wrote:Be careful,

You have the most powerful mac mini. The 2014 mac mini's are around 40% less powerful than the late 2012.

I hope they go back Quad Core on the next mini. The 2014 One was a huge step back. I know somebody with one and It was struggling when I was using some VSTis although it worked decently for recording audio. They basically made it more like a MacBook Air which isn’t much.


The 2012 One was a pretty nice box when fully maxed out. CPU use was pretty decent for VSTis. Don’t know how it compared to the MacBook Pros but I did a lot of projects with it relatively smoothly.

And what about the dual core Intel Core i7 3,0 GHz coming with currently availabla on official apple distribution "new" Mac Minis? Is it going to be enough for live performance with some backing tracks, synths played live and some live instruments?

And by the way, any solutions for an on stage monitor for mac mini, like on top of the amplifier or the like?

Unless the "mini" range isn't going all but dead, it's perfectly logical if Apple release new models in 2018. I'd rather wait.

WWDC June 4...

Live is a pig, that's all there is to it. Live responds well to raising the buffer to get more CPU like all DAWs do.
Like people have mentioned flatten parts when you're done working them, raise the buffer when you're done playing software instruments live. I can get by on 128 or even 256 without too much issue. A ten foot guitar cord adds roughly 10ms latency for comparison so don't think you NEED to play at 64 bit buffers.

Live responds best to faster CPUs, I don't think Live is that great at using 6-10 core systems, not in my experience, it's not bad, but faster initial CPUs seem to work better, at least for me.

Geekbench is your friend, I've found in general their scores add up to better performance in DAWs etc. You will be surprised to learn how some older laptops etc. fair pretty well.

Machinesworking wrote: Live responds best to faster CPUs, I don't think Live is that great at using 6-10 core systems, not in my experience, it's not bad, but faster initial CPUs seem to work better, at least for me.

it comes down to the weakest link in the chain. if you have a machine with 1000 processors, even if only 1 of those 1000 can't keep up the work thrown at it, then you will get an audio glitch...

Machinesworking wrote: Live responds best to faster CPUs, I don't think Live is that great at using 6-10 core systems, not in my experience, it's not bad, but faster initial CPUs seem to work better, at least for me.

Unless you made sure to not overload one core, then how valuable is this observation? What does "faster initial CPUs" even mean here? Because if you have faster, but fewer cores, then those cores can also be overloaded when you stack tracks on them. And fewer cores and many tracks needed to distributed over these can manifest overload and the resulting glitches quite fast.

The only valid way to make a surefire test comparison is to find very similar, if not identical computers, with different core counts and where single core work capability are about the same. Then use more tracks than cores until they start to glitch.

More cores does add some overhead for sure, but I think this is not likely the reason behind your experience. More likely is the fact that CPUs with more cores tend to be a little slower per core and thus faster to overload.

Stromkraft wrote:

Machinesworking wrote: Live responds best to faster CPUs, I don't think Live is that great at using 6-10 core systems, not in my experience, it's not bad, but faster initial CPUs seem to work better, at least for me.

Unless you made sure to not overload one core, then how valuable is this observation? What does "faster initial CPUs" even mean here? Because if you have faster, but fewer cores, then those cores can also be overloaded when you stack tracks on them. And fewer cores and many tracks needed to distributed over these can manifest overload and the resulting glitches quite fast.

The only valid way to make a surefire test comparison is to find very similar, if not identical computers, with different core counts and where single core work capability are about the same. Then use more tracks than cores until they start to glitch.

More cores does add some overhead for sure, but I think this is not likely the reason behind your experience. More likely is the fact that CPUs with more cores tend to be a little slower per core and thus faster to overload.

you have a special skill at creating arguments out of nothing.

a single core that is more powerful can cope with a single track that is more heavily loaded. obviously there are no hard and fast rules here, it will depend on how you use Live.

Stromkraft wrote:

Machinesworking wrote: Live responds best to faster CPUs, I don't think Live is that great at using 6-10 core systems, not in my experience, it's not bad, but faster initial CPUs seem to work better, at least for me.

Unless you made sure to not overload one core, then how valuable is this observation? What does "faster initial CPUs" even mean here? Because if you have faster, but fewer cores, then those cores can also be overloaded when you stack tracks on them. And fewer cores and many tracks needed to distributed over these can manifest overload and the resulting glitches quite fast.

The only valid way to make a surefire test comparison is to find very similar, if not identical computers, with different core counts and where single core work capability are about the same. Then use more tracks than cores until they start to glitch.

More cores does add some overhead for sure, but I think this is not likely the reason behind your experience. More likely is the fact that CPUs with more cores tend to be a little slower per core and thus faster to overload.

In my experience Live is not as good at loading processes to multiple CPUs as other DAWs are on my system. I have a 12 core 3.34ghz Xeon Mac Pro and in my experience Live doesn't behave that much better on it than on the 4 core 2.7ghz i7 MacBook Pro here.

Live literally comes in dead last CPU wise compared to Logic, DP, and Reaper. Getting 70% of the track counts of DP and Logic, and damn near 60% of the track counts of Reaper. I'm out of town at the moment, but I'm more than willing to do some tests when I get back to Seattle. I could be wrong and Live could be just generally CPU hungry but my experience has been that Live is more likely to fall apart CPU wise, not spread CPU efficiently among cores etc. I have not tested or experienced this with v10 though so...

Machinesworking wrote: In my experience Live is not as good at loading processes to multiple CPUs as other DAWs are on my system.

OK, but aren't these DAWs generally more powerful in any fair comparison against Live? And even if you do find they are, you still feel they are even more powerful when it comes to utilizing multiple cores compared to Live? If so, that's indeed very interesting. Thanks for sharing.

Machinesworking wrote: I'm out of town at the moment, but I'm more than willing to do some tests when I get back to Seattle. I could be wrong and Live could be just generally CPU hungry but my experience has been that Live is more likely to fall apart CPU wise, not spread CPU efficiently among cores etc. I have not tested or experienced this with v10 though so..

Please do and share the tests with us if possible. I can't compare with more than 4 cores myself at the moment, but I'd be willing to test Reaper and Pro Tools First on my quadocres and share the results. I have both Live 9 and Live 10.

Stromkraft wrote: Please do and share the tests with us if possible. I can't compare with more than 4 cores myself at the moment, but I'd be willing to test Reaper and Pro Tools First on my quadocres and share the results. I have both Live 9 and Live 10.

I have Live 9, 10, Reaper, DP9, and logic X for comparison. It's a lot harder to tell how much CPU issues are coming from a DAWs bad multi core capabilities rather than general CPU issues for sure. With general CPU use 'to failure' tests are a great indicator. I've given up on proselytizing on reasons why the Live 9 CPU test thread is not accurate, and I do share one distinct thing in common with you in that I would rather have proof than someones gut feeling. Trying to figure out a good way to perform this sort of test is not easy though. In the past with DAWs like Logic 7 for instance which had separate CPU meters for each core, you could see Logic choke out a single core, stopping and starting the sequence would generally wake up whatever code was dropping the ball there and Logic would evenly distribute the load.
I think looking around online for a core meter for OSX, then adding as many tracks as there are cores, with a virtual instrument playing a simple MIDI note line, then loading them with FX is a good start. In a decently coded for multi core DAW you should see a fairly even distribution of resources, unlike older versions of Logic or other DAWs from that era.

I will say this though, I wish somehow Ableton would hire Justin Frankel or be bought out by him. Barring it's lack of anything resembling UI standards Reaper is as leanly coded as they come, in CPU tests it soundly destroys Live, just trounces it, and it has some of Lives glitch free operation, doesn't choke if you load a plug in with the sequence running, has Session View type "clip" firing in quantized marker setting etc. Live as a lean machine with as many customizable options as Reaper has and with VCA faders, MPE, spectral editing, VST3, 32 bit bridge, etc. etc.

Machinesworking wrote:

Stromkraft wrote: I think looking around online for a core meter for OSX, then adding as many tracks as there are cores, with a virtual instrument playing a simple MIDI note line, then loading them with FX is a good start.

the Activity Monitor has two core meter bridges—look in the Window menu...

I'm in a real confusion here. For the very moment I'm buying an old Macbook Pro late 2011 (i5 -2435M, 2,4GHz, 16gb ram) because my laptop literary is falling apart. But I'm about to get a subvention for my business activity and I'm planing to buy a new laptop from authorised seller.

I want to spend the money wisely, so pending like 3k$ on a new Macbook which will only have a two-core CPU is, I think, not the way to go, I want to spend less on a laptop, because I have to buy other gear too.
I don't know what to do, because I need a realiable machine for live performance (Ableton Live), but I also work with Logic Pro X so...MacOs.

Should I just have two machines, the 2011 macbook for Logic, and get the latest i7 4-cores windows laptop for Ableton? Which CPU would you recommend then? I know there's millions of laptop models, so I'm not asking about a specific model, just the CPU, more GHz or more cores... or ... Perhaps those two core i7 in new MBP's are ok, they have this hyper threatening so four virtual cores in total?

I will wait for the Apple conference, at least out of curiosity, that's just a few days now.