New Logic Multitrack Benchmark Test

[paulcristo]

...there is a HUGE GAP between speculation and fact... just worth remembering...

My thoughts aren't speculative, they're observational. And you didn't answer my question regarding Apple's incentive to improve Logic.

[fisherking]

everything is working well for me here. either way, there is a HUGE GAP between speculation and fact... just worth remembering...

 

I agree - but there's a number of facts that are pretty good to make one thing there'd be a lack of that incentiveness (is that a word?). And that's got nothing to do with Logic working well or not.

 

facts? and different from the 'facts' of bugs in pro tools, cubase, etc? and how is it different from logic "working well or not"? even if you feel you have these facts, you're still speculating about the logic team...

[fisherking]

...there is a HUGE GAP between speculation and fact... just worth remembering...

My thoughts aren't speculative, they're observational. And you didn't answer my question regarding Apple's incentive to improve Logic.

 

yes, they're YOUR observations... which, again, are not, by default, facts. and what's the question exactly? perhaps worth asking the logic development team (ie, instead of, say, the ios team, or apple in general).

[Sascha Franck]

facts? and different from the 'facts' of bugs in pro tools, cubase, etc? and how is it different from logic "working well or not"? even if you feel you have these facts, you're still speculating about the logic team...

 

First off: I'm not speculating about the Logic team. But I *do* measure them by certain undeniable facts. Such as "Has the EXS seen an update seen a noteworthy update during the last decade?" and the (factual) answer clearly being a "No". I could obviously come up with several more examples.

So, while I obviously have no idea whether this is something the Logic team is interested in or not, for me as a user it's clearly "they're not doing anything regarding issue XYZ". And even if they're working hard on the very issue behind the scenes, that doesn't make any difference for a user. Because fact is that the EXS is still looking pretty much identical as it did around 10 years ago. And in software land, 10 years of no improvements equals something being rendered abandonware, like it or don't.

And that's pretty much all there is about it. And as said, for a mere user it's completely irrelevant whether the team is still working hard on the EXS. There's also no need to speculate about whatever they're working on, feel like doing (or not doing) and what not. It's completely irrelevant. Factual evidence says "it seems to be like this". Nothing more, nothing less.

[fisherking]

facts? and different from the 'facts' of bugs in pro tools, cubase, etc? and how is it different from logic "working well or not"? even if you feel you have these facts, you're still speculating about the logic team...

 

First off: I'm not speculating about the Logic team. But I *do* measure them by certain undeniable facts. Such as "Has the EXS seen an update seen a noteworthy update during the last decade?" and the (factual) answer clearly being a "No". I could obviously come up with several more examples.

So, while I obviously have no idea whether this is something the Logic team is interested in or not, for me as a user it's clearly "they're not doing anything regarding issue XYZ". And even if they're working hard on the very issue behind the scenes, that doesn't make any difference for a user. Because fact is that the EXS is still looking pretty much identical as it did around 10 years ago. And in software land, 10 years of no improvements equals something being rendered abandonware, like it or don't.

And that's pretty much all there is about it. And as said, for a mere user it's completely irrelevant whether the team is still working hard on the EXS. There's also no need to speculate about whatever they're working on, feel like doing (or not doing) and what not. It's completely irrelevant. Factual evidence says "it seems to be like this". Nothing more, nothing less.

 

yes, some things need updating. but nothing here is impeding my workflow (i actually use new sonic arts nuance as my main sampler anyway). does the EXS need an update? sure. but perhaps your needs are not considered priority with the logic team? and since we don't know what goes on there (am picturing a dr. frankenstein lab with maniacs in lab coats)... we can only speculate; that doesn't alter the facts you mention, but it doesn't mean that their priorities are somehow 'wrong'... they're just not in sync with your priorities, something you mention about every 15 minutes on this forum.....

[skijumptoes]

The crazy thing about EXS is why did thy never utilise redmatica to advance it, they actually ended up stripping away functionality from users with their buy out as existing RM products were left to expire.

 

I find that totally bizarre, to think that Keymap Pro could’ve been an integral part of Logic today!

 

But there’s plenty of cheap third party solutions so can’t get too caught up it, I’d rather Logic team were weak at something that we can address via third party tools, than fixed components which are a REAL obstacle.

[Sascha Franck]

they're just not in sync with your priorities, something you mention about every 15 minutes on this forum.....

 

Fwiw, it's not my priorities. I'm using Kontakt and Battery for my sampling needs. I would like to use the EXS more often because it's so efficient, but in the end, I don't care all too much. Might even be better to stick with 3rd party tools as I'm not all too sure about my future in Apple land.

So it's just been one example.

Anywhere, there's quite some patterns among Logic users about what they'd like to see adressed. That also includes the EXS and Ultrabeat - but several other things, too. Some of them being very longstanding things that other companies simply seem to adress in a way more focused manner.

[Sascha Franck]

I find that totally bizarre, to think that Keymap Pro could’ve been an integral part of Logic today!

 

That is in fact quite bizarre. As it seems, the only folks profiting from the Redmatica deal are Mainstage users as it apparently has RMs autosampling technique built in.

[fisherking]

they're just not in sync with your priorities, something you mention about every 15 minutes on this forum.....

 

Fwiw, it's not my priorities. I'm using Kontakt and Battery for my sampling needs. I would like to use the EXS more often because it's so efficient, but in the end, I don't care all too much. Might even be better to stick with 3rd party tools as I'm not all too sure about my future in Apple land.

So it's just been one example.

Anywhere, there's quite some patterns among Logic users about what they'd like to see adressed. That also includes the EXS and Ultrabeat - but several other things, too. Some of them being very longstanding things that other companies simply seem to adress in a way more focused manner.

 

i hear you, really. but again, the logic team probably has it's priorities, and i, for one, am okay with that; everything i use in logic (and, full disclosure, 97% of my synths and effects are 3rd-party) works. also, my pro tools-using collabs have issues sometimes, as does my (one!) friend who uses cubase.

 

which developers specifically 'address in a more focused manner' their issues? seems all DAWS (as with all software) have issues, good features, missing features... etc.

[Sascha Franck]

which developers specifically 'address in a more focused manner' their issues? seems all DAWS (as with all software) have issues, good features, missing features... etc.

 

I am absolutely aware that there's no software as complexed as a sequencer coming without bugs and other issues. Yet, as an example, last time I had a look at Cubase 10s feature set, large portions of it looked like a copy of my Logic wishlist, some of those wishes reaching back a *long* time - and again, while I'm saying "my wishlist", I am absolutely sure that plenty of these things are very high on others lists, too.

 

Oh yeah, I hear you - different priorities and all that. But then: What exactly *are* the priorities of the Logic team? To me it doesn't look like a clear path or whatever. I mean, on one side, they're coming up with, say, those track stacks. Which are just excellent. Or with Drummer. Another excellent thing (could do with some improvements, but it's really pretty well thought out). And on the other hand there's Drum Machine Designer, hopelessly hiding their inability/unwillingness to fix the source thing, namely Ultrabeat. With DKD being a bit better but still not exactly great at all, let alone the Studio Horns/Strings. For me, there's quite some contradiction. On one hand, they're adding features asked for by professionals (and they're doing that at least sort of well), on the other hand there's stuff that looks as if it'd be programmed by amateur programmers (just imagine purchasing a 3rd party plugin like DMD and it looked like it actually does under the hood...). Really, to me there seems to be a lack of a strong focus.

[fisherking]

which developers specifically 'address in a more focused manner' their issues? seems all DAWS (as with all software) have issues, good features, missing features... etc.

 

I am absolutely aware that there's no software as complexed as a sequencer coming without bugs and other issues. Yet, as an example, last time I had a look at Cubase 10s feature set, large portions of it looked like a copy of my Logic wishlist, some of those wishes reaching back a *long* time - and again, while I'm saying "my wishlist", I am absolutely sure that plenty of these things are very high on others lists, too.

 

Oh yeah, I hear you - different priorities and all that. But then: What exactly *are* the priorities of the Logic team? To me it doesn't look like a clear path or whatever. I mean, on one side, they're coming up with, say, those track stacks. Which are just excellent. Or with Drummer. Another excellent thing (could do with some improvements, but it's really pretty well thought out). And on the other hand there's Drum Machine Designer, hopelessly hiding their inability/unwillingness to fix the source thing, namely Ultrabeat. With DKD being a bit better but still not exactly great at all, let alone the Studio Horns/Strings. For me, there's quite some contradiction. On one hand, they're adding features asked for by professionals (and they're doing that at least sort of well), on the other hand there's stuff that looks as if it'd be programmed by amateur programmers (just imagine purchasing a 3rd party plugin like DMD and it looked like it actually does under the hood...). Really, to me there seems to be a lack of a strong focus.

 

perhaps, because there are 3rd-party options (so, 'pros' don't have to use the included 'strings' (shudder), or ultrabeat, or... whatever.... the focus is elsewhere. or not. again, without knowing what the logic team is about, we can ONLY speculate; no harm in that, as long as we're clear that we're making assumptions... and not stating 'facts'.

 

anyway, will bow out of this 'discussion'. but why not perhaps move to cubase? just seems you have a lot of issues with logic (but... whatever). have fun. make some great music. the rest is just details..........

[Sascha Franck]

Fwiw (before I bow out of this discussion, too, at least at this particular place), no I'm not having a lot of issues with Logic. No more and no less than what I'd likely have to go through with anything else at least.

It's just that these days the combination of Logics state of development and Apples business tactics isn't exactly floating my boat at all, to put it conservatively. If one of the two would make me smile, I'd happily deal with the other. But that's just not happening, at least not right now. We'll see.

[Sascha Franck]

Btw, out of fun I decided to run a few "sort of" benchmarks on my own. Randomized things, as said, this was just for fun:

 

- Zebra 2, 85 tracks playing a 6-voice chord using factory preset "HS Esoteric Garden".

 

- Synth 1, "Initial Sound" patch. 10-voice chord, 521 tracks. There's your reason for the higher track count already...

 

-TAL Noize Maker, preset "KB Pop Pluck TUC", followed by TAL Chorus LX (default patch) and TAL Reverb III (default patch), 6-voice chord, 72 tracks.

 

- Kontakt 6, factory library, layer of "Fudge Pad" and "String Ensemble", 6-voice chord, 85 tracks.

 

- Massive, "Clean Sweep Pad", 6-voice chord, polyphony set to 32, 65 tracks.

 

And some Logic stuff:

 

- ES2, patch "Purity" followed by Space Designer loaded with a 3.4sec Stereo-IR, 6-voice chord, 138 Tracks, hence 828 synth voices and 138 SD instances. Without SD the track count would go up to something around 250.

 

- EXS24, selfmade patch sampled in tritones, filter off, amp envelope release at 0 (helps with the cycle jumps), "virtual memory" off (highly recommended in case you got plenty of RAM and something such as Kontakt for larger stuff), 12-voice chord, 504 tracks. Right, that's 6048 sampler voices.

 

Everything done at 44.1 and 64 samples buffer size. Everything supposed to play multiple 4 bar cycles without problems.

 

Not bad for a 9 year old machine, huh?

[skijumptoes]

Not bad for a 9 year old machine, huh?

..or for a DAW that you paid for 6 years ago!

[Sascha Franck]

..or for a DAW that you paid for 6 years ago!

 

Well - LP9 would possibly deliver the same results.

But anyway, regardless of my constant (yet IMO reasoned) anger with Apple/Logic, you're of course right. It's an incredibly efficient DAW and so far, the two Macs I owned have been working pretty well for what has to be considered a very decent lifespan.

[PangHu]

Test on my new 2019 MacBook pro 15'' with i7, the spec of this 9gen i7 CPU is similar to the 8gen i9 last year and the result is basically the same.

Mine got a result of 67 tracks at 128 I/O buffer size, 69 tracks at 64 I/O buffer size. The processor buffer is all medium.

Compared to the 2019 i9 version of MacBook pro which reaches 88 tracks, I strongly regret that I should have bought an i9 version of MacBook.

It looks like logic is a DAW that demands multicores, which in windows is not the case, I believe Cubase and Pro tools are all demanding single-core performance.

If I have seen this post before my purchase, I would have definitely bought the i9 version.

[skijumptoes]

You will find that the i9 will perform better in a short benchmark test - it's basically how Apple price their goods. But it's not a valid indicator of how it will perform in the real world.

 

When running for a more sustained period the i9 isn't going to be so much different performance in use to your i7, possibly worse in some cases as the i7 is better balanced to utilise turbo boost where the i9 is a powerful beast in a small space - hence thermal build up and the systems reaction is to throttle the CPU.

 

Me personally i'd have the i7 everytime over the i9. If you're not hitting the capabilities of the i7 then don't concern yourself about it, these benchmarks are just a quick indicator. How it runs with a loaded project in the summer months is most valuable.

[Ploki]

the i9 has 8-cores, his i7 has 6.

 

that does make a difference.

 

However, i7 Mini 6-core is on par with the i9 8-core in the macbook, because you know, they're hot as f*%@.

[Dedtime]

I got up to 93 tracks looping for a couple cycles. I have the newest 15" Macbook pro with the 2.4 ghz 8 core i9 and 32gb of ram. At 94 tracks it stopped around bar 9. My i/o buffer is at 128. With the buffer at 1024 I had 100 tracks playing. My cpu hates me now.

[nightjar]

With my 2018 6-core i9 MBP using Logic Pro 10.4.6

 

At 44.1 and 128 Buffer, built in audio

 

I'm getting 76 tracks.

[TTOZ]

Getting a 16" MBP here so hopefully everything will be ok with Catalina and I'll give it a go as soon as it comes.. Will be the 2.4ghz 8 core, 32GB ram. My 2014 macbook pro often just locks up and 2 minutes later is suddenly at the login screen LOL.. sometimes I'm watching a video and it's all corrupted, it's just screwed.. I got the 15" but the 16" came out a week later so I returned it.. Anyway very curious to see if the "improved thermals" make a difference for Logic.

[kingminotaur]

With "custom built" desktop hex core 8770 water cooled cost $1500 16gb men crucial 1TB ssd built 2 years ago getting about 300 tracks

without error. activity monitor showing about 700%

 

Geekbench 4 gives about 5000 single core 25000 multi core.

 

edit: it cost about $300 in consultant costs to build.

[Ploki]

jesus thats insane

[Sascha Franck]

That's a machine people would expect from Apple. But not at 4x the price.

[logicben]

15 tracks

 

1024 samples buffer size

Logic 10.4.8, MacOs 10.14.6

MacBook Pro (mid 2012), i7, 2.9 GHz, 2-core, 16GB Ram

[TTOZ]

With "custom built" desktop hex core 8770 water cooled cost $1500 16gb men crucial 1TB ssd built 2 years ago getting about 300 tracks

without error. activity monitor showing about 700%

 

Geekbench 4 gives about 5000 single core 25000 multi core.

 

edit: it cost about $300 in consultant costs to build.

 

my iMac pro gets 5300 GB4 single core and 30,400 multi.

 

it gets 155 in this test.

 

You are doing something wrong and not unmuting tracks.. i bet you you just duplicated to 300 tracks but only the default amount are playing.

 

What you are saying is literally impossible.. there is no way an 8700 can give 300 tracks in this test.

 

a 9900K gives a little more than mine, and that's all 8 cores at 5GHZ vs mine at 4GHZ. Around 170 from memory.

 

A 9900K is way more powerful than an 8700. In fact a 9900K is the best consumer chip intel make. over 6000 GB4 single core and 32000+ in the multi.

 

an 8700 giving almost TWICE that of a 9900K? Yeah, not likely. I flat out don't believe you to put it bluntly but will give you the benefit of the doubt that you are testing incorrectly. OR, you could prove it with a simple video from your cell phone (so there is no overhead with a screen recorder). LOL.

[TTOZ]

That's a machine people would expect from Apple. But not at 4x the price.

That result is fake.

 

This is like barefeats saying the iMac pro 10 core gets 508 vs 251 on the 8 core (with the previous benchmark).

 

Incorrect testing.

[Ploki]

true.

[Sascha Franck]

That's a machine people would expect from Apple. But not at 4x the price.

That result is fake.

 

This is like barefeats saying the iMac pro 10 core gets 508 vs 251 on the 8 core (with the previous benchmark).

 

Incorrect testing.

 

Yeah, you're probably right.

[angelonyc]

I appreciate these benchmarks. however. I can cripple my 2015 - 6 core mac pro with 8 instances of a U-He virtual instrument, with automation on it. I can get 100 or more tracks with less CPU demanding virtual instruments, and audio tracks. Some UVI instruments also can get pretty demanding.

 

Turning virtual instruments to audio, and saving channel strips, then nulling out, I can get over 100 tracks going. A lot depends on which virtual instruments. I have never been able to get an Apple salesman to certify how many tracks a machine can run, because of the extremely wide variation of CPU demanded by different 3rd party VI's.

 

And rest assured, as soon as the 2019 gets more units sold, 3rd party plug-ins will get more cpu demanding. I cringe at how many processes are running in the background, with extremely cryptic names, so you can never figure out what they are.

 

I once had my machine, completely bogging down a lot. I sent some core dumps to Apple. Came back with the message to cut out some processes, one was Native Instruments, access.