Gearslutz

Gearslutz (https://www.gearslutz.com/board/)
-   Apple Logic Pro (https://www.gearslutz.com/board/apple-logic-pro/)
-   -   Anyone interested in my new benchmark to replace the dated Evan's test? (https://www.gearslutz.com/board/apple-logic-pro/1228406-anyone-interested-my-new-benchmark-replace-dated-evans-test.html)

TNM 25th August 2018 09:31 AM

Anyone interested in my new benchmark to replace the dated Evan's test?
 
NewLogicBenchmarkTest.zip - Google Drive

Hi...

I have created a new benchmark with scultpure at 96K, playing 4 notes in sequence over 16 bars looped..

Very pleasant sound, just the default sculpture loads with, all volumes have been adjusted sensibly, and I have put a Logic limiter on the master bus for safety.

I have included one audio track, which should be highlighted when test is done, so no cores are "live".

This was born out frustration of people over at logic pro help forums sincerely believing that the 10 core imac pro got over 500 tracks in evan test vs 251 for the 8 core..

Of course this is impossible, as many here have stated, since Logic only allows 255 unique instrument tracks anyway.

Sculpture is a lot more CPU heavy, so heavy in fact, that with just 40 tracks of this test at 96K, my quad core 2.8ghz i7 macbook pro is maxed out.

Which means that even an 18 core imac pro won't be able to do 255.

If y'all think 96k is not the way to go, i can make it a 4 note chord at 44K instead of a single note ascending.

Just let me know.

It's time to put the dated test to rest, thank Evan for his hard work..

I am happy to pass on all responsibilities of the test to someone else if they like, this is not about it being my test.. You can put your name on it and say it's yours completely.

This is simply about having a realistic test to use on modern machines, and is made in Logic 10.4.1 which i believe anyone on Logic X will be using.

(Correct me if I am wrong, but maybe Logic 9 can open this as it does have sculpture? I am not sure on that as i can't install Logic 9 right now as i have the physical media version of 9.0.0 and don't even have a DVD drive anymore).

There is just one caveat.. sculpture idle actually uses CPU..

so say I have 64 tracks of sculpture loaded but only one with any midi on it, all cores get hammered.

So there is a tiny bit of work involved on your part.

I will be happy to provide tests with 32, 64, 96 and 128 tracks with sculpture and midi pre loaded.

You will just need to deleted tracks, NOT mute, till you get the right max number for your machine,

OR, right click duplicate the last track one by one if you need to add more (and copy the midi to it, which is why i have made it easy in one long 16 bar joined file for each track).

If i should proceed with this, let me know, i'll be testing my imac pro 8 core this week as i don't veen have logic installed on it right now, and am physically going to an apple store to test the ten core.. yes i will admit, the guy got me that irritated but it's more about people believing that the imac pro 10 core has more than double the power of the 8 core with Logic, which is absurd.
This will finally put that to rest, and i am happy if anyone knows anyone at barefeats to submit the new test to them too.

Let me know, and i'll get it all uploaded ASAP if there is interest.

Cheers.

PS let me know 44 or 96K and i'll tally the votes and choose the most popular.

I believe all macs for years have 96k internal sound, so personally i think setting to internal sound and 96k is the fairest easiest way to compare all macs.

NewLogicBenchmarkTest.zip - Google Drive

TNM 25th August 2018 10:46 AM

Ok I have made some changes, I am adding a low cut (logic EQ) and multicomp(logic multipressor) to each track, and have made it a 4 note chord (musically in key don't worry LOL).

So now even at 44.1 it's an intense test..

and I have made it 128 sculpture tracks with mute tool active on project load and the tracks not playing midi are now using reasonable resources.

All good.

They are all numbered.. the problem is, the default sculpture patch has a nasty resonance at 44K that only disappears at 88K and higher. I have notched it out as best as I can with the multiprocessor. You see, if I change the patch, logic changes the whole track name and the numbers all get screwed up.. Right now, the numbers make perfect sense and no one has to count their tracks, they can just see the number playing

OR, I can just set sculpture to higher res quality.

The problem is, this is a new feature in 10.4, so not everyone will have it.

so i need to hear from people whereas they think that's ok, or leave the default sound..

so.. as it stands, 4 note chord, 44K, default sculpture patch...EQ, and multipressor on each track.

Edit, i have edited the default sculpture patch to remove the resonance...

Still an EQ and multipressor on each track..

I would have put the vintage EQ as it's a hog but not everyone has that.

I will probably put a space designer as well.....

Will be finished within half hour and upload everything.

TNM 25th August 2018 11:45 AM

well this is it,

I really think I have created the perfect new benchmark test.. and props to Evan, didn't realise how long it would take to get the balance right.

On my 2015 macbook pro 2.8hghz quad, i get 37 tracks with turbo disabled or 38 with turbo enabled.. The really is something to this turbo boost switch software.. difference is.. CPU is at 74 degrees instead of 95!
Turbo boost enabled will initially get 45 tracks for about 8 bars but then overheats and drops the cpu frequency.. the macbook pro just can't maintain it.

I will inform the logic test topic as well as other forums. Space Designer just took up too much ram for so many tracks so i replaced with platinum verb, which WILL still load just fine for everyone even though it's "hidden" in Logic 10.4

Details of test:

44K
Tracks on load : 128 instrument tracks each with sculpture, EQ, multipressor, Chorus, Auto Filter and Platinum Verb, playing pleasant 4 note chord

Tracks with their midi enabled by default: 50 (i came to this figure for people with new hex core macbook, hex core mac pro users and so on, it's very easy to enable more just like with Evan, mute tool will load with project so just click the midi clips to enable or disable one by one!)

Logic total ram use from Fresh Launch and project load : 1.1 GB

Project Length: 32 bars looping at 130 BPM (just what my default is.. if you don't like this let me know and i'll change it)

Audio track selected on Load so you can just play away and all cores will be distributed correctly. You literally don't need to do anything other than mute and unmute clips on a track by track basis.

My idea is that it has to successfully play the 32 bars AND make the loop point without the overload message appearing, for it to be considered a valid result.

My results:

Crystal well 2.8ghz macbook Pro i7 Quad, 16GB ram, 1TB SSD, Nvidia 750M

Internal macbook audio. Buffer: 128, Process Buffer Range: Medium.

Turbo Disabled: 37 tracks

Turbo Enabled: 38 tracks

This is obviously a much more intense test and much more suitable for modern machines.

If a machine out there really CAN play the whole included 128 tracks, it's very simple, just highlight the last track and press command D to duplicate it and copy the midi clip over till you reach your maximum tracks.

Cpu load is spread perfectly across the test, real cpu is low 80's % per core.. Logic is hammering away...

I'll test my imac pro this week, but I am guessing it will be 100 tracks...

I am happy to make it 255 tracks default if that's what you all prefer, and anything else you want changed, just let me know,

so we can get it to be the new default test.. it makes so much more sense with computers being as powerful as they are today.

Cheers

Fernand 25th August 2018 01:32 PM

Wow, I was going to run it on my i7 4 core hack, but remembered that
I run Yosemite/Logic 10.2.4 on this machine.

Also can't download from Dropbox for some reason. Is it a zip? It's
showing up as separate folders and won't download.

I don't think requiring 10.4 is ideal, because some people don't want to
bother with OS "upgrades", and 10.2.4 is a very nice and stable release.
I would hope you can make it 10.2.4 compatible, most everybody can run that.

Assuming of course they had the sense to save that executable.
I don't know how performance compares between the two.

TNM 25th August 2018 05:21 PM

well, if Logic can not open songs in older versions, I had no idea.

I have installed 10.4.1 for the first time straight on this machine, so that's all the apple store gives me access to.

it's a zip file alright, i just checked the link with a different browser where i am not logged into dropbox, and the download this file option is present.. go to the top right of dropbox in the drop down menu and choose direct download..

That said, before, "download this file" showed on dropbox.. so I might just upload the logicx file and re submit the link.. as ideally i want the download option to appear when the link is pressed immediately.

Nothing I can do about older versions.. this is exactly why I didn't use effects or options only in 10.4, as i presumed older logic, even Logic 9, could open the song. With pro tools you can open many versions back.

TNM 25th August 2018 05:26 PM

Edit, post no longer required, one link for all at google drive

TNM 25th August 2018 05:29 PM

NewLogicBenchmarkTest.zip - Google Drive

Google correctly gives a download button when link is accessed :)

If someone has got 10.4 and also an older logic and can transfer it, i would be most grateful

Deleted User 25th August 2018 09:56 PM

2014 Mbp dual core i5 base model = 9 tracks... :lol:
Haswell quad core i5 4690k hackintosh 35 tracks....
I like the older test.... :lol: gooof
In all seriousness, good work thanks!

TNM 25th August 2018 10:34 PM

Quote:

Originally Posted by wegface (Post 13486745)
2014 Mbp dual core i5 base model = 9 tracks... :lol:
Haswell quad core i5 4690k hackintosh 35 tracks....
I like the older test.... :lol: gooof
In all seriousness, good work thanks!

in all seriousness also, believe me when the original test came out, many of us had macs that could do 9 tracks also! :lol:

:) Thanks for trying it!

Fernand 26th August 2018 01:20 AM

GREAT! Well done!
Download works
Project loads and runs in 10.2.4
Easy to use, pleasant sound, good interface. Bravo.
We can't very well call it "new benchmark"
or "latest benchmark". How about TNMBench?

Are we supposed to adjust any settings? I forget,
are the sample rate and buffers set by .logicx file?
Are you specifying any standard settings?

I think that if this is to be a measure of the hardware
we should be specifying all parameters, including
using only the built-in audio for a reference run.

Then people can try all their special audio interfaces
and settings, but first getting a "standard" would be good?


On a 4 Ghz i7 4790K w/ 32 GB RAM running 10.2.4 in Yosemite
while running a lot (25GB) of other apps & stuff, I can run 62 tracks
on the built-in audio with the default 44.1K and a 128 buffer.

jwh1192 26th August 2018 02:46 AM

both computers - Logic 10.4.1 / MacOS 10.12.6

MBP mid 2014 16gb ram .. 500gb SSD - about 26 tracks ..

MacPro 2012 24gb ram .. 256 SSD boot drive / 256 SSD audio drive - about 48 tracks ..

ValliSoftware 27th August 2018 05:50 AM

Quote:

Originally Posted by TNM (Post 13485908)
Hi...

I have created a new benchmark with scultpure at 96K, playing 4 notes in sequence over 16 bars looped..

Very pleasant sound, just the default sculpture loads with, all volumes have been adjusted sensibly, and I have put a Logic limiter on the master bus for safety.

I have included one audio track, which should be highlighted when test is done, so no cores are "live".

This was born out frustration of people over at logic pro help forums sincerely believing that the 10 core imac pro got over 500 tracks in evan test vs 251 for the 8 core..

Of course this is impossible, as many here have stated, since Logic only allows 255 unique instrument tracks anyway.

Sculpture is a lot more CPU heavy, so heavy in fact, that with just 40 tracks of this test at 96K, my quad core 2.8ghz i7 macbook pro is maxed out.

Which means that even an 18 core imac pro won't be able to do 255.

If y'all think 96k is not the way to go, i can make it a 4 note chord at 44K instead of a single note ascending.

Just let me know.

It's time to put the dated test to rest, thank Evan for his hard work..

I am happy to pass on all responsibilities of the test to someone else if they like, this is not about it being my test.. You can put your name on it and say it's yours completely.

This is simply about having a realistic test to use on modern machines, and is made in Logic 10.4.1 which i believe anyone on Logic X will be using.

(Correct me if I am wrong, but maybe Logic 9 can open this as it does have sculpture? I am not sure on that as i can't install Logic 9 right now as i have the physical media version of 9.0.0 and don't even have a DVD drive anymore).

There is just one caveat.. sculpture idle actually uses CPU..

so say I have 64 tracks of sculpture loaded but only one with any midi on it, all cores get hammered.

So there is a tiny bit of work involved on your part.

I will be happy to provide tests with 32, 64, 96 and 128 tracks with sculpture and midi pre loaded.

You will just need to deleted tracks, NOT mute, till you get the right max number for your machine,

OR, right click duplicate the last track one by one if you need to add more (and copy the midi to it, which is why i have made it easy in one long 16 bar joined file for each track).

If i should proceed with this, let me know, i'll be testing my imac pro 8 core this week as i don't veen have logic installed on it right now, and am physically going to an apple store to test the ten core.. yes i will admit, the guy got me that irritated but it's more about people believing that the imac pro 10 core has more than double the power of the 8 core with Logic, which is absurd.
This will finally put that to rest, and i am happy if anyone knows anyone at barefeats to submit the new test to them too.

Let me know, and i'll get it all uploaded ASAP if there is interest.

Cheers.

PS let me know 44 or 96K and i'll tally the votes and choose the most popular.

I believe all macs for years have 96k internal sound, so personally i think setting to internal sound and 96k is the fairest easiest way to compare all macs.

NewLogicBenchmarkTest.zip - Google Drive

It's not just about the machine.
A real test is using the instruments you use in a test, but then which library too.
Most people use Kontakt and which library? Can vary and it matters because some load big samples (eats ram) and some use a busy script (eats cpu).

Then there's AU Instruments like Diva. I've notice that doing a test with just Diva doesn't help because a factor is the actual preset as well. I've seen some presets that just peg the CPU, and I'm not talking the single core because I don't have the Diva track selected, I have an empty blank track selected.

In the end, each person just has to optimize their own computer to adjust to loaded applications and 3rd party plug-ins and libraries.

Deleted User 27th August 2018 08:53 AM

Quote:

Originally Posted by ValliSoftware (Post 13488485)
It's not just about the machine.
A real test is using the instruments you use in a test, but then which library too.
Most people use Kontakt and which library? Can vary and it matters because some load big samples (eats ram) and some use a busy script (eats cpu).

Then there's AU Instruments like Diva. I've notice that doing a test with just Diva doesn't help because a factor is the actual preset as well. I've seen some presets that just peg the CPU, and I'm not talking the single core because I don't have the Diva track selected, I have an empty blank track selected.

In the end, each person just has to optimize their own computer to adjust to loaded applications and 3rd party plug-ins and libraries.

Yes, but we all do that anyway by using our computers.
The point of these tests (which you seem to have missed) is to make an even benchmark for comparison purposes.

Fernand 27th August 2018 12:50 PM

Of course, the idea is to test the computer itself.
Users will have different needs and configurations.
==============================

Mac Pro 2009 4,1 flashed to 5,1 2010-2012 Classic tower.
Dual 3.46 Ghz X5690 Xeons, 12 cores/24 threads.
RAM: 64GB. Boot SSD on PCIe. RAID0 6TB rotary.
GPU: HD 7950 3GB, dual 4K 60Hz displays.
External Behringer FCA1616 in FW mode.
44.1k, Buffer 128 Large.

Performance: about 116 tracks.

But are we settling on 96k, internal audio? Buffer 128 Medium ?

ValliSoftware 27th August 2018 04:11 PM

Quote:

Originally Posted by wegface (Post 13488611)
Yes, but we all do that anyway by using our computers.
The point of these tests (which you seem to have missed) is to make an even benchmark for comparison purposes.

This video shows that certain presets make the CPU go high, it's not even the AU instrument per se.


So explain to me in detail how these benchmark tests address this issue.
Nobody is just going to use the built-in synths in Logic. Everyone already has their favorite AU instrument they're going to use which these benchmarks don't address.
So again, please explain in detail how these benchmarks address the issue of presets making the CPU go high and how you're take into consideration different AU instruments regarding these benchmarks.

Deleted User 27th August 2018 04:32 PM

Quote:

Originally Posted by ValliSoftware (Post 13489025)
This video shows that certain presets make the CPU go high, it's not even the AU instrument per se.


So explain to me in detail how these benchmark tests address this issue.
Nobody is just going to use the built-in synths in Logic. Everyone already has their favorite AU instrument they're going to use which these benchmarks don't address.
So again, please explain in detail how these benchmarks address the issue of presets making the CPU go high and how you're take into consideration different AU instruments regarding these benchmarks.

Benchmarks (this included) do not try to address any issue, except create a standard which is used to compare machine performance with each other.
The single core issue is resolved by selecting an empty audio track (by the way set as default in this benchmark).
Many people use the plugins built into daws, they are often very good, and logic is no exception. How can you have a benchmark of something not every user has installed??

:cop:

TNM 28th August 2018 08:02 AM

Quote:

Originally Posted by ValliSoftware (Post 13488485)
It's not just about the machine.
A real test is using the instruments you use in a test, but then which library too.
Most people use Kontakt and which library? Can vary and it matters because some load big samples (eats ram) and some use a busy script (eats cpu).

Then there's AU Instruments like Diva. I've notice that doing a test with just Diva doesn't help because a factor is the actual preset as well. I've seen some presets that just peg the CPU, and I'm not talking the single core because I don't have the Diva track selected, I have an empty blank track selected.

In the end, each person just has to optimize their own computer to adjust to loaded applications and 3rd party plug-ins and libraries.

Sure, and choosing s standard Diva preset would still be a scientifically accurate test as long as the preset was the same an all machines tested. It's all relative.

I guarantee you that machines that score higher in this test will also get more diva polyphony overall, maybe not per core, depending on clock speed, but overall.

I have done diva tests as well, but that was to test the DAW's against eachother. With the "deep space" preset at 16 polyphony and divine quality, playing a 4 note "real world" progression with long release times, diva multi both activated and deactivated, I learned a lot..One, that you should only use a plugin's own multi processing if you have lots of cores and still only for two instances, otherwise let the DAW handle it, and 2, for playback tracks on the high internal buffer, Pro Tools beat every other mac DAW I tested, including Logic/Reaper/S1/Cubase

It's still a useful test..

None of these tests are going to be perfect and translate to every scenario, but I was just wanting to make something that stressed the machine a bit more to replace a test that many use, but their machines can max out the test.

If people are interested in a more varied test, using a variety of instruments and FX, I am happy to make one, just let me know, and which plugins should be included (they'd have to have a free demo without sign up needed to be considered). Or i can do one with a variety of Logic instruments that will work for everyone. I was trying to avoid session orchestral, camel space/phat, and Vintage EQ's cause not everyone uses Logic 10.4

TNM 28th August 2018 08:06 AM

Quote:

Originally Posted by Fernand (Post 13486947)
GREAT! Well done!
Download works
Project loads and runs in 10.2.4
Easy to use, pleasant sound, good interface. Bravo.
We can't very well call it "new benchmark"
or "latest benchmark". How about TNMBench?

Are we supposed to adjust any settings? I forget,
are the sample rate and buffers set by .logicx file?
Are you specifying any standard settings?

I think that if this is to be a measure of the hardware
we should be specifying all parameters, including
using only the built-in audio for a reference run.

Then people can try all their special audio interfaces
and settings, but first getting a "standard" would be good?


On a 4 Ghz i7 4790K w/ 32 GB RAM running 10.2.4 in Yosemite
while running a lot (25GB) of other apps & stuff, I can run 62 tracks
on the built-in audio with the default 44.1K and a 128 buffer.

Hi.. I really think the name is fine, but if others also mention it, then I'll change it. Sample rate and buffers are set by the user.. and I already said what my personal settings were, 128 buffer with medium process buffer. I also suggested that people use apple internal sound to "standardise" the test, but I am not going to dictate anything regarding that, just like people had freedom to choose their own interface and buffers for Evan's test.

I can post recommended guidelines again, but I doubt everyone would follow it.

For the record, my recommended guideline is to use Mac internal audio, 128 buffer, medium process buffer, 44K.

i think since this test scales so well at higher SR's, it would be nice if people wanted to just quickly change the SR in the transport, i have made the project transport so that is just one click away, and post 96K results also. Completely up to you!

Deleted User 28th August 2018 08:09 AM

The test you made already serves the purpose fine. :cool:

Fernand 29th August 2018 09:41 AM

@ TNM , I don't know if you can explain basics to someone who apparently has
never used benchmarks. Or offering to make specialized tests. Your test is fine.

I don't understand. Do you like TNMBench or you think NewLogicBenchmarkTest
is a name with a future?

The point of "recommendations" is to get others to post results with the same
settings. They can always run with other configs, but especially post the lowest
common denominator config results for others to compare. I'll rerun with internal
audio, 128 Medium buffer, 44.1k.

TNM 29th August 2018 12:18 PM

Quote:

Originally Posted by Fernand (Post 13491686)
@ TNM , I don't know if you can explain basics to someone who apparently has
never used benchmarks. Or offering to make specialized tests. Your test is fine.

I don't understand. Do you like TNMBench or you think NewLogicBenchmarkTest
is a name with a future?

The point of "recommendations" is to get others to post results with the same
settings. They can always run with other configs, but especially post the lowest
common denominator config results for others to compare. I'll rerun with internal
audio, 128 Medium buffer, 44.1k.

TNMBench makes it about me which is the last thing I want.. I am happy even for someone else to claim the test as their own and improve on it in future if they like, etc.

I was quite happy with newlogicbenchmarktest but as i said, if someone else agrees with you, ok.

Deleted User 29th August 2018 01:46 PM

Quote:

Originally Posted by TNM (Post 13491838)
TNMBench makes it about me which is the last thing I want.. I am happy even for someone else to claim the test as their own and improve on it in future if they like, etc.

I was quite happy with newlogicbenchmarktest but as i said, if someone else agrees with you, ok.

how about "make-you-want-a-new-computer-and-therefore-a-divorce-benchmark"

:lol:

I think TNMBench is a good name.

dven 31st August 2018 03:39 PM

On a new MBP i7 2.2/560X w/32GB I get 72 tracks reliably at 128, medium. This is LPX 10.4.1 on 10.13.6 with both supplemental updates for the new MBPs.

Cheers!
Dorian

TNM 1st September 2018 05:28 AM

Quote:

Originally Posted by dven (Post 13495151)
On a new MBP i7 2.2/560X w/32GB I get 72 tracks reliably at 128, medium. This is LPX 10.4.1 on 10.13.6 with both supplemental updates for the new MBPs.

Cheers!
Dorian

Awesome! the 8750H is the best processor for the new macbooks as has the least heat of all of them and can maintain it's turbo in many cases. :) You got the best combination macbook for sure, the better graphics with the base processor and full ram.. that's the macbook to get!

I am going to try my imac pro this weekend, finally installed Logic and the 80GB library too! Will try it tonight, I am curious how different the 6 core will be from the 8 core, we both have the same all core turbo of 3.9.

it's also interesting to see, that in real world, the 6 core is twice as fast as my older 4 core.. that shows improvements in clock for clock performance and cpu architecture, but obviously the all core 3.9 turbo has a lot to do with that here too :)

I have to say, from playing around with alchemy and things, the new imac pro is running Logic so well, I am thinking of switching back to it after all that, it's been a pleasure to use so far and the quietest machine i have ever owned, whether windows or mac.

fastlanephil 2nd September 2018 01:02 AM

2009 Mac Pro flashed to a 5,1
3.46GHz Xeon 12-core CPU/32GB of Memory
500GB Boot drive using a PCIe card.1TB SSD PCIe card samples drive.
Radeon 5770 GPU
128 buffer

116 tracks

TNM 2nd September 2018 04:39 AM

Quote:

Originally Posted by fastlanephil (Post 13497422)
2009 Mac Pro flashed to a 5,1
3.46GHz Xeon 12-core CPU/32GB of Memory
500GB Boot drive using a PCIe card.1TB SSD PCIe card samples drive.
Radeon 5770 GPU
128 buffer

116 tracks

WOW the fatal error I made is not using all 255 tracks but now it's too late.. if a 12 core mac pro from that era can get so close to the 128, then an 18 core imac pro will absolutely smash it. So there you go.. I really thought the poor single core performance of those machines would play a bigger part here! I was wrong.

Michaeltn86 4th September 2018 01:10 PM

1 Attachment(s)
MacBook Pro (Retina, 15-inch, Late 2013) 2.6 GHz Intel Core i7 16 GB 1600 MHz DDR3

at 44 only 50 tracks appear. other appears mute by default. processing threads 8 / buffer range large / multithreading playback live tracks / summing standard precision 32 bits / rewire off

with 512 buffer size I was able to play WITH glitches 38 tracks.

with 512 buffer I could play 15 track with small glitches.

with 512 buffer I could play 13 tracks with no glitches

could go down to 32 buffer without glitches with 13 tracks. but some nearly cpu clipping spikes


at 96, I could play with 512 buffer 8 tracks.at 128 it started glitching.

at 96, could play with 64 buffer 6 tracks but cpu nearly clipping. not stable. I attached a print screen. it doesn't matter the number of tracks, even if I go to 1 track my cpu behavior according to logic is the same. so, not stable for tracking. the last core is always near clipping.

there are some things that needs to be improved in the file test imo. the cpu meter should be on and displayed. the audio source should have a very smooth attack, as the person doing the test might confuse it with glitching (I did, so I had to double check if it was the audio or the cpu).

I might have done something wrong and will read the topic again to make sure I didn't make a big mistake.

edit: re read the thread and will redo the test again. I used my MOTU 1248 as IO. will try again later with built in sound card.

I forgot to thank u for the test and the people who posted their results. the new macbook result seems impressive against mine haha. I also would like to suggest to add an additional midi track where u can test if it's possible to record without glitches / cpu spikes

edit 2: with respect (and in hopes to improve the test) btw, I used the file in the 1st post. it's too confusing with different links in different posts. as soon as u have sorted out all potential problems, and what users are looking for (for example, I suggested the midi track to record because all I want is a stable machine to record at 96 - I also suggest u to suggest testing first with the internal IO and then later with the interface io - this should give a nice idea about how logic and interface drivers behave too, and it'd be easy and fast if instructions are well laid out). so, as soon as everything is sorted out, a new topic should be nice :]

thank u for ur efforts. I appreciate it a lot, as I'm constantly double guessing if it's time to upgrade.

TNM 4th September 2018 09:10 PM

hi, i'll make the correct links in the first post only. Fair enough. Secondly, the performance meter is NOT saved with logic song state unfortunately, this drives me crazy too as i have to hit my keyboard shortcut for it every time i launch Logic.

only 50 tracks being active is correct at launch since many people have laptops that can't play the 50 tracks. That is why the mute tool is active on load so you can just click on clips to disable or enable extra tracks. This is how the original logic benchmark test was also designed so i stuck to the same formula. Cheers

johannes2510 5th September 2018 09:38 AM

Hackintosh
i7 8700K
32 gigs RAM
44K
buffer 128 medium


CPU stock, 3.7 ghz - 4.7 ghz turbo: 102 tracks

CPU overclocked at 4.9 ghz: 117 tracks.

http://i67.tinypic.com/avlr0z.png

Quote:

Originally Posted by fastlanephil
2009 Mac Pro flashed to a 5,1
3.46GHz Xeon 12-core CPU/32GB of Memory
500GB Boot drive using a PCIe card.1TB SSD PCIe card samples drive.
Radeon 5770 GPU
128 buffer

116 tracks
I was especting more from my hackintosh due to fastest clock cores.
If I remember well in Evan's benchmark the 8700K has some better results than the xeons of the Mac Pros cheesgraters 5.1. In this test appear that are quite similar.

PS
Somebody knows why the first logical (?) core on the right on the Logic's cpu meter is so low respect all the others?

TNM 5th September 2018 04:22 PM

Logic reserves the last core for live VI's and rewire.