Hackintosh leads to different result true Peak and LuFs - same project

Find topics on computers, studios and music-related hardware.
Post Reply
weirdtrash
New Member
Posts: 8
Joined: Mon Jan 31, 2011 1:10 pm
Contact:

Hackintosh leads to different result true Peak and LuFs - same project

Post by weirdtrash » Fri Jun 21, 2019 1:42 pm

Hi, maybe someone can bring some light to my question.

New build hackintosh, Z390 DESIGNARE | i9 9900K | Vega 56 8GB
Mojave 10.14.5

All setup and running.

DAW Cubase Pro 10.0.20

UAD Apollo Twin mkII and octo satellite via TB 2 - TB 3 mac adapter

Opened an existing project that was mixed on an iMac 27"
wanted to do a null test, then analyzed the audio and to my suprise, the results were not the same. I got higher true Peak Level and higher LuFs.

Thats leading to my guess that faster CPU and faster TB3 Bus leads to different results.

I then checked a bounce on both machines without any plugins inserted, and the results where the same.

Anyone else recognized this issue or is it something else, that I'm missing?

Thank you for your replies on that topic.

Cheers, Micha
Cubase Pro 10.0.40

-Gigabyte Z390 Designare (LGA 1151, Intel Z390, ATX)
-Intel Core i9-9900K (8, LGA 1151, 3.60GHz),
-be quiet! Dark Rock 4 (15.94cm)
-Sapphire Pulse Radeon RX Vega 56 (8GB, High End)
-Predator DDR4 64GB 4-Kit 3600MHz
-1 Samsung 860 EVO Basic (1000GB, 2.5")running OS/1 Samsung 860 EVO Basic (2000GB, 2.5") for audio project data/1 Samsung 860 EVO Basic (2000GB, 2.5") for backup via ICY BOX JBOD
-OSX 10.14.5
-BIOS f6n

soundcards via apple thunderbolt 3 to thunderbolt 2 adapter:

-1 UAD Apollo Silver (thunderbolt card)
-1 UAD Apollo Twin (thunderbolt)
-1 UAD Octo Satellite (thunderbolt)

Controller:
Steinberg CC121
Softube Console 1

User avatar
MattiasNYC
Grand Senior Member
Posts: 3831
Joined: Thu Dec 16, 2010 9:27 am
Contact:

Re: Hackintosh leads to different result true Peak and LuFs - same project

Post by MattiasNYC » Fri Jun 21, 2019 3:15 pm

CPU and TB has nothing to do with levels. My hunch is that something is off somewhere else.

- What do you mean by "wanted to do a null test, then analyzed"? which steps did you take specifically?

- How large was the difference of TP and LUFS?

- What do you mean by "bounce"? Actual Cubase/Nuendo "bounce" or "export audio mixdown" (Pro Tools "bounce")?

- You said after "without any plugins inserted", did you mean first before the bounce you had plugins inserted and you tried without when you did the "bounce"?

- You said then "and the results where[sic] the same"; did you mean you still had that difference?


Intuitively I'd say check your routing. It's not uncommon for people to accidentally set up new systems differently and not notice. Your project should load the same, but it's I guess hypothetically possible for your "studio setup" or whatever it's called to be different and unaffected by the project. So check that you didn't accidentally double up on outputs/monitoring paths etc. And check where you're measuring levels.

Also make sure automation plays the same if you're using it. And check all faders and trims etc.

Really need more clear information about what you did and how though.
Nuendo 7.1.4 / Lynx TWO-B / Windows 10 Pro 64-bit / Ryzen 1700 3.7GHz (oc) / 16GB Corsair Vengeance DDR4@3200MHz / Nvidia GTX 660 / ASUS x370-A mobo/ 500GB WD Blue system drive / Crucial BX100 250GB SSD media / spinners for library/backup ::::: iZotope RX / Phoenixverb Surround / DaVinci Resolve / Faderport / Applied Acoustics UltraAnalog / my pet pony Frank

weirdtrash
New Member
Posts: 8
Joined: Mon Jan 31, 2011 1:10 pm
Contact:

Re: Hackintosh leads to different result true Peak and LuFs - same project

Post by weirdtrash » Fri Jun 21, 2019 5:31 pm

Hi MattiasNYC

thanks for taking time!

- What do you mean by "wanted to do a null test, then analyzed"? which steps did you take specifically?


I exported from the same project the stereo bus on each machine - then imported the 2 stereo tracks into a new project and flip the phase on one stereo track - the result should've been silence - to be sure the hackintosh is getting the same result, which is not the case. So I run the statistic on each stereo track, see attached picture

- How large was the difference of TP and LUFS?


see attached picture

- What do you mean by "bounce"? Actual Cubase/Nuendo "bounce" or "export audio mixdown" (Pro Tools "bounce")?

yes sorry, export audio mixdown

- You said after "without any plugins inserted", did you mean first before the bounce you had plugins inserted and you tried without when you did the "bounce"?


After realising that it's probably a latency issue with a hole mix I opened a new project on both machines, and imported a stereo file and export audio mixdown without any plugins on both machines

- You said then "and the results where[sic] the same"; did you mean you still had that difference?[/i]

When I compared those exports of a stereo track from each machine there was no difference at all - so all good without any plugins


Intuitively I'd say check your routing. It's not uncommon for people to accidentally set up new systems differently and not notice. Your project should load the same, but it's I guess hypothetically possible for your "studio setup" or whatever it's called to be different and unaffected by the project. So check that you didn't accidentally double up on outputs/monitoring paths etc. And check where you're measuring levels.

I've imported all the preferences from the old machine into the new machine (Library/Preferences/Cubase) should be fine

Also make sure automation plays the same if you're using it. And check all faders and trims etc.

Really need more clear information about what you did and how though.


the automation is the next thing I will check

Thanks again for taking time
statistic.png
(140.81 KiB) Not downloaded yet
Cubase Pro 10.0.40

-Gigabyte Z390 Designare (LGA 1151, Intel Z390, ATX)
-Intel Core i9-9900K (8, LGA 1151, 3.60GHz),
-be quiet! Dark Rock 4 (15.94cm)
-Sapphire Pulse Radeon RX Vega 56 (8GB, High End)
-Predator DDR4 64GB 4-Kit 3600MHz
-1 Samsung 860 EVO Basic (1000GB, 2.5")running OS/1 Samsung 860 EVO Basic (2000GB, 2.5") for audio project data/1 Samsung 860 EVO Basic (2000GB, 2.5") for backup via ICY BOX JBOD
-OSX 10.14.5
-BIOS f6n

soundcards via apple thunderbolt 3 to thunderbolt 2 adapter:

-1 UAD Apollo Silver (thunderbolt card)
-1 UAD Apollo Twin (thunderbolt)
-1 UAD Octo Satellite (thunderbolt)

Controller:
Steinberg CC121
Softube Console 1

User avatar
MattiasNYC
Grand Senior Member
Posts: 3831
Joined: Thu Dec 16, 2010 9:27 am
Contact:

Re: Hackintosh leads to different result true Peak and LuFs - same project

Post by MattiasNYC » Fri Jun 21, 2019 10:21 pm

Try two exports from the same machine with all plugs on both exports. If you have plugs that play back ever so slightly differently every time that could explain the difference. For example auto-planning not synced to the project tempo, or something like that. That would happen on consecutive exports on the same system.
Nuendo 7.1.4 / Lynx TWO-B / Windows 10 Pro 64-bit / Ryzen 1700 3.7GHz (oc) / 16GB Corsair Vengeance DDR4@3200MHz / Nvidia GTX 660 / ASUS x370-A mobo/ 500GB WD Blue system drive / Crucial BX100 250GB SSD media / spinners for library/backup ::::: iZotope RX / Phoenixverb Surround / DaVinci Resolve / Faderport / Applied Acoustics UltraAnalog / my pet pony Frank

weirdtrash
New Member
Posts: 8
Joined: Mon Jan 31, 2011 1:10 pm
Contact:

Re: Hackintosh leads to different result true Peak and LuFs - same project

Post by weirdtrash » Sat Jun 22, 2019 6:55 am

that‘s a good idea, will do that.

Have a great weekend
Cubase Pro 10.0.40

-Gigabyte Z390 Designare (LGA 1151, Intel Z390, ATX)
-Intel Core i9-9900K (8, LGA 1151, 3.60GHz),
-be quiet! Dark Rock 4 (15.94cm)
-Sapphire Pulse Radeon RX Vega 56 (8GB, High End)
-Predator DDR4 64GB 4-Kit 3600MHz
-1 Samsung 860 EVO Basic (1000GB, 2.5")running OS/1 Samsung 860 EVO Basic (2000GB, 2.5") for audio project data/1 Samsung 860 EVO Basic (2000GB, 2.5") for backup via ICY BOX JBOD
-OSX 10.14.5
-BIOS f6n

soundcards via apple thunderbolt 3 to thunderbolt 2 adapter:

-1 UAD Apollo Silver (thunderbolt card)
-1 UAD Apollo Twin (thunderbolt)
-1 UAD Octo Satellite (thunderbolt)

Controller:
Steinberg CC121
Softube Console 1

weirdtrash
New Member
Posts: 8
Joined: Mon Jan 31, 2011 1:10 pm
Contact:

Re: Hackintosh leads to different result true Peak and LuFs - same project

Post by weirdtrash » Thu Jun 27, 2019 2:38 pm

Hi again

sorry for my late response.
I did an audio mixdown twice from the same song on the iMac where the mix was done, and the result is not the same
stats.png
(476.03 KiB) Not downloaded yet
So nothing to do with the hackintosh then, probably plugin related see all used
plugins.png
(345.5 KiB) Not downloaded yet
plugins in attachment).
Cubase Pro 10.0.40

-Gigabyte Z390 Designare (LGA 1151, Intel Z390, ATX)
-Intel Core i9-9900K (8, LGA 1151, 3.60GHz),
-be quiet! Dark Rock 4 (15.94cm)
-Sapphire Pulse Radeon RX Vega 56 (8GB, High End)
-Predator DDR4 64GB 4-Kit 3600MHz
-1 Samsung 860 EVO Basic (1000GB, 2.5")running OS/1 Samsung 860 EVO Basic (2000GB, 2.5") for audio project data/1 Samsung 860 EVO Basic (2000GB, 2.5") for backup via ICY BOX JBOD
-OSX 10.14.5
-BIOS f6n

soundcards via apple thunderbolt 3 to thunderbolt 2 adapter:

-1 UAD Apollo Silver (thunderbolt card)
-1 UAD Apollo Twin (thunderbolt)
-1 UAD Octo Satellite (thunderbolt)

Controller:
Steinberg CC121
Softube Console 1

User avatar
MattiasNYC
Grand Senior Member
Posts: 3831
Joined: Thu Dec 16, 2010 9:27 am
Contact:

Re: Hackintosh leads to different result true Peak and LuFs - same project

Post by MattiasNYC » Sat Jun 29, 2019 2:12 am

Well that's good news actually!

Like I said, I'd suspect any plugin that runs an oscillator that isn't perfectly synced to your project tempo, or any "nonlinear" emulation plugin that introduces randomness or whatever...

Fortunately it doesn't really make that much of a difference in terms of meeting delivery specs. You'd likely (!) place a True Peak limiter last in the chain anyway so a half dB difference is really not an issue with that limiter after. And the average levels are essentially the same.
Nuendo 7.1.4 / Lynx TWO-B / Windows 10 Pro 64-bit / Ryzen 1700 3.7GHz (oc) / 16GB Corsair Vengeance DDR4@3200MHz / Nvidia GTX 660 / ASUS x370-A mobo/ 500GB WD Blue system drive / Crucial BX100 250GB SSD media / spinners for library/backup ::::: iZotope RX / Phoenixverb Surround / DaVinci Resolve / Faderport / Applied Acoustics UltraAnalog / my pet pony Frank

Post Reply

Return to “Computer/Studio Hardware & Setup”

Who is online

Users browsing this forum: Google [Bot] and 2 guests