Skip to main content

Found this an interesting read on mixing & mastering at 96kHz and thought it would be worth sharing.

http://www.ryanschwabe.com/blog/96k
 

Comments

kmetal Fri, 09/16/2016 - 19:51

pcrecord, post: 441302, member: 46460 wrote: That's a good point.
This makes me think... Is upsampling less damageable than downsampling ? I'm tempted to say yes. Adding zeros is easier than removing some no ?

That's interesting. I'm not sure. I would imagine your just adding blanks and preserving the original data in the case of upsampling. I think boz explained it in his post.

Sean G, post: 441315, member: 49362 wrote: What DAW are you planning on using Kyle?...was it Samplitude??

Samplitude pro x, PTHD are gonna be the main ones. Eventually Sequoiabecuase of its broadcast features, and ability to run different sample rates and formats in the same session.

Maybe Chris audiokid would have a better knowledge of how Samp handles the SRC from a DAW perspective when it comes to the SRC.

Im curious in the case of vsti if the daw itself is doing the upsampling or if it's done within the sample player. I'm inclined to think the sample player is responsible, but that's just a guess.

Boswell, post: 441322, member: 29034 wrote: This probably arises through a misunderstanding of what happens when upsampling. Going from 44.1 to 96 KHz is not a simple process, but involves digital padding and then filtering. The 96KHz result does not contain the same samples as the 44.1KHz waveform, so will not null when subsequently down-converted to 44.1 KHz. What it does contain is an approximation to the same original waveform, but it certainly does not have any more information in it than the original. This means that if the original was 44.1/24, then there will be no information in the frequency range 20-40KHz in the 96/24 version of the track. If the original was 44.1/16, then although the 96/24 (upsampled) version will typically have non-zero values in the bottom 8 bits of each sample, the track will not have any more than 16 bits of information in each sample. This is not immediately easy to comprehend, but Shannon information theory says this is so. You can expand this reasoning to 192KHz.

If the original analogue waveform was sampled at a lower rate than the DAW processing rate, the lower rate imposes a limit on the amount of information captured, and no amount of processing can restore what was not captured in the first place. You can, of course, add information during processing (such as deliberate distortion), and this will occupy the full bandwidth available within the DAW rate.

It's interesting that in the transfer of old cylinder and 78rpm recordings to CD, it's preferable to leave some of the system hiss in the result, as that fools the ear into thinking there was more high-frequency content in the original recording than there actually was. Restricting the output bandwidth to the known bandwidth of the recording makes the transfered result sound very dull, even though it's a more accurate capture than the version that has more hiss.

Right.

There's no more information than in the 44.1/16 version of the track.

It would change to a 96/24 file because that's what you set it to do. There would be no more information than in the original recording. There would indeed slightly less information, as any sampling rate change process is lossy.

Well said! Gonna have to re read that one to fully understand!

audiokid, post: 441328, member: 1 wrote: Great stuff Bos.(y)

Some of this gives reason why I do a two DAW "uncoupled" pass. I do not hear analog gear (even a straight wire for that matter) as improving the audio, but "just the right amount", especially for harsh mixes recaptured on a second box adds (smears or filters) enough of the edge, creates a faux finish that simply sounds more natural, pleasing in comparison to a cleaner digital mix that is too truthful per-say. It can also help broaden up the stereo spread.

That being said: The reasons I might include a straight wire or added character gear pass could be exclusively to "help" distract the brittle edge of poor conversion. The analog pass between two DAW's (example: 96k to 44.1 or simply both at 44.1 for that matter) can improve the stereo perception too. As Bos touched on, even some dirty hiss recaptured on a second DAW will give a perception that the mix has more high end (broader bandwidth) (more FM like) after we re-process a mix through a two DAW system and some analog EQing.
The addition of analog gear in a pass is more about smoothing out the sharp edges, fitting the changes (altering or down-sampling audio back together (left, right and centre) in a way that doesn't seem to sound as good just digitally.
The addition of analog (OTB mixing) is never about making the audio clearer and more pristine.To me its more about changing it while trying not to sound like I blurred it up. Blurring happens right away so you really need to be listening. Thus, the benefit of transformerless summing and mastering tools .
Its super easy to fool yourself into thinking a change of smear is the improvement while missing you just crossed the transients over and actually blurred up a clients mix.

Remember the days before we had a DAW, when we used samplers? Well, now we have DAW's and the ability to do even more.

  • Two uncoupled DAW's makes it easier to study cause and effect.
  • Two uncoupled DAW's with some character analog gear gets us closer to the world of Sound Design.

class="xf-ul">

Also well said.

I don't mix harshly in general (although I've got one that I did that's brutal on the ears. Shouldn't have mixed for 4 days straight and slept at the studio for that one. Ooops)

I'm more concerned with gaining (or not losing) a sense of dimension, which is the downfall of some summing busses and too much pluggin processing (particularly inserts).

I've a/b'd the bounces w the session, even at the same sample and bit rate before and the bounces have sounded smaller and flatter (less 3D)

The thing I noticed when you processed a bit of one of my 4tracks recordings yeas ago was the depth and openess. Not space or open in the reverb sense, but dimension.

It was the same tracks I used on my mix so I knew it was something to do w you or your system immediately. The bricasti and 2 daw were it, and the dangerous which I didn't know you used at the time.

I've been flapping my gums to every engineer I talk to about the 2nd daw and none of them tried it. None lol. They're loss. Finally I have the money for two computers and comverters, just waiting on my space which I'm building now. (Small writing room in yard). Still missing monitor controller and summing box.

I found the differnce to be not subtle. And otherwise unattainable. Like plugging into a neve vs a typical decent interface pre. They're both good, but one just can't sound like the other.

audiokid Fri, 09/16/2016 - 21:39

kmetal, post: 441335, member: 37533 wrote: The thing I noticed when you processed a bit of one of my 4tracks recordings yeas ago was the depth and openess. Not space or open in the reverb sense, but dimension.

It was the same tracks I used on my mix so I knew it was something to do w you or your system immediately. The bricasti and 2 daw were it, and the dangerous which I didn't know you used at the time.

thanks

kmetal, post: 441335, member: 37533 wrote: I've been flapping my gums to every engineer I talk to about the 2nd daw and none of them tried it. None lol. They're loss. Finally I have the money for two computers and comverters, just waiting on my space which I'm building now. (Small writing room in yard). Still missing monitor controller and summing box.

I often feel we should not say a word. It is without question, something I would keep silent years ago, especially when I was trying to make a name for myself, but the other side of me is saying, share the things we learn.

Brother Junk Sat, 09/17/2016 - 07:35

audiokid, post: 441338, member: 1 wrote: I often feel we should not say a word. It is without question, something I would keep silent years ago, especially when I was trying to make a name for myself, but the other side of me is saying, share the things we learn.

So you are talking about using two computers and two daws, and summing them?

I sent you a pm...I'm lost on the double daw thing.

audiokid Sat, 09/17/2016 - 12:26

Brother Junk, post: 441344, member: 49944 wrote: So you are talking about using two computers and two daws, and summing them?

I sent you a pm...I'm lost on the double daw thing.

Hi, I am just swamped with work so my time right now is very limited. For starters:
There is so much on this topic within our forums already posted.
To my understanding, (humbling speaking) Bos and I are globally leading in this so you will find everything about 2 DAW systems within recording.org.
Start searching:

  • Uncoupled DAW's
  • 2 DAW systems
  • 2 box
  • Mixing into a master
  • Capturing on a second DAW
  • Hybrid summing systems
  • Mixing OTB
  • Mixdown systems

class="xf-ul">

Brother Junk Sat, 09/17/2016 - 13:11

audiokid, post: 441348, member: 1 wrote:
Hi, I am just swamped with work so my time right now is very limited. For starters:
There is so much on this topic within our forums already posted.
To my understanding, (humbling speaking) Bos and I are globally leading in this so you will find everything about 2 DAW systems within recording.org.
Start searching:

  • Uncoupled DAW's
  • 2 DAW systems
  • 2 box
  • Mixing into a master
  • Capturing on a second DAW
  • Hybrid summing systems
  • Mixing OTB
  • Mixdown systems

class="xf-ul">

I already started searching it, thank you!!

kmetal Sat, 09/17/2016 - 15:51

audiokid, post: 441338, member: 1 wrote: I often feel we should not say a word. It is without question, something I would keep silent years ago, especially when I was trying to make a name for myself, but the other side of me is saying, share the things we learn.

I've always felt this conflict, especially when I was training a good friend who would eventually possibly take hours from me.

The way I see it is spread the word. Most people won't listen anyway. And I also look at it as a way of keeping myself ahead of the curve and learning new stuff, that way I'm not giving away my whole show, becuase there's always new stuff I'm learning.

Either way I'm glad you and boz are open about this technique becuase it's going to make an instant improvement. I like things that make kings better just by pluggin in!

audiokid Sat, 09/17/2016 - 16:10

kmetal, post: 441351, member: 37533 wrote: The way I see it is spread the word. Most people won't listen anyway. And I also look at it as a way of keeping myself ahead of the curve and learning new stuff, that way I'm not giving away my whole show, becuase there's always new stuff I'm learning

Indeed. (y)

EDITED:
My new system is going to be so awesome. I'm pretty thrilled about it all. I share exactly what you just said because the gear enables advanced study but you still need to understand what its about and why monitoring is so important, and how two DAW's can work together. So like you say, we are always learning more about things as we go. Without the history, its not that easy to just dive in and start being a pro.

Beside the gear itself, this is a lot about hearing small changes and not missing these subtleties that have big impact.
Two DAW's make it possible to learn steps I would never hear (or possibly discover) on one DAW. Yet I do believe one DAW is all we need, two DAW's is like having more cutting blades or better lighting to create a painting.

Two DAW's make it easier to check work, check how gear effects sound good and badly, how gear and plug-ins differ, how to model them, check mixes and masters and more.

kmetal, post: 441351, member: 37533 wrote: Either way I'm glad you and boz are open about this technique becuase it's going to make an instant improvement. I like things that make kings better just by pluggin in!

I think we often forget this business is all about hearing, containing, capturing and preserving. If we miss our cues, get lost in the clutter, we spin in circles, endlessly chasing hype and dead ends. Gear is so important, but hearing and being able to put it all together without getting distracted... thats where this is going as well.

kmetal Sat, 09/17/2016 - 17:19

audiokid, post: 441354, member: 1 wrote: My new system is going to be so awesome

Dare I ask what you got up your sleeve? (I know your busy I'm okay w just a teaser... :))

audiokid, post: 441354, member: 1 wrote: Beside the gear itself, this is a lot about hearing small changes and not missing these subtleties that have big impact

+1,000,000

audiokid, post: 441354, member: 1 wrote: Two DAW's make it easier to check work, check how gear effects sound good and badly, how gear and plug-ins differ

I'll be doing some example on the same pluggin w the same settings in Sam vs PT and at different sample rates for one of my YouTube vids. I think the results will be very informative especially when the end user listens to say the soundcloud on the phone vs the .wav in their studio.

audiokid Sat, 09/17/2016 - 17:52

kmetal, post: 441358, member: 37533 wrote: Dare I ask what you got up your sleeve? (I know your busy I'm okay w just a teaser... :))

It's ongoing but the core of the 2 DAW hardware system is the Folcrom, Universal Audio, Millennia Media, Dangerous Music, Bricasti, Eventide, Pulse Techniques, SPL.

Both DAW's are Sequoia 13 with Ableton as a secondary

  1. DAW 1: I'm unsure of the multitrack conversion and interfacing

  2. DAW 2 is Lavry Blacks - AES EBU.
    class="xf-ul">

audiokid Sat, 09/17/2016 - 18:06

kmetal, post: 441358, member: 37533 wrote: I'll be doing some example on the same plugin w the same settings in Sam vs PT and at different sample rates for one of my YouTube vids. I think the results will be very informative especially when the end user listens to say the soundcloud on the phone vs the .wav in their studio.

looking forward to that.(y)

audiokid Sat, 09/17/2016 - 18:15

Same DAW system which are both W7 - I may upgrade to W10. I'm using W10 pro with my mobile laptop and really like it but my 2 DAW system is flawless so why change unless drivers or software start dating, forcing me to update.

DAW 1 is an 8 core 965 i7
DAW 2 is an AMD server that I've been using for 5 years now. The mixdown DAW can be more modest.

kmetal Sat, 09/17/2016 - 21:50

audiokid, post: 441362, member: 1 wrote: Same DAW system which are both W7 - I may upgrade to W10. I'm using W10 pro with my mobile laptop and really like it but my 2 DAW system is flawless so why change unless drivers or software start dating, forcing me to update.

DAW 1 is an 8 core 965 i7
DAW 2 is an AMD server that I've been using for 5 years now. The mixdown DAW can be more modest.

Cool thanks! My current daw1 is gonna become a vsti host eventually so I'm kinda planning the real daw 1.

I agree w the OS thing. If it ain't broke.... Remote access to the desktop is the main feature I've noticed between w 7 and 10. I'll be in w10 pro becuase I'm starting from scratch.

audiokid, post: 441363, member: 1 wrote: Just committed Lavry for DAW2. Unsure what interfacing I will go with for them.

Congrats!!! Lavry does not mess around, and the prices and configurability make it an excellent real world option.

DonnyThompson Sun, 09/18/2016 - 04:55

kmetal, post: 441353, member: 37533 wrote: Oh ditto for samplitude man!!!!!! If only more people heard there DaW's..... I though I was crazy when I used to think Adobe auditon sounded better than most others.

Come to find out all digital is 1's and 0's, but not created equally.

I used to be one of "those" guys; who thought that all DAW's were created equal and that they all sounded the same, the only differences being the features between them.
I'll admit, I was dubious when Chris ( audiokid ) was nudging me towards a move to Samplitude... because I was clinging to the same mindset I'd had for so many years, that all DAW's were essentially nothing more than feature-rich samplers, and that "44.1 was 44.1" no matter which DAW I used, and that the differences would be in the hardware and not the software.

So, I finally tried Samp, and I was blown away by the differences in the audio quality from that of my previous platform, Sonar ( vers. 8 PE).
Suddenly, the audio had more clarity, more definition; and with the exception of the software, I hadn't changed a thing... same i/o, same pre, same computer, same monitors...

I'm no software coding expert, so I can't explain why Samp sounded better... I'm sure it has to do with something in the coding of the audio engine, ( I don't know if it's in the playback or the recording, perhaps both) or the other various processes in the software - I couldn't tell you what those would be - but there was no doubt in my mind ( or to my ears) that Samp sounded far better than my previous DAW did.

All I can say is that I became a convert at that moment. Now...little did I know I was in for a pretty serious learning curve with the program, and I'd made the mistake of switching platforms in the middle of an album project ( note to self... don't ever do that again...LOL), but in the end it was worth it.
(Finding Martin's ( Kraznet) instructional videos on YT was a life saver, btw.)

After working with the program for almost two years now, I can confidently say that I really don't see myself switching to another platform from here on out.

Brother Junk Sun, 09/18/2016 - 07:39

Samplitude huh?

I'm researching this double daw thing now. I have (installed) Logic 10, PT10/11, I'm pretty sure I still have Ableton on there. My studio experience is obviously zero, relative to all yours, but I have never seen anything but PT in studios.

This is the first time I have even heard of samplitude.

kmetal Mon, 09/19/2016 - 11:14

Brother Junk, post: 441370, member: 49944 wrote: Samplitude huh?

I'm researching this double daw thing now. I have (installed) Logic 10, PT10/11, I'm pretty sure I still have Ableton on there. My studio experience is obviously zero, relative to all yours, but I have never seen anything but PT in studios.

This is the first time I have even heard of samplitude.

Pt is most common in large comercial studios for various reasons mostly financial investment and logistics.

As far as pro artists go I see a lot of people mentioning logic and cubase.

Samplitude is an excellent program. I'm using it in my new system I'm putting together as the main daw.

As far as the dual daw thing the software is t nearly as important as the two sets of comverters/interfaces and summing box. For what I know about it anyway.

DonnyThompson Tue, 09/20/2016 - 01:17

Brother Junk, post: 441370, member: 49944 wrote: This is the first time I have even heard of samplitude.

Chris (audiokid ) or Tim Dolbear (TimDolbear ) would have to check me on the history/origins of the platform, but as far as I know, Samp is the offshoot of Sequoia, which has been a popular mastering platform choice for pro mastering engineers for quite a few years now.
It was developed as a multitrack production platform version of Sequoia, and while it's nowhere near as popular as Avid, more than just a few previous PT users have ended up switching to it, because of its sonic accuracy, and its object-based editing/mixing approach.
If I had to guess, I'd wager that a large part of this migration happened when PT went 64 bit, and the early versions wouldn't support the 32 bit processors that so many people had spent lots of money on, and early 64 bit versions of platforms like PT -and even Sonar- required an additional "bit bridge" program to adapt 32 bit plugs to the new platforms... And while they were supposed to allow you to use older 32 bit plugs, there were some that wouldn't work...

While Samp's coding for their 64 bit platform allowed immediate and seamless integration with those older 32 bit plugs that DAW users had invested so much money into up to that point.
Most of those DAWs have since improved this 32 bit plug adaptation to their 64 bit platforms, but it wasn't like this during the early days of the 32 to 64 bit changeover. Samp might not have been the only platform to do this, but they were certainly one of the few firsts who did.

Samp Pro X is not an "easy" program to learn; it's pretty deep and feature rich;, even for veterans it can seem to be a bit "daunting" at times... but once you get a handle on it, it's a highly intuitive and powerful program.

FWIW
-d.

kmetal Tue, 09/20/2016 - 02:07

It's funny becuase PT is slowly moving towards object based editing. They have clip gain, and in PTHD they have clip eq and maybe compression. Super basic clip functions, especially relative to Sam.

With quick bounce PT is moving towards printing tracks i.e. The 'commit' function, and track freezing. This is gonna allow for even more powerful pluggin processing higher bit and sample rates, and more remote realitime collaboration.

According to SOS samplitude started way back in the early 90's!!

http://www.soundonsound.com/reviews/magix-samplitude-pro-x

"Like many other audio applications, including Steinberg's Cubase, Emagic's Logic, Ableton's Live and Presonus' Studio One, Samplitude began life in Germany — but unlike them, it has its roots in the former Communist bloc. Tilman Herberger and Titus Tost met at university in the former German Democratic Republic, where they managed to get their hands on the first Commodore Amiga microcomputer to make it through the Iron Curtain. It was for this platform that early versions of Samplitude were developed and, following the reunification of the country, released commercially. From the beginning, it was a fully native application, and in 1994, Herberger and Tost ported Samplitude to the Windows platform. They remain very much in charge of its development.

The Magix company was founded in 1993, and alongside professional tools such as Samplitude, developed a formidable range of multimedia products in the music, video and photo markets, currently employing around 330 people. The main company headquarters is in Berlin, but Herberger and Tost still head up a large software development team in Dresden, where I met them in January this year. "

Chris Perra Tue, 09/20/2016 - 05:22

I did a test. Recording drums at 48 k.. Then importing them fader settings and all.. no plug ins though into a new project set to 96k.

Cubase did the conversion.. I did a render at 16 bit 44.1 for each.. they almost null completely.. Where they don't is in and around 20 k It's totally inaudible for me.. Even cranked way up and is at -60 db or so...

Then I threw one Uad Shadow Hills comp on the output buss..
The difference increased.. now the difference was audible if you cranked it up.. around -38...

So.. I wonder if Uad or other developers work at 44.1, 48, 96, 192, to get the exact modeling characteristics.

Although....just because we can see a difference.. I doubt anyone could hear the difference in a blind test..

Attached files

audiokid Sat, 01/28/2017 - 15:21

Namin, post: 447036, member: 50331 wrote: Its interesting that thats what Mike Shipley also had had to say in one of his rare interviews.

+1

Dan Lavry used to chime in here and talk about conversion years back. 88.2 is an ideal SR to track if you are bouncing down to 44.1/16 for CD. But since I am no longer bouncing down on the same DAW, I've essentially eliminated the reasoning behind 88.2.

Namin Sun, 01/29/2017 - 00:03

audiokid, post: 447038, member: 1 wrote: +1

Dan Lavry used to chime in here and talk about conversion years back. 88.2 is an ideal SR to track if you are bouncing down to 44.1/16 for CD. But since I am no longer bouncing down on the same DAW, I've essentially eliminated the reasoning behind 88.2.

Thanks Chris! I have been reading everything you and Kyle have been posting on this wonderful site! I also remember Mike Shipley saying that plugins perform better at higher sample rates in the same interview. Yes of course, this was all before he passed away in 2013. But I remember him saying that now that the converters are all getting better and better, he can pretty much stay in the box.
Anyways, I have been recording at 48khz for the last 3 years or so and have been intrigued to move up to 96 after reading about 2 DAWs here in Recording.org. So the first purchase this year was an RME HDSPe MADI card which I have now and am saving up for the Dangerous Music ST and either the Orion 32 or the Ferrofish 32. Not sure of the summing box, but maybe the Burl Audio B32 Vancouver. or the SPL MixDream. Friedeman Tischmeyer was a huge advocated of the SPL products. Or would a summing box even be necessary if I were to go with the 2 DAW setup? I have not been entirely sure here reading various threads on the topic of analog summing with respect to 2 DAWs. Does hybrid here refer to an analog summing box of some kind. Again, if anyone can help me choose either the Orion or the Ferrofish, that would be a great help. Ferrofish has even Dante connectivity incorporated this year. Your thoughts and recommendations can be of great help.

audiokid Sun, 01/29/2017 - 00:24

Namin, post: 447051, member: 50331 wrote: Thanks Chris! I have been reading everything you and Kyle have been posting on this wonderful site! I also remember Mike Shipley saying that plugins perform better at higher sample rates in the same interview. Yes of course, this was all before he passed away in 2013. But I remember him saying that now that the converters are all getting better and better, he can pretty much stay in the box.
Anyways, I have been recording at 48khz for the last 3 years or so and have been intrigued to move up to 96 after reading about 2 DAWs here in Recording.org. So the first purchase this year was an RME HDSPe MADI card which I have now and am saving up for the Dangerous Music ST and either the Orion 32 or the Ferrofish 32. Not sure of the summing box, but maybe the Burl Audio B32 Vancouver. or the SPL MixDream. Friedeman Tischmeyer was a huge advocated of the SPL products. Or would a summing box even be necessary if I were to go with the 2 DAW setup? I have not been entirely sure here reading various threads on the topic of analog summing with respect to 2 DAWs. Does hybrid here refer to an analog summing box of some kind. Again, if anyone can help me choose either the Orion or the Ferrofish, that would be a great help. Ferrofish has even Dante connectivity incorporated this year. Your thoughts and recommendations can be of great help.

Thank you for the kind words.
Good choices! You can't go wrong with any of the products mentioned. However, I would start off with two DAW's and uncouple them before buying a summing box. Get to know how that works and then if you want to start adding external hardware a summing box like the MixDream would definitely be a good step.

Brother Junk Mon, 01/30/2017 - 07:32

audiokid, post: 447038, member: 1 wrote: +1

Dan Lavry used to chime in here and talk about conversion years back. 88.2 is an ideal SR to track if you are bouncing down to 44.1/16 for CD. But since I am no longer bouncing down on the same DAW, I've essentially eliminated the reasoning behind 88.2.

Bc the SR is cut exactly in half?