Hybrid setup, summing mixer questions

Discussion in 'Summing / Mastering consoles' started by Gahlord, Sep 9, 2012.

  1. Kurt Foster

    Kurt Foster Distinguished Member

    Joined:
    Jul 2, 2002
    Location:
    77 Sunset Lane.
    Speaking of SPL / Mixdream, the SPL / Neos 1010 looks very cool. :cool: In Short: Sound Performance Lab

    Faders and inserts and compression ! Oh my!

    I don't think we're in Kansas any more Toto!
     
  2. Boswell

    Boswell Moderator Distinguished Member

    Joined:
    Apr 19, 2006
    Location:
    UK
    Home Page:
    This is an interesting thread! I'll chime in here to clarify one aspect of what is being talked about, and that's SRC (sample rate conversion).

    SRC is needed only when the recording sample rate is different from the output target sample rate. If you are recording at 44.1KHz with the aim of producing a CDs at the end of it, then SRC does not feature in the process, and I do many recordings of demos and other material at 44.1KHz. However, my preferred method for high-quality recordings is to capture all tracks at 96KHz, usually on a pair of Alesis HD24XRs. I will then replay these 96KHz tracks into an analog mixer and capture the 2-track mix both at 96KHz back to a couple of spare HD24XR tracks but also split to a conventional audio interface into a computer at 44.1KHz.

    Using this method, the mix is done in analog from the 96KHz tracks, and I need no digital SRC to go down from 96KHz to 44.1KHz. I have not found a way of generating better CD-quality stereo results, whatever the quality of the individual pieces of gear used. But ahead of actual gear, there are two things to which I attribute the sound quality of the method:

    (1) the avoidance of a digital SRC. None the SRCs I have tried are completely transparent. I think Chris reported that he immediately noticed an improvement in his mixes when he adopted this source-mix-capture method.

    (2) Not mixing at 44.1KHz. The addition of tracks that all have brick-wall anti-aliaising filters at around 20KHz produces a tiring top octave. This effect is not dissimilar from the bedroom-recordists results of tracking everything using a single bright condenser microphone. When the top octave is 20-40 KHz, the 10-20KHz region is much cleaner, and having just the stereo 20KHz filter for the mix capture is a tolerable necessity.
     
  3. audiokid

    audiokid Chris Staff

    Joined:
    Mar 20, 2000
    Location:
    BC, Canada
    Home Page:
    Nicely put, Bos and thank you for making me aware of this when I first got interested in hybrid!

    I just got a Korg MR2000SBK for this. Boswell, can you explain more about DSD and why we aren't seeing larger scale, 8 and 16 channel DSD recorders like this?
     
  4. audiokid

    audiokid Chris Staff

    Joined:
    Mar 20, 2000
    Location:
    BC, Canada
    Home Page:
    Bos, have you tried and compared numerous analog consoles and found one over another more favourable? Has headroom been a factor? I find threads where people try summing otb, who use an older analog console and they eventually succumb to ITB. As noted in a few threads back, I personally trust they are not going about it well but also wonder if their console lacks headroom too.

    The MixDream and Dangerous Master have high headroom. I've never actually tried to push them to the limit.

    Kurt, the NEOS would be stunning! There is a long thread on it where people joke about it being overkill. Understandably why, it goes right over most peoples heads. I would love to have that. It would be the ultimate summing system. oohhh la la.
     
  5. Boswell

    Boswell Moderator Distinguished Member

    Joined:
    Apr 19, 2006
    Location:
    UK
    Home Page:
    I could be cynical and say that because editing DSD streams is not supported by PTHD, there's no market for DSD recording devices. However, this just masks the fact that DSD edits are difficult and complex. I had a hope a couple of years back when DSD started to make its mark that a compromise editing scheme would be worked out where you made PCM versions of your DSD input tracks and worked out all the edits, effects and mix that you wanted using the PCM tracks. You would then set off an overnight batch job that built a stereo DSD mix from the original DSD inputs using the parameter set you had decided upon from your work on the PCM tracks. Sadly, this does not seem to have happened.

    Headroom you can sometimes work around, usually at the expense of noise floor. However, purpose-built analog high-level mixers like the two you mention are in a different league.

    For me, the big difference between using nice-sounding conventional analog mixers over cooking varieties for mixdowns is in things like the musicality of the EQ circuits and whether the line inputs are constrained to go through the pre-amps. This sort of thing does not need to be very expensive, although it's not in the bottom-level price bracket. For example, I get great results from an A+H Zed-R16, ironically used purely in its analog mode, and from some of the older Midas boards like the Venice and Verona. All these are excellent-sounding products for analog mixdowns as long as you take care to avoid the pre-amps by using the insert returns for your input signals where the line inputs are not separately routed to the main mix.
     
  6. Gahlord

    Gahlord Active Member

    Joined:
    Sep 9, 2012
    Location:
    Burlington VT
    Boswell: Thank you for adding that!

    Just to I can make sure I understand what's happening, your signal chain looks something like:

    sounds->preamps->AD at 96khz->DAW1->DA at 96khz-->Analog stuff-->Split to: AD at 44.1 on different recording device/DAW2 for CD --and-- AD 96khz on DAW1? or DAW3? for archival

    Question: How many recording devices are you looking at here? For example-- DAW1 is the ITB mix. DAW2 could be anything (tape/DAT/wax cylinder/Logic/PT/etc) so long as the AD being sent to it makes sense. DAW3 could be either non-existent (sending the stereo sum back to the original DAW1) or else it could be anything.

    It seems a waste to be running three full computers for this given that two of them are simply catching a stereo track. But I suppose computers for that purpose could be relatively inexpensive.

    The purpose of the final 96khz stereo track is so that you can then re-output it to DA 96khz and re-capture it at whatever sample rate you need for different distribution formats, right? Sort of like "re-amping... for masters"?

    Question: Is the headroom you and audiokid discuss the same kind of headroom I concern myself with on my bass--having 1600 watt power stage but never needing to turn up even half way--to get a clear sound? In other words, is one element of quality in the summing mixer configuration dealing with having a lot of available gain?

    Thanks again for your insights here. This thread is really clearing things up for me.
     
  7. Gahlord

    Gahlord Active Member

    Joined:
    Sep 9, 2012
    Location:
    Burlington VT
    Also, glad you are liking this thread too Boswell. My hope is that it becomes the go-to thread for people who express an interest in moving towards an OTB process. I am continually updating the top post so people don't have to trawl through the entire dialogue to find the highlights.
     
  8. Boswell

    Boswell Moderator Distinguished Member

    Joined:
    Apr 19, 2006
    Location:
    UK
    Home Page:
    The main point of the recording/mixing chain that I described I use for high-quality mixes is that no DAW is involved. The recording is done to Alesis HD24XRs, which are 12/24 channel 24-bit digital hard disk recorders that can be stacked in blocks of 12 or 24 channels (12 at 96KHz, 24 at 4.1/48KHz).

    For mixdown, I replay the HD24XRs via their D-A converters into the analog desk and then capture the two-track analog mix out. I usually capture the mix at both the original higher rate (96KHz/24-bit) to spare tracks on one of the HD24XRs, and simultaneously at the target rate, which is most often CD standard 44.1KHz/16-bit but could be video standard 48KHz/24-bit.

    The CD-rate capture can be done on something like an Alesis Masterlink, but I usually use an RME FireFace800 and a simple data capture program.

    I can transfer the higher-rate mix to DVD-audio format, and then use it for blowing the socks off the people I know who have Hi-Fi systems incorporating a DVD player with DVD-audio replay capability. Through hearing DVD-audio like this, they have come to realise that commercial CD-quality leaves a lot to be desired. Since I don't do commercial mastering, the higher-rate 24-bit mix is also useful as a target for the mastering houses who are tasked with squashing the CD-mix to something they regard as releasable.
     
  9. Gahlord

    Gahlord Active Member

    Joined:
    Sep 9, 2012
    Location:
    Burlington VT
    Thanks Boswell. So in the chain above DAW1 and DAW3 would, in your case, be the same device Alesis HD24XR (or a few of them). But the same device may be both playing audio out to the console and receiving the stereo pair back from the console. I'm assuming you run the CD-rate from a monitor out or other out from the console that is the same signal the HD24XR is receiving?

    Super helpful. And great setup eliminating the screen entirely but keeping digital as a medium. I have to ask though, do you use the Alesis HD24XR instead of tape simply to avoid the maintenance and supplies issues of tape? Or because of other advantages in keeping the tracks in a digital medium?

    Also regarding previous post about a DSD edit path: That is exactly how many higher end film edit stations work. All of the editing on screen is done using lower resolution files which creates something called an EDL (edit decision list). That EDL is simply an XML file that gets run against the full resolution file when the time comes to make a print. It's a workflow that should very easily transfer over to audio files--could probably hack film editing software to do it with a few lines of code. I don't know anything about DSD but will have to investigate now. Lol.
     
  10. Boswell

    Boswell Moderator Distinguished Member

    Joined:
    Apr 19, 2006
    Location:
    UK
    Home Page:
    Yes, that's the chain. The mix simply produces a single 2-channel (stereo) analog out; I split this to a pair of HD24XR inputs and to the FF800 or whatever device is capturing the 44.1KHz stream.

    I know only a little about audio processing for film, but was aware of the use of EDLs. The big difference between film and pure audio is the emphasis on editing and sync rather than on levels, effects and dynamics. These latter things play a part in film sound, of course, but you can get away with a great deal more when there is a screen to watch, provided the timing is accurate.

    When I first started this game in the 1960s, tape was what it was all about, but I rarely use tape now. I have shelves of 10.5" boxes that I know I should go through and transfer to digital format, but somehow the time to do it is difficult to find...
     
  11. mightyeskimo

    mightyeskimo Active Member

    Joined:
    Sep 11, 2012
    Location:
    West Michigan
    Hi guys! I'm new to OTB and I just posted a new thread before reading this one. Can you guys share your calibration routine? I'm using Logic Pro and a MOTU 828mk3 Hybrid to output 4 stems to a passive summing mixer. Then from the mixer back to pre's through a compressor and then back into the MOTU. I'm trying to decide on a cal level that works best for me.
     
  12. RemyRAD

    RemyRAD Member

    Joined:
    Sep 26, 2005
    Cal to what? Studio equipment, since it was invented, from the beginning of its existence, calibration reference level of 1.23 V across 600 ohms a.k.a. +4 Dbm, is still the standard in real studio equipment. Everything else was just everything else. That's a normal operating level, where your peaks will be +10 to 15 DB and beyond. And only the professional stuff can do that. So when asking about the best reference level, that is the best reference level. So your preamps, along with your compressors and your passive summing mixer. That should have an active output, are all most likely intended to be used with each other at +4 DB average working level outputs and inputs.

    Now if you are asking where to set the knobs on your various pieces of equipment, there are only general guidelines. Some devices have a marker about two thirds of the way up on some volume controls. General guidelines with rotary controls put that marker to the two o'clock position. This is where most input sources, will equally equal output, a.k.a. unity gain. While at the same time, being able to provide variable gain. From the gain trim, which are two different things. Gain trim affects the internal operating parameters of the amplifier, where the variable gain actually refers to a passive control at the output of the amplifier. Both are interactive. Utilizing them the right way is always good. Utilizing them the wrong way, is frequently good. But that means that you have to understand how to do it right before how you can do it wrong the right way. Wrong the right way is what we refer to as saturation and overdrive. Done right, it can present a more upfront and/or aggressive quality to the sound tonality. Doing it wrong means. It's choking in noise and unlistenable distortion. Sometimes even that can be right. For instance, you certainly wouldn't want the same kind of guitar overdrive plug-in for use on an operatic soprano. Unless you really thought she was that awful and wanted to make sure she couldn't get a head, but rather need to give it?

    The same can be said of your compressors. When you push certain dynamic range devices beyond their rated norm, other cool things can come from it. Otherwise you crushed the crap out of everything leaving your self nowhere to go. And each type of compressor imparts its own unique coloration. If it's an optical unit, peak sensing, average or RMS sensing type and whether it has variable attack and release times or no adjustments for that. They'll create their own magic or, crash and burns. What peak limiter is good for the goose ain't good for the RMS Gander. Or maybe it is? Only you can find out.

    If I told you how to get good sound... I'd have to kill you.
    Mx. Remy Ann David
     
  13. mightyeskimo

    mightyeskimo Active Member

    Joined:
    Sep 11, 2012
    Location:
    West Michigan
    Good point. Sometimes I overcomplicate the theory.
     
  14. RemyRAD

    RemyRAD Member

    Joined:
    Sep 26, 2005
    Overthinking the theory is actually a good thing to do. You need to think that you are a sound wave and don't know how to go home. So you have to put yourself into those little wavelets and pretend you are swimming through your electronics to find the out door. It's sort of like, LET'S MAKE A DEAL TV show, where you get your choice of what's behind door number one, door number two, or door number three. One's a very nice prize and the other two are just stupid funny. I find a lot of our recording equipment today to be just stupid funny. And they are trying to sell you stupid funny telling you, honey, they want your money. And then trying to rationalize their own stupidity in what they are trying to sell you. That's another reason why I like attending the AES shows. They give you all of this technical blather that really has nothing to do with good sound. Then you can just laugh and walk away from them. Knowing that you know better. Let's face it, no one's willing to give a sucker an even break.

    I like Tootsie Pops best. Because in the middle there is something you can dig your teeth into. Not just a stick.
    Mx. Remy Ann David
     
  15. Kevin Big Jam

    Kevin Big Jam Active Member

    Joined:
    Oct 3, 2012
    Location:
    Kent UK
    Home Page:
    This is a great thread I am learning a lot from seeing as I'm about to go Hybrid for the first time as indicated in my first post here. Many points have been cleared up for me already. Boswell, I'm very interested in this capturing method but would you mind explaining to me how it would be different from capturing the St master out from a console into a daw as a 96KHz file and then playing that 96KHz recorded file back through the desk and re-capturing it as a 44.1KHz file.

    Thanks in advance.
     
  16. Kevin Big Jam

    Kevin Big Jam Active Member

    Joined:
    Oct 3, 2012
    Location:
    Kent UK
    Home Page:
    Hi another question from me if you don't mind. The ZR16 is one of my considerations as a summing mixer / new audio interface. It's reading very well indeed and it would seem you confirm that is is a very good sounding unit used in a specific way by avoiding the pre amps & using insert returns for inputs. Would you mind elaborating on that a little more please baring in mind I have no console experience? For info, all my sounds apart fro my vocals and guitars are generated ITB at the moment and I already own a very good pre amp for capturing these. Thanks.
     
  17. Boswell

    Boswell Moderator Distinguished Member

    Joined:
    Apr 19, 2006
    Location:
    UK
    Home Page:
    The simultaneous capture at two different rates results in better audio quality at 44.1KHz, as it saves one set of A-D-A conversions. If it were just a stereo track, you would not in any case need to put it back through a desk. Simple replay at one rate directly captured at another is all that would be needed.

    The pre-amps in the Zed-R16 are excellent, especially given that you get 16 of them in a mid-price console. However, as with many live consoles of this type, the line inputs are attenuated and put through the microphone pre-amps. The insert returns by-pass the pre-amps, but are unbalanced, so you have to deal with that in the wiring or use something like high-quality bal-un transformers. I try to stick to a rule of not feeding line-level signals through pre-amps of any sort unless either it's unavoidable in routing terms or I'm going for a special sonic character such as transformer saturation.

    It's a bit ironic using the Zed-R16 as a purely analog console by ignoring its A-D and D-A converters, but it does perform well in this role, and also the digital side of it is not designed to run at the higher conversion rates.
     
  18. Kevin Big Jam

    Kevin Big Jam Active Member

    Joined:
    Oct 3, 2012
    Location:
    Kent UK
    Home Page:
    Thanks for the swift reply Boswell. Yes of course. My morning head had not even considered that! Funny how we try to over complicate things at times.

    Understood, are the firewire inputs into the ZR16 from the Daw processed through the mic pre-amps as you explained? As an audio interface, I thought the zR16 ran upto 96KhZ? You have already been helpful, however if you get a chance please offer your thoughts on my upgrade options http://recording.org/hybrid-recording-forums/53420-major-upgrade-itb-hybrid-advice-appreciated.html
    Thank you yet again.
     
  19. Boswell

    Boswell Moderator Distinguished Member

    Joined:
    Apr 19, 2006
    Location:
    UK
    Home Page:
    Yes, the D-A converters come in at the same level as the insert points, i.e. after the pre-amps.

    I was a bit terse in describing the digital side of the Zed-R16 as not running at higher rates. It's slightly complicated and restricted, and you should study pages 24-25 of the manual. Basically, the FireWire 2-bus interface and all the ADAT I/O are limited to 44.1/48KHz operation, whereas the individual channel A-D and D-A will run via FireWire up to 96KHz. So you can use it as a mixer with 96KHz digital sources coming in via FireWire but with an analog 2-bus result, or as a multitrack digital recorder via FireWire at up to 96KHz.

    The Alesis HD24XRs that I use for recording have analog and ADAT I/O up to 96KHz, but no FireWire. Their A-D and D-A converters are good, and so the easiest and best-sounding way of working for me is to have the HD24XRs produce high-bandwidth analog outputs that are mixed in analog by the Zed-R16. I could use the Zed-R16 FireWire host to give me a stereo 44.1KHz result, but actually I choose to split the main 2-bus analog outputs to go back to an HD24XR at 96KHz and also to a computer via an RME FF800 or via another independent box at 44.1KHz for the CD-rate capture.
     
  20. Kevin Big Jam

    Kevin Big Jam Active Member

    Joined:
    Oct 3, 2012
    Location:
    Kent UK
    Home Page:
    Hey Boswell, thank you again. I read the manual regarding this and I understand the limitations. They are fine for my planned usage of the unit at this stage. I would be sending 16 x 96KHz audio stems / instruments from logic via the 16 firewire channels to the unit to complete the final mix in the analog domain. Capturing the final mix file on a device like the HD24XR is a very good piece of advice for a transparent final file. A second bounce to that machine could also produce a 44.1KHz version avoiding any SRC inside a DAW at any time. My main query about this unit is of the D/A converters sound as good as other high end audio interfaces like the Apogee Ensemble or Apollo.
    What's worth noting, especially for those producing there own masters, is that you can supply 96KHz masters to I-tunes for distribution as they have optimised there conversion software to create the AAC and MP3 files from this resolution. There is a very good article about that here http://images.apple.com/itunes/mastered-for-itunes/docs/mastered_for_itunes.pdf including info on how to use there software to test how your tracks will sound after the conversions.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice