Routing analog & digital audio between two DAWS

Discussion in 'Computing' started by kmetal, Sep 9, 2016.

  1. bouldersound

    bouldersound Real guitars are for old people. Well-Known Member

    Location:
    Boulder, Colorado
    I know a little about the video side. I've been using Sony's software since before Sony bought it from Sonic Foundry, starting with CD Architect (version 2, I think) and Sound Forge 4.5, then getting into Video Factory, Vegas Video 3, Vegas 6 and now Vegas Pro 13. I've been recording and mixing multitrack audio since Vegas 3. I think this all started for me around 2000, 2001. I'm waiting to hear what Magix does with Vegas.
     
    audiokid likes this.
  2. kmetal

    kmetal Kyle P. Gushue Well-Known Member

    Location:
    Boston, Massachusetts
    Home Page:
    Strangely the engineer did this on the new Coldplay record according to his 'inside track' interview in SOS. Maybe last marchs issue I think. I recently read it and was kinda surprised. Both that he had a capture rig, and more so he used it to capture at a higher rate. lol I'm like you got it backwards bro... but that's why he gets the big bucks. He addressed it in he article saying he'd used a higher sample rate to track/mix but the computer can't handle that along w all the plugins.

    He captured thru Burl and cranesong DA not sure about the capture ad.

    It's difficult for me to cryitisize someone working on that level, but really, between the backwards capture and the million and a half plugins, his method seemed kinda lame. Especially considering all the high end gear he had to track and mix with.

    All that other stuff seems like bullocks to me..

    Buffer underuns are a common plague in PT, particularly the older versions like 7,8 and 9 to a lesser extent. They re wrote the newer versions to better support native (non dsp) computers. It's still the most common error messsage in PT and often un justified. PT doesn't like any sort of changes to its settings for some reason. It's a good idea to restart the computer when you change any settings with buffers or sample rates particularly in PT.

    24/96 shouldn't overload that level of computer even tho it's getting on in years. I would throw a shot in the dark and say you should get at least 24trks at that sample rate without error, especially at that high a buffer.

    I'd try it in one of your other DAW programs and see if you have a similar result. PT has its good merits but rock solid relability, and CPU efficiency are it.

    If your other daw(a) react the same way then it's something to do w the system.

    Your running two HDD right? One for you programs and one just to record audio too?

    It's being in 24 bit that makes the most audible differnce.

    If the destination is cd then many will track at 24/44.1. I've done most of my work at 24/44.1 for various reasons. I've done a little bit at 24/96.

    It's basically a good idea to use 24 bit word length, that affects some important things like gain staging and headroom.

    Once your in 24 bit. The sample rate makes far less of an audible differnce.

    Like Audiokid said not all converters sound better at higher sample rates. Most comverters in general were designed w 96k in mind (even if they go higher) and that's usually what their specs are published on. Prosumer stuff is undoubtably designed w 44.1/96 in mind even if they go higher. It's unlikely someone using a budget interface would have the other gear necessary to take advantage of ultra high sample rates.

    As long as your 24 bit you should be doing just fine at any given sample rate like 44.1.
     
  3. kmetal

    kmetal Kyle P. Gushue Well-Known Member

    Location:
    Boston, Massachusetts
    Home Page:
    I'd replace that mbox with a focusrite Scarlett or preferably and RME babyface PRO. RME is affolradle professional walking the prosumer line on the side of pro. MOTU sits in between focusrite Scarlett and RME. MOTU is prosumer at its best, and is suitable for modest professional work.
     
  4. Boswell

    Boswell Moderator Distinguished Member

    Location:
    UK
    Home Page:
    Was that was the article about Rik Simpson? I remember reading that and thinking Uh - huh, not the best solution to that particular problem. However, what he was getting from the HEDD was harmonic addition, and since that added detailed things to the mix, it's not a problem to have the output generated at a higher SR. I guess he was using digital in (at 48K) and digital out (at 96K) of the HEDD, so no DA - AD process involved there.

    For many reasons, it doesn't fit my concept of a two-box uncoupled system.
     
    kmetal likes this.
  5. Brother Junk

    Brother Junk Active Member

    I think so...I use externals. PT is installed on the OS X drive though, is that wrong?

    Maybe just for clarification, how would you accomplish that...just by putting the whole PT file (.ptx, audio files, etc) on the externals? I have internals as well, that o8 Mac Pro has 8tb in it lol (best $150 computer ever!)

    The reason I do the external thing is because that's what I've seen done at studios. My external is connected with ethernet though instead of USB...maybe that's the problem? The only reason it's setup that way is because I need to find another one of that style USB.

    I ask for the clarification bc I recall seeing this thing setting, buried deep somewhere in the PT menu, where PT will try to split the writing between two drives. At least, that's how I recall it...almost like a raid format. But I was told that this isn't necessary past PT9. It was more for the days when scsi drives etc were in use.

    The RME isn't that much $...gracias. This? (It looks sweet!) http://www.markertek.com/product/rm...mat-mobile-usb-2-0-high-speed-audio-interface

    And it's frickin tiny. I had to look at all the views and zoom in to make sure it had all the i/o I need.

    So, this whole decoupled dual daw thing, I find super interesting. Feel free to correct me where I'm wrong. But it seems to combine the best of everything. And I felt like, I pretty much achieved a rudimentary understanding of it. Until I read what you wrote and @Boswell wrote...lol Although, it sounds like this doesn't apply to a pure, uncoupled system, but I am curious as to...

    What would be the point of a 48k in, and a 96k out? I've devoted a solid 1o minutes to that question and I can't think of why one would do it? E.g. Why would he not be 96k in/out? Or 96k in, 48k out? Bos, are you saying his particular equipment setup led to some sort of pleasing harmonic anomaly? E.g. like a Tube amp vs solid state? Basically, he took advantage of a useful gimmick he found?

    So would the "the less SRC, the better" rule of thumb still apply?
     
  6. Boswell

    Boswell Moderator Distinguished Member

    Location:
    UK
    Home Page:
    Usually the point of the uncoupled two-box method is that you capture the mix as accurately as possible at the destination sampling rate. As I've suggested several times in the past, keep in your mind the model of a stereo microphone being recorded by box 2. Everthing up to that point is concentrated on generating the virtual stereo microphone output.

    In the case of a HEDD in the SOS article, it's acting as an effects box on the main mix. If I understood the article correctly, he was using it to add subtle colouration to his 2-bus mix, so there was no sense of its output being an accurate capture of the input. If the subtle additions had significant components in the top octave of the incoming sampling rate (10 - 20KHz), then it's fair enough to capture the result at a higher rate. An exact doubling of the rate is indeed easier in processing terms than (say) going from 44.1KHz to 96KHz, but it still involves significant processing rather than simply filling each of the missing samples with the average of the two samples either side. There is also the point that if the box will go both up and down in rates (in -> out), then it probably uses a general SRC algorithm for all the possible rates, and will only go transparent if the input and output rates happen to be the from the same clock. The faults in this way of thinking are exposed when you realise that the captured output probably has to come down to 44.1KHz at the end.
     
    kmetal likes this.
  7. kmetal

    kmetal Kyle P. Gushue Well-Known Member

    Location:
    Boston, Massachusetts
    Home Page:
    Yup! I love 'inside track' articles becuase it's like being a fly on the wall on some pretty awsome high end sessions. I love seeing what those guys use and don't use.

    Here's a link to the article (for anyone interested) , and a snapshot of the particular summing section of it. I subscribe digitally, but I'm not sure how much non-subscribers will be able to read.

    Surprising to me was his use of parelell master/mixdowns, where he blends various 2trk mixdowns to taste. Also the use of izotope ozone was a bit surprising too.

    http://www.soundonsound.com/techniques/inside-track-coldplay-hymn-weekend

    IMG_3197.PNG


    IMG_3198.PNG
     
  8. kmetal

    kmetal Kyle P. Gushue Well-Known Member

    Location:
    Boston, Massachusetts
    Home Page:
    lol no shortage of storage. If you've got an internal HDD then it'll be better than using an external HDD for audio. This due to faster transfer rates of sata 3 vs USB or FireWire.

    So you've got your system drive which has all your programs and applications installed, and an audio drive where you save all your PT/daw sessions. This will save your recorded audio there by default. You also would have your samples on the audio drive, or on a third dedicated sample drive.

    Yes that's the interface I was referring to. B&H sells it for 700$ even, 50 cheaper than most everywhere else.

    I've found that interface to be the best compromise of quality / features / price, available right now. If you don't require high I/o. It's expandable I/o his the adat. The closest direct comoeteitor would be apogee duet. Which is similar in quality and channel count, but Mac only, no dsp, and a bit more money.

    The others I've compared to offer higher channel and lower quality for a little cheaper, or higher channel count and similar quality for double the price.

    On the cheaper side it's MOTU traveler, and focusrite scarelett. On the higher channel count, double price, and similar quality it's the apogee quartet, and the antelope zen tour.

    Once you get above those price points you get into the agogee symphony, antelope Orion, and mytek, and other high end devices.

    Yes as far as I know it holds true across the board for digital.

    I'm confused about this, could you explain it a bit more?
     
  9. DonnyThompson

    DonnyThompson Distinguished Member

    Location:
    Akron/Cleveland, OH
    Home Page:
    @audiokid, @Sean G, @Boswell, @kmetal , @pcrecord @bouldersound , @Brother Junk , @dvdhawk , @Davedog

    Like Kyle, this is where I also get confused.

    I'm certainly not against using plugs; I think most here would consider me to be an advocate for their use - if they are needed - and in that regard, if they accent your mixing style in a positive way, and make your mixes sound better than what you'd get without them, then they can be useful tools. After all, not all of us have the luxury budgets available to have Neve/SSL/API/Dangerous/ pre's, with Cranesong, Apogee or Antelope conversion... and personal indulgences aside, I'm not sure it's even financially worth it any more; to invest in that stuff as a business move; we know that more than just a few of the big, pro (some even famous) studios - have been shutting their lights off and locking their doors in the past few years, ( some would even say at an alarming rate) and that many of these places have fallen directly as a result of smaller studios popping up nearly everywhere... and "budget" recording rigs being what they are, I think that this is where the plug-in market flourished. And, I do think that plugs can be useful; whether it's for a particular "modeled" sound, or for more forensic-type frequency or gain tuning.

    But... for those who do still have and use the "creme de'la creme" of audio capture devices, with top-notch gain chains... well, I guess I just don't see it.

    If you have very nice OB pieces available - say, real LA2's, 1176's, Pultecs, Focusrite Red compressors... and you had fantastic-sounding front-load capturing through the nicest, top of the line mics and preamps available, then why should you need to "fix" anything? I'm not saying that tweaks shouldn't be applied, that's what mixing is all about; getting multiple tracks and takes to sound good all together - but why would you reach for a $20 LA2A plug-in if you have an actual LA2a in your rack?

    At that point, do we perhaps need to look at the engineer's lack of talent in getting the best capture possible? That's not rhetorical, gang... I'm really asking here...
    Personally, I don't know how else to look at it, because if you had all the aforementioned pro-level equipment, and you still feel the need to "fix" certain tracks, then I think you have to look at the possibility that the fault lies with the engineer, because it's certainly not the gear, right?

    Again, I'm not being rhetorical... my question(s) really are sincere; I'd like to know what my peers think... ;)
     
    kmetal likes this.
  10. Sean G

    Sean G Well-Known Member

    Location:
    Sydney, Australia
    I cannot understand why you would use a plug-in if you had the real thing present...but then again there are those who hold the view that once you are in the box, stay in the box.
    You only have to look at guys like Andrew Scheps who have gone down the totally ITB road now, and its not like these guys who are at the pointy end don't have access to the very best outboard gear available at their disposal...if they didn't already own it in the first place.

    Another perspective is that it may just be a workflow thing too...everything has to be done at 10 times the speed now compared to even a decade ago...the budgets are gone when you could spend half a day just dialling in an audio chain or auditioning hardware...who needs that when you can just drop a plug-in onto a track and audition it in 30 seconds and if you don't like it you can remove it and just reach for another. You can set a whole template up in a DAW as we all know that is there from the first second...theres' no patching in like the analog days.

    Then there's the whole recallability factor...next time you open your session its all there as you left it...nobody leaves the faders on a console untouched for 3 days waiting for a revision to come back to them..that is if you still have or are using a large format console today.

    With the amount of things like plug-in latency, phasing issues, smearing and digital distortion that come hand-in-hand with plug-ins and the chug-alug effect in chaining in multiple plug-in after multiple plug-in on track after track after track, you need to eliminate as much of that as possible starting with the capture. Fix it in the mic, not in the mix I say.
    That way, if it sounds like ass (thanks to Kurt I now love that saying) then it can only come down to two things...shitty equipment or shitty technique.

    Get both of those things right and you are on the home straight.
     
    kmetal likes this.
  11. Brother Junk

    Brother Junk Active Member

    Ty, that's about what I assumed. Colouration is a better word than gimmick, but you basically understood what I was asking

    I use Ozone! lol. I love Ozone actually. And they have a vocal one that I like but I can't think of it atm. Not Alloy, it will come to me. I think that's actually my favorite Izotope plug-in...and I can't think of the name of it.

    I don't know what Parallel master/mixdowns means but I'll look it up. I'm assuming it's not like a wet/dry track mix.
    Yes, what has me wondering though isn't the Sata/USB/firewire speeds. It's that I don't know if the ethernet (often called gigabit ethernet, so 1,000,000 bytes per/s) applies the same speed to local connections. In other words, I'm not sure if the 1gb ethernet connection is still limited to that speed on a local network. I'm not saying it does or it doesn't. I'm saying I genuinely don't know. But if it is Sata 3 (that's around the right time, I'll look today) it's hard to beat that speed. I wonder why some studio's only use externals? Or maybe they are connecting it with Sata 3....I never examined them from the back.

    Yeah, I will def pick that up in the future. I don't require many more i/o than I have, nm 12 (or 24, whatever it was).

    That's the first thing I thought of. But you guys are the ones who have toyed with the real thing. I've never messed with a real compressor in my life. But I've always thought of the plug-ins as the secondary, or lesser option. But that's an assumption...I've never actually gotten to compare.

    To be completely honest, I still don't fully comprehend how to use compressors. I mean, I can make it work for me, but I've seen people who hear a track, take a second, and then set knee, ratio, threshold etc....just bam, bam, bam.

    Just out of curiosity, have any of you ever compared hw to the plug-in that imitates it? If so, what did you find?

    What is the chugalug effect?

    **Edit, the pi I was thinking of is Nectar/2. I'm a hack, unlike the rest of you, so, I don't know if it would make your cut quality wise, but I love that plug-in. That, Melodyne and PT and I can do vocal tracks pretty quick.
     
  12. Sean G

    Sean G Well-Known Member

    Location:
    Sydney, Australia
    Yep...I have a Warm Audio EQP-WA...a Pultec clone, now whilst it doesn't sound exactly like a Pultec EQP-1A, its pretty damn close if you read the reviews, and a plug-in of the same just cannot replicate what running through Cinemag transformers and real tubes can do IMO. I like it so much I have another on back order from a month ago which is finally arriving tomorrow.

    Here is an SOS review where they put the Warm Audio EQP-WA up against an actual Pultec EQP-1A in a studio comparison, and compare it to plug-in versions of Pultecs.
    http://www.soundonsound.com/reviews/warm-audio-eqp-wa

    Lol...thats' what I equate multiple plug-ins doing on a track ...chug-alugging down the track like a train...chug-alug....chug-alug....chug-alug....dragging down your cpu performance and adding latency to a mix....add to that the multiplier effect by the number of tracks loaded with plug-ins as well:D
     
    kmetal likes this.
  13. Sean G

    Sean G Well-Known Member

    Location:
    Sydney, Australia
    Don't get me wrong, there are some really good plug-ins out there, and some are pretty close to the hardware they emulate. There is a trade-off with using anaolg hardware as well when you are coming out of the box and then going back in again...thats why IMO you want to have your hardware going in as part of your chain or in the middle between 2 DAWs...no going back into the same box as guys here like @audiokid & @Boswell will tell you.

    There will always be a degree of degredation of the audio signal...thats a given and impossible to avoid. Its how you manage and minimise it as much as possible that which makes the difference.
     
  14. Boswell

    Boswell Moderator Distinguished Member

    Location:
    UK
    Home Page:
    Gigabit ethernet gets its name from the propagation rate of the measured unit. "Giga" = 10^9 and "bit" = bit, not byte. So the rate on the ethernet cable is 10^9 bits/sec or 1,000,000,000 b/s. This corresponds to 125,000,000 bytes/sec or 125MB/s. Note the capital B when referring to bytes and the lower case b when referring to bits.

    This is the rate that the bits within a packet of information would travel. Given that there will be multiple layers of wrappers round each packet and also gaps between packets, the end-to-end data rate of the payload could well be less than half the maximum bit rate of the transmission medium.

    One of the difficulties in using ethernet as a digital audio transmission medium in a multipoint network is that the underlying hardware offers no guarantee (a) of the end-to-end transmission time, (b) packets will arrive in the order in which they were sent, due to being routed on a per-packet basis, (c) a packet will arrive at all and (d) a packet will arrive uncorrupted. Because of issues (b) - (d), one of the higher protocol layers takes care that a long message can be assembled correctly from shorter packets, often involving re-transmission of lost or corrupted packets. All this bodes badly for real-time audio, but is fine for transmission of audio data files. These problems do not apply to point-to-point ethernet links where there is no other traffic.
     
  15. Brother Junk

    Brother Junk Active Member

    I had totally forgotten about this. And the capital B thing. I never set up networks, nor has ethernet ever been an option that I was concerned about.

    So, is this why the studios I've seen use the externals? It would be considered point to point I think...

    **Edit, I'm an idiot. They are doing it for the portability.

    The question arose bc externals are expensive, and they have older Mac Towers like mine in some rooms. You could buy a standard drive for half the cost of the externals they are using...maybe even less than half.

    So why the external route? I asked the RE just to see if he knew and he said the owner takes them home a lot.

    Question asked and answered...hoorah! That one has been bugging me for a long time.
     
  16. dvdhawk

    dvdhawk Well-Known Member

    @DonnyThompson

    This could probably be a separate thread too. But to give my answer to your question, I think you and I have a similar views on this.

    If someone is getting good results, and having some success using a particular approach - I'm all for it, whatever works for you. The SOS guy probably acquired one widget at a time and applied them on top of what (one would hope) was a pretty quality recording to begin with -given the level of gear and expertise. Each new plug-in probably gives it something he finds .1% more pleasing to his ear. I would hope he doesn't need them for grand sweeping adjustments, or to compensate for poor tracking.

    I try to use plug-ins very sparingly, but like a lot of you I usually have a pretty clear vision of where the mix is going to end up when I'm tracking - so I don't hesitate to print EQ, or even modest compression if I know that's going to stick. We all know that you can have your kick, snare, hi-hat, and bass guitar forming the absolute perfect pocket in the mix, but if you solo'ed any one of them they might (as @Kurt Foster would say) 'sound like ass'. For me, it's always better and more efficient in the end, to spend an hour trying different mics and find the sweet spot to aim them, versus fighting the mix every hour after that. Most of the tracks, I might not need any EQ on them unless it's for a specific effect in a specific song. Better signal in -> better signal out. Garbage in -> plug-ins -> filtered garbage out. (no matter how many times the folks on the ISS filter the water…. they're still drinking urine).

    That being said I do routinely use plug-ins as needed, primarily for EQ, compression, delay, and reverb. I'm always mindful that there's going to be a trade-off when algorithms are involved. Computational error, even if it's usually not noticeable, is sure to leave a cumulative pile of artifacts if you overdo it.

    As far as the plug-ins themselves, I'm under no illusion that a $50 - $300 plug-in can perfectly emulate every nuance of a $30,000 piece of hardware, but that doesn't mean they're of no value. And as it's been said before, no two pieces of hardware are truly identical either. I've never had my hands on a Fairchild or Pultec, so how would I know? All I know for sure is that I like what a BF LA-2A plug-in sounds like and use it more than the stock compressor. I like the Pultec EQ plug-in that I have, and I use it in certain situations, but less often than the stock parametric in StudioOne.

    I've personally been doing a version of the decoupled DAW thing for a long time when a project merits it. I have a buddy with some upscale hardware, and I do the editing / mixing ITB, and we pass that stereo mix in realtime through his rack hardware and record the resulting 2-track on a separate DAW at 44.1kHz. The capture DAW will usually have a limiter on the inputs, but basically we're setting levels as if we were going to DAT, or any other 2-track recorder. Ideally, we won't need to nudge any levels once it's been captured into the second DAW.

    The core piece of hardware in that process being my buddy's Avalon 747. I haven't found anything yet that doesn't sound noticeably better just by virtue of passing through it - even before you engage any of its functionality. If it's from a cold start, you do have to let it warm-up for 30 minutes or so, but then it's rock-steady after that. The tubes give the sound instant gravity, the tube compression circuitry is great for what we do. It's not overly dense or dark, but I can see where some might not like it for classical music. Luckily, we're not recording the Frogtown Philharmonic. If you haven't used a 747 you might not believe the icing on the cake is the 6-band graphic EQ. The center-frequencies, the Q, and the amount of cut/boost of each band have been carefully tailored individually (by someone with exquisite taste) so that each band is perfect and incredibly musical. You can sweeten the track, you can completely change the character of the track with radical settings, but you cannot ruin a track (even if you're trying to for sake experiment) with the stupidest comb-filtery looking 3-up / 3-down EQ settings you can think of. The character will change, but the mix will not come undone on anything we've tried.
     
    kmetal and DonnyThompson like this.
  17. bouldersound

    bouldersound Real guitars are for old people. Well-Known Member

    Location:
    Boulder, Colorado
    A decent compressor plugin is way better than a run of the mill analog compressor. Really high end hardware compressors do things that can be hard to emulate digitally. Actually, all compressors to things that are hard to emulate, but what normal compressors do isn't worth emulating.
     
    audiokid likes this.
  18. audiokid

    audiokid Chris Well-Known Member

    Location:
    BC, Canada
    Home Page:
    +1


    To add: I have owned some of the best and I wouldn't waste a dime on any hardware compression when it comes to hybrid mixing or mastering now. They are all a complete waste of money and time.
    ITB is 100% better in all respects including being able to side-chain better than any hardware comp can ever do. Which is where it all comes of age.
    Tracking though, love them. Especially the down right dirty UA tube and tranny stuff.
    Clean comps, ITB is once again, superior.

    To my ears... The higher end the hardware, the more you realize digital is better.
     
    kmetal likes this.
  19. Kurt Foster

    Kurt Foster Distinguished Member

    Location:
    77 Sunset Lane.
    or the better the hardware the better anything sounds. that's a no brainer. i don't believe it can be attributed to solely digital however. i just don't agree. digital is fine. it works. but it's not better. just different. perhaps you like it more and that's fine. there's nothing digital has to offer that analog won't do as far as actual documentation of a performance. digital is easier to edit. that's the only advantage i can see. but then, i know how to pull down a fader at the end of a track. :ROFLMAO:
     
    kmetal likes this.
  20. audiokid

    audiokid Chris Well-Known Member

    Location:
    BC, Canada
    Home Page:
    +1 for tracking.
    -1 for mixing or mastering.
    :cool:

    Once ITB, stay ITB. the love of analog compression lives in tracking and ends at mixing. You won't see money on analog mixing or mastering compressors ever again. For front end though, love them :love::love::love: so we are both going to be smiling on that.

    Until you get up to great conversion, I can't say I blame you but once you have the best of both worlds it goes like this....

    Digital compression is extremely accurate and very fluid.
    Digital compressors do all sorts of functions from extremely subtle to broad stroke aggressive impact, brick-wall limiting to surgical de-essing and triggering other freq's in the most creative and dynamic, or delicate ways, no analog compressor could ever compete.

    ... and stereo digital compression rocks. :D
     

Share This Page