1. Register NOW and become part of this fantastic knowledge base forum! This message will go away once you have registered.

Audio export to .wav. Is there a better way?

Discussion in 'Recording' started by Tony2Tone, Aug 6, 2006.

  1. Tony2Tone

    Tony2Tone Guest

    I'd like to ask what is the best way to export audio tracks from programs into .wav files while maintaining the highest audio quality.

    I'm tracking in Cubase SX 3 and have been using the EXPORT function to create .wav file mixes of each individual track but is there a better way that keeps the audio as pure as possible?

    I've read about using plugins such as TapeIt, which are plugin inserts that write to a .wav file in real time.

    I need to export all audio tracks to take to be mixed and don't want to compromise audio quality.

    Any advice?

    Thanks
     
  2. RemyRAD

    RemyRAD Guest

    Today is your lucky day! Ignorance is bliss.

    First, your .wav file is the highest quality file type to save to. You can however select certain parameters that can both destroy or enhance your completed masterpiece.

    You work with Cubase SX3 and you have been doing everything correct in using your "export" function to create your stereo ".wav" file mixes. I would hope that when you track, that you record at no less than 16 bit, 44.1kHz sampling rate? Most average consumer oriented multi-track recording interfaces offer no less quality than CD-quality which is what I have just indicated. There are however newer more high-resolution analog to digital converters and digital to analog converters that offer sampling frequencies up to 192kHz at 24-bit resolution. Does that mean that if your software allows you to export your tracks to 192kHz 24-bit that you should?? The answer can be both yes and no. If you have recorded everything at 16 bit 44.1kHz, there is no actual advantage to up convert or " transcode" your 16-bit 44.1kHz sampled recordings to the higher resolution. There are technical advantages to doing so if you understand how to make the most of that up conversion but it is actually not increasing the resolution of your original tracks. It can however give you greater latitude for your mix. The only hitch in the get along is, you still want CDs to be 16-bit 44.1kHz otherwise they won't play in a conventional CD player, then why not mix down to 16-bit 44.1kHz, so that you can create CDs without any further conversions necessary. Some folks like myself will mix down to 16 bit 44.1kHz so that they can be directly transferred to CD. Other people will want their mixed master's tube is the highest resolution possible so they will mix down at 24 bits at 192kHz sampling. Later they will have to down convert that to 16 bit 44.1kHz so CDs can be produced. Everybody does it differently to suit their own needs. You'll have to discover what works best for you.

    Never but never record or master anything to .MP3, unless it is for just a reference to be e-mailed to somebody. To be used for a memory chip music player or iPod. ".MP3" can be used when you are collaborating with musicians around the world via the Internet which is becoming more and more popular daily.

    .wav files are all written in real-time. So the plug-in that you believe is called tapeit makes no sense since you are already recording in real time to ".wav" files. So what you think you have read about is about as reliable a piece of information as being indicated to place a firecracker up your butt and light it because it's good for a laugh? It's incorrect.

    When you mix down all of your tracks to a stereo .wav file, you are not compromising audio quality unless you are reducing the bit rate or sampling frequency. So do not select settings such as "8-bit 22.05kHz", "16-bit 22.05kHz". That is a very low quality setting and should never be considered for any kind of serious audio work. if you're not multitrack interface and software allow for 24-bit, 96kHz or 192kHz, you may want to consider cutting all your tracks at that highest resolution available. When you mix down, you mix down to the same high-resolution format. After that, you will trans-code or downconvert that 216-bit 44.1kHz for a CDDA, audio CD release. I find that with reasonable equipment and proper recording techniques, 16-bit 44.1kHz across-the-board is quite adequate but for a novice who wants the highest quality they may be capable of producing, a high-resolution recording system certainly couldn't hurt.

    The reason many people like 24-bit recording now is because it offers a substantially larger dynamic range than 16-bit. 16-bit only allows for a 96 DB dynamic range whereas 24-bit recording allows for a greater than 120 DB dynamic range. But due to inexperience and technical misunderstandings, a 24-bit recording can be overloaded and distorted as easily and quickly as a 16-bit recording. How could that be? You can't record louder on a 24-bit systems, any louder than on a 16-bit system but you can't record quieter on a 24-bit system then you can I 16-bit system. So tell me, when was the last time you heard a pianissimi of some instrument in the middle of a rock-and-roll song drop 96 DB into the background and then back to full volume? Right, I didn't think so. So you want the highest resolution? Do you also never settle for driving in anything but a Rolls-Royce? A Chevrolet would just be so inadequate and embarrassing.

    I hope you used a Neve console to record all of your tracks because using anything else wouldn't be as worthy and certainly not have the ability to deliver the resolution, sweetness and roundness of sound that an old Vintage console that has recorded most every rock and roll it known to man has been recorded upon. And so if you use something like a Mackie or a Barringer to record your project, that's not high-resolution and therefore it probably means your music is not worth listening to?

    I hope you know I'm only trying to pull your bit rate?
    Ms. Remy Ann David
     
  3. sheet

    sheet Well-Known Member

    1. There are differences in bounces between manufacturers, no matter what file format is used. I don't care what anyone says, there IS an audible difference in some programs. I have seen a difference in file size too between manufacturers.
    2. There are also differences in dither...BIG TIME DIFFERENCES between software manufacturers. There are some blind listening test CDs out there. Find them. Very interesting. Just as converters mess with the sound, so does the dither. There are two that do the least damage and Cubase ain't it.
    3. For file transfers, I recording in Broadcast Wave format, unless I am sending AIFF files to the Avid guys. Broadcast Wave time stamps the audio. It will be the same "everytime". This is very useful if you are working in multiple studios, on multiple DAWs, or needing to add tracks later.
     
  4. dterry

    dterry Active Member

    I'll guess you've already addressed this, but....Outside of just burning your recorded wav files to data disc, you should just export to broadcast wave at 24-bit and whatever sample rate you are using. You could use 32-bit float if the app you are transferring to supports it, but that's usually overkill in terms of file sizes. Broadcast wave just gives you the advantage of time stamping the file, if the DAW you will be using supports placing them at their origin. Probably safer to bounce from time 0.

    The only downside to exporting tracks with Cubase SX is the lack of batch processing - solo and export one at a time is the only way. If you are just bouncing recorded tracks, no plugins, no panning, no level changes - just set all your faders to unity and you should get the same thing you started with (bounce mono tracks to mono; stereo to stereo; of course).

    The differences between DAW bounces has to do with what someone does in the mix, pan law, etc. Lynn Fuston's DAWSUM sampler put that argument to rest. I checked personally - mixes bounced from Nuendo, ProTools HD, Fairlight, Pyramix, and Samplitude all cancelled with each other to complete null (I've also compared several native DAWs on my own). These were unity gain tracks, bounced/exported at unity gain.
     
  5. JoeH

    JoeH Well-Known Member

    I've heard the same thing about differences with some software programs in terms of sonic quality in the final mixdown. I didn't say it, but others claim it's so, that Samplitude/Sequioa sounds among the best of them when bouncing from multitrack to stereo wav files. Something about the software's summing capability? (I'm of course happy about that, since I use it all the time....)

    As for the benefits of 24 bits - these days it's become the preferred method to use during production; there's simply more detail available for better crossfades, fadeouts, reverb tails, and of course dynamic range. One can indeed distort a 24 bit recording as easily as a 16 bit recording, but there's more overall internal headroom when working with multiple tracks, EQ and so on.

    The conventional wisdom thesedays is to record all raw tracks at 24 bits, work with that bit-depth throughout the production, and then bounce to 16/44 for CD, using the best dithering you can get your hands on. (I use PowR-3)
     
  6. dterry

    dterry Active Member

    Lynn Fuston tested the summing/bounce theory with a lot of DAWs and put out a DAWSUM sampler. I tested the majors as listed above - they all cancelled to their full 24 bits.

    I also use Nuendo and Sequoia and have tested this myself. There is no difference between most DAWs in summing unless one is coded incorrectly, or unique characteristics are included in the process. The differences do come into play with pan law, fader taper (Samplitude/Sequoia seem to be exponential, at least in the way the GUI operates, vs. a more linear taper in Nuendo), and of course plugins, which throw the whole debate out the window. 2+2 is 4, even in the digital realm. ;-)

    The difficulty in such testing comes when people just pull up a few tracks and say, "wow, this sound better!". Better? Different? How? Was the test identical? How much time elapsed between hearing the two DAWs - a minute, an hour, a month?

    At first with Lynn's sampler I thought there was a minute difference in imaging and depth between a couple of DAWs, but then realized the bounces were off by as much as 0.6dB. Once I leveled them off (in Sequoia), I couldn't hear a discernable difference. Phase cancellation confirmed there was none. One other difference in Sequoia/Samplitude is a form of soft limiting in the mix engine - you can drive it into the red and not get harsh clipping as with most DAWs (Paris had this too). Granted this isn't necessarily good practice, but it can be useful, or at least a safety net of sorts if you don't have a good limiter on the main at the time.

    I also use POW-r 3, as well as Cranesong's Analog dither (probably my favorite), though POW-r dither is very good also.

    Regards,
    Dedric
     
  7. Tony2Tone

    Tony2Tone Guest

    Bouncing from Cubase SX

     
  8. RemyRAD

    RemyRAD Guest

    I've heard that if you take a green magic marker and color the shiny side of your CDs, the sound will be better! NOT!

    You know you can get pregnant when a boy kisses you. NOT!

    You can make free long-distance telephone calls by whistling into the telephone. NOT! Any more like the old days......I loved Captain Crunch!

    If you fart enough, you can power your car on the methane. Ooooooopppsy Pooooopsy!
    Ms. Remy Ann David
     
  9. dterry

    dterry Active Member

    Paul, there is no difference in the audio engines of SX3 and Nuendo 3.
     

Share This Page