The Mysteries of the Quiet CDR

Discussion in 'Mastering' started by David French, Jun 13, 2003.

  1. David French

    David French Well-Known Member

    Could someone please explain why a CDR that I burn on my PC, even with heavy limiting, always sounds a few dB quieter than a commercial release? Will a Masterlink burn a commercial level CD?
     
  2. joe lambert

    joe lambert Distinguished Member

    The simple answer is because relative volume is different than peak volume. And it's the proper EQ,Compression/limiting that gives you the big sound (average level) not just cranking the l1 or l2.
     
  3. David French

    David French Well-Known Member

    So you're saying that if I do a bit for bit extraction of a commercial CD track, then burn the resultant .wav file onto a CDR with no processing whatsoever, the CDR will be just as loud as the original CD?
     
  4. falkon2

    falkon2 Well-Known Member

    You got that right. Welcome to the wonderful world of signal processing. :D
     
  5. David French

    David French Well-Known Member

    Forgive me, but do you gentlemen realize that assuming I don't know what i'm doing is insulting? I love RO because it is a place where you can ask honest questions without fear of being insulted by a disgruntled AE like at ProSoundWeb :roll: . I am no idiot. I have made careful study of many many commercial tracks with respect to peak level, RMS level, frequency distribution, and stereo distribution. I strive to match these statistics as much as I can when preparing to burn a CD. I do not claim to be a mastering engineer in any sense of the word and I greatly appreciate the talents of a good ME, but differenced in RMS and frequency distribution alone could not account for the differences in level that I am experiencing; the snares are even louder. I cannot believe that the "proper EQ, Compression/Limiting" is responsible for individual snare hits being louder on a commercial CD than on my burned CDRs. It seems to me that there is something inherently quieter about a CDR than a commercial CD. In fact, I have a vague memory of a discussion right here at RO some time ago about burning CDs without the "-3dB drop". I searched for it, but couldn't find it. I believe Bill Roberts was involved in the discussion. Perhaps he could shed some light on the situation?
     
  6. falkon2

    falkon2 Well-Known Member

    If you were referring to my post, forgive me. No slight was intended. In fact, if you got down and dirty with the facts, I probably can claim the title of "Member with absolute least hands-on experience" on RO.

    What I DO know is - the fact is, CDs are all digital - that is, 1's and 0's. The information is interpreted by the player the instant it's played, and is not when the CD-R is burned. This means that if the CD player's D/A was fed with the exact same bits from the reading lens, be it a commercial silverback CD or a cheap USD0.20 CDR in the tray, the volume will be the same.

    Try a simple test - I don't claim to have done it, but I would be very surprised if the results aren't what I hope they would be.
    Get a commercial CD, rip a track out with a reliable ripping program like EAC to make sure that you get a bit-for-bit copy of the CD (brief tutorial found here), then burn the resulting .wav file.

    Play the two CDs one after another, or better yet, get someone to do a blind A/B with... The volumes SHOULD be the same. (though after reading your post, I'm not 100% sure anymore)
     
  7. falkon2

    falkon2 Well-Known Member

    P.S. I think I need to write this because my post sounded a little too lordy for my own liking. I'm not trying to "command" anyone to try out the test, like I already knew the answers. I really want to know the results, and I can't currently perform this test on my own because I am stuck with a laptop and rock-bottom monitoring systems for the forseeable future. If my assumptions about the volume were false, I REALLY want to know!

    Again, no slight was intended.
     
  8. doulos21

    doulos21 Member

    im however not as nice cheap burners = lower sound get a decent burner made for audio should be like an extra 100 bucks at most then rip an entire cd and load the thing through a true stero pre so you know the levels are the same rec the cd 2 wav your riped version of one song and the commercial cd if there is any diffrence in volume ill eat my mic cord
     
  9. realdynamix

    realdynamix Well-Known Member

    :) Hi David! Let's use a snare hit as an example of what Joe said. In mixing, the snare hit is very fast, so fast that it's peak may not even show up on a VU meter by the time it is over. What would the average level of that snare be? I know you have been through all this, and understand it.

    If we used that hot snare as the absolute maximum point we can go, will there be room for all the other instruments to blend?

    Even if the snare is reduced in level into the mix, will it have the same desired impact?

    You say snare hits are louder on a commercially produced CD. How are you getting your instrument and vocal music levels up there in average to work effectively with the snare?

    If you limit the snare, it doesn't sound like a snare anymore, and if you attempt to get a hot snare sound in a mix, and go to compress and limit that mix, the snare is going to control all your dynamics, and defeat your attempt at average loudness.
    IMHO it has nothing to do with burning except, some stand alone recorders might offer a bit of protection in the metering.

    Any help?

    --Rick
     
  10. falkon2

    falkon2 Well-Known Member

    Take a deep breath! No, really!
    What exactly do you mean by cheap burners giving "lower sound"? Lower volume? Lower quality?
     
  11. horowizard

    horowizard Guest

    Greetings All,

    Perhaps my take on this subject will shed some light toward someplace where you might find some answers to the baffling question of "Why don't my CDs sound as loud as the commercially released ones?". The first thing you should take into account overall is the mix. How similiar is the material that you are attempting to capture for the entire world to hear to the material that you are comparing it to? You may noitice that many comercial CDs also lack low end or bass in the mix, while others have tons of it. If your mix is bass heavy when compared to the commercial CD, you are going to have to do a lot of work to get the same slamming volume levels. It can be done if you are willing to take a few extra steps in the process. If limiters seem to be squashing your sound into something no longer recognizable to your original concept, try mixing some of the unaffected track in with it. Don't take it all the way. Do some mastering compression on your 2 Mix Bus. Then in the Mastering Program (I assume all you guys are working in the Digital Domain, right?) use a Multiband Compressor, which breaks the sound up into different frequency bands and compresses each one differently. When dialed in properly, a Multiband Compressor can make the sound literally leap out of the speakers without adding any EQ. However, be careful to apply it properly so it suits the style and mood of the track you are working on. You wouldn't treat a Ballad in the same way you would treat a Disco song. You wouldn't serve up Salsa the same way you would present Rock. Remember, don't overcook it. Only do what you need.
    Still not loud enough? Bring the file from the Mastering Program back into the Editing program and try some more Limiting and Compression/Mastering Compression there. You will find by comparision that you are getting closer and closer in volume and yet still maintain the integrity of your original mix, if you do it this way, in small steps.
    On another note, some mention of burners and CD Writers should be discussed. There are so many different ones out there coupled with so many different programs and they all have their assets and liabilites. I won't mention specifics, but suffice to say that there are differences and you may have to experiment for a while before you hit upon the right combination that best suits your needs. You will also find that differnt writing speeds will give you different results. A stand alone audio CD burner may give you product that has more depth to the sound that the burner in your PC. I personally don't take much stock in the quality of sound that comes from using the burner mounted in your average run of the mill pedestrian PC. I have yet to hear a disc which came out of a PC that didn't sound like crap. Harsh words, yes, but I can only speak for myself. Take it as one man's opinion.
    As far as your DAWs are concerned, don't always flip on that "Auto Mastering" switch. It may give you the volume you are seeking, yet totally destroy your ability to understand the lyrics.
    Finally keep in mind that commercial CDs are mastered with equipment which is the finest of the fine. There are brick wall limiters out there that will knock your socks off, but let's face it, these toys are not cheap and usually out of the realm of the home recordist's budget. You get what you pay for folks, and you will not get that big sound without spending some big bucks.
    In the meanwhile, keep trying.
     
  12. Terry G

    Terry G Guest

    David, PM me, I will give you my email and the answers.
     
  13. falkon2

    falkon2 Well-Known Member

    This has been what's been baffling me for quite some time. Other than the jitter effect causing some bits to be misread during playback by the CD player on bad burns, how can a set of 1's and 0's burned onto one surface sound different from an identical copy on a different surface?

    Could anyone shed some light on this?
     
  14. Ethan Winer

    Ethan Winer Active Member

    David,

    > Could someone please explain why a CDR that I burn on my PC, even with heavy limiting, always sounds a few dB quieter than a commercial release? <

    I'm surprised nobody mentioned the best method of all for maximizing volume - using a program like PeakSlammer. Unlike limiting, which uses attack and release time constants and always adds a sound of its own, PeakSlammer and programs like it act on single cycles only. So you don't have to add an obnoxious amount of compression to increase the levels. Here's a link to the page that describes PeakSlammer:

    http://www.scrollworks.com/products/slammer/

    The key is it reduces the volume of any single cycle that exceeds a threshold you define. But it does so without clipping, so you can get a healthy increase in perceived level with no side effects. I typically go for 4 dB. of reduction, but you can go even higher than that before it starts to sound affected.

    --Ethan
     
  15. Ethan Winer

    Ethan Winer Active Member

    Falcon,

    > how can a set of 1's and 0's burned onto one surface sound different from an identical copy on a different surface? <

    You are absolutely correct, and even the effect of jitter is grossly overstated. Bits is bits, and nobody can tell the difference between one error-free CD and another in a blind test. I'm not saying all CD players or D/A converters are the same. But the raw data on the CDs surely is.

    --Ethan
     
  16. David French

    David French Well-Known Member

    Let me start by saying what a pleasure it is to have finally created a hot thread! From past experience, I was beginning to believe that I was a thread killer!

    falkon2,

    First, it's all good, bro. :(

    Rick Hammang,

    I think you are basically agreeing with Mr. Lambert, but you managed to do so in a respectful way, which I appreciate. What you said, however, seems a bit zen-like to me. It seems to me that all you did was present a kind of paradox. Maybe I just don't get it. :confused:

    horowizard,

    The material I am putting onto CD is quite similar to the material I am comparing it to. This is how I learn; I find a precedent for every track I do and try to match it as close as possible. I have a voracious appetite for learning audio and I am in the immitative stage of my education. I spend a great deal of time studying and emulating professional work.

    Ethan Winer,

    Is PeakSlammer truly inherently better than plugins like Waves L2, TCNative L, and Sony Oxford Inflator? And on the point of the different sounds from different CDs, I have always seen it your way as well, as logic would dictate, yet I have heard guys arguing over which CDR 'sounds' better! Quite strange.

    So gentlemen, let's keep going until we get to the bottom of this :c:
     
  17. Jon Best

    Jon Best Active Member

    Um, huh? If your burner's a piece of crap, then maybe you'll get skips and errors, but any volume difference is *not* going to be attributable to the CDR itself. If you load the tracks from a CD into a burning program, and burn them on a CDR, and then play the CDR in any CD player (that it works in- there are CD players that won't play CDR's well, but that doesn't have any bearing here), any volume difference you hear can only be due to a) your ripping/extracting hardware/software is ^#$%ed up, b) your burning software is ^#$%ed up, or (remotely) c) your playback hardware is ^#$%ed up in a really interesting way.

    AFA the 'true stereo pre' goes, I'm not sure what you mean. If you go out analog, go through some sort of pre, and then rerecord, you've invalidated the whole test- your levels could be anywhere.

    The point is, cheap burner or not, if you burn a data identical CDR and it actually plays in your player, you should not hear any volume difference. Slight tonal differences may be possible, but I think the jury's still out on that one. It's certainly never been proven, and it certainly won't be major.

     
  18. Jon Best

    Jon Best Active Member

    I really don't mean to be disrespectful, but I ain't buyin' it. This kind of statement bothers me, as it comes up so often, and there's very little base for it.

    If you burn a CDR on a half dozen different players, you will definitely end up with differing error rates.

    Your playback mechanism will either be able to correct them, or it will skip and pop.

    Nobody has ever, to my knowledge, done a successful double blind test that shows that anyone can tell the difference between data-identical disks, although there is *some* anecdotal evidence of slight differences.

     
  19. David French

    David French Well-Known Member

    Well, I emailed Terry G, and this is what he had to say:

    A very interesting new dimension to the debate.

    What do you all think of this?

    At this point, I thought it might be appropriate to bring up the wise words of Benjamin Franklin - "If you argue and rankle and contradict, you may achieve a victory sometimes; but it will be an empty victory because you will never get your opponent's good will". Although I have no authority here whatsoever, I would like to urge everyone to keep a cool head and to be respectful of eachother at all times. I remember several threads, recently a certain thread about the HR824s, that have really gotten out of hand because everyone has such strong beliefs. But aren't we after the truth, not to prove ourselves right? Please, be cool. :cool:
     
  20. falkon2

    falkon2 Well-Known Member

    Now THAT is an interesting turn of events.

    With this new info in mind, I did a simple test - I'm not too sure if I can draw these conclusions from it, but here goes anyway.


    A good way to test if two tracks are identical is to load them up side by side in a multitrack/DAW and invert phase on one. If they're identical, the result of playback will be silence. (if f(x) - g(x) = 0, f(x) = g(x))

    With that in mind, I performed a couple of burns/rerips and rip/reburn/rerips, and tested the end results against the originals.

    I generated white noise, normalized and compressed it till the level analyzers said -5dBRMS, then I burnt it and reripped. When loaded up into Cakewalk side by side, silence.

    I generated a normalized square wave, and the dB meters gave positive values (since rms is based on a normalized sine wave). I burnt, reripped, compared. Silence.

    I ripped a couple of commercial CDs with decent rms levels (-10.5 and -8 respectively). Reburnt, reripped, compared. Silence.

    It is totally possible in the third case (with the commercial CDs) that the ripping process was somehow influenced by the -2dB limit as well, and thus the first .wav that ended up on my hard drive was already crippled before being reburned and reripped. If that were true, then both .wavs would be identical, no question, and yet the -2dB limit could still be there.

    However, the first two tests yielded silence as well... which means after burning and reripping, the copy was identical. Now, if we think about this carefully, the reverse process might also be true. Could CD drives somehow gain back -2dB from CD-Rs burnt with that limit? That would imply that the data itself was unchanged, but there's a limit that goes into play during playback, rather than when the CD-R is burnt.


    Or it could simply be that the limit is an unfounded rumor.

    Discussion?
     

Share This Page