Cause of pops on CDRs?

Discussion in 'Audio Hardware' started by Patrick M, Mar 14, 2002.

Thread Status:
Not open for further replies.
  1. Patrick M

    Patrick M Subgenius Thread Starter

    Location:
    US
    That's the best subject line I could come up with.

    I'm a branch for a trading tree for a live show. I got my two discs in the mail from the seed a couple of days ago.

    Here's the deal: pops between tracks on certain tracks. Seems to mainly in the left channel.

    There are no two second gaps or anything. I don't think it's a TAO problem.

    This show came from a webcast, so it should be one continuous feed. IOW, I don't think the pop is from butting two .wavs up to each other.

    So what's going on? The guy who burned it said he has noticed random pops on the originals he has. On mine, the pops are in the same place, between the same tracks (but not all tracks), consistently. I've tried two CDPs already with the same results

    Last night, I extracted some of the tracks with EAC and pulled them up in CEPro. The pop is right there in the .wav. Here's an example:

    http://sector7g.org/sandbox/dogman-whatisthis-pop.wav

    (It's only 75k.)

    Any ideas? I don't want to pass these on to my 4 leaves with the pops intact, but I have a feeling if the guy above me burns me another copy, they will have the same problems.
     
  2. Sckott

    Sckott Hand Tighten Only.

    Location:
    South Plymouth, Ma
    Actually, it IS a TAO problem, Patman. Maybe not directly.

    Really depends on just how the disc is duplicated. A lot of people use EZCD creator to "copy" a disc, some manually rip it to the HD and the .WAV files will have errors on the 1st curve (so slightly) on certain waves, or the CDROM will not stream a beginning data stream at absolute 0 (or it will when it's not supposed to).

    Meaning, well, there's lots of factors here, but if the wav at the end at track one doesn't connect EXACTLY to the beginning of track two, you have a gap that's not sewn tight. The .01 second gap will create a pop.

    Using EAC would help if your original is good to begin with, but if it's not, yep, gotta cherry-pick those clicks where the beg/end places don't match up because either ripping or CDR software refuses to pair and sew seperated wavs together.

    Another reason why EAC does so well. One seamless wav, and a cue with allll the layout.


    Here's the place where the two wavs don't meet:[​IMG]

    On your end, Pat, you're gonna have to pluck. Oh, F***k.

    Notice how the beginning/end of the wav WANTS to start at absolute digital 0? Hmm.... that ain't right, but that's what's happening.


    Here's a little embellishment:

    [​IMG]

    Imagine you're driving in a big Harley and Leatherface is coming to get you (or it's Peter Mew on his way to negotiate some mastering costs and you're trying to get there before he does).

    Yeah, this is the bump. It might sting a little :p
     
  3. Grant

    Grant Life is a rock, but the radio rolled me!

    Not only could TAO be the problem but the DC offset may be waay off from 0.

    I suggest you load each track into Cool Edit, remove the DC bias, turn the snapping (fine) on (right click on the ruler), choose the Display Time Format as Compact Disc 75 fps, edit out any clicks, and burn a new copy with DAO only.
     
  4. Patrick M

    Patrick M Subgenius Thread Starter

    Location:
    US
    Thanks for the replies.

    Now the practical questions:

    1) Why would only some tracks be affected?

    2) Are we all in consensus that this is a burn problem?

    3) If 2) is yes, isn't it the responsibility of the dude above me to get me clean copies and not me to fix his errors?

    OK, #3 is just opinion, but what do you think?
     
  5. Sckott

    Sckott Hand Tighten Only.

    Location:
    South Plymouth, Ma
    What one would have to find is, what's happening at the level above your's.

    Roxio and other CDR sofware products DO this sort of thing, depending on how it's done. You can make a burrito a million ways, but it's still a burrito. You can make a CDR track-by-track duplication (Roxio, CDRwin) or use something LIKE EAC and truely have DAO duplications and burnings.

    Sounds like you'd need to whip some education on your fellow trader-friend. For you Patrick, I'm sure you know how to correct it. Pain in de butt.

    This phenomenon happens when the CDR software is choosing .wavs that normally WOULD sew together, but the software, even at DAO, treats each track like a seperate stop/go product. It's rarely the ripping, unless it's Mp3 sourced.

    But if it WERE Mp3 sourced, the gaps would be much wider. Every Mp3 encoder starts wavs at 0 for about 1 or 1.5 milliseconds. It's not a whiplash-crack.

    This is CDR software that isn't playing nice. It's happened to me a few times, trying to a LP rip, then seperating tracks. I gave it to EZCD, and even though DAO is used, it does not sew the tracks perfectly together. Like I said, it almost treats each wav like seperate products, and occasionally, the stream stops for a millisecond.

    Suck!

    The question is, would you look a gift horse in the mouth? Truely your buddy may not be into tweeking. Some people copy CDRs like "duh" when actually there is a difference in how you do it. Trust me on that, and about 2 dumpsterfulls of blasted CDRs that didn't make "the cut"! ;)
     
  6. Patrick M

    Patrick M Subgenius Thread Starter

    Location:
    US
    Speaking of the gift horse in the mouth thing

    I think myself and another guy serving as a branch are starting to piss off the guy who started the tree. He even said something like, "I bet you bitch when the toilet tissue is hung the wrong way."

    Here's how I look at it:

    People have different standards for what they consider acceptable. I got 6 CDRs in trade recently that were done on a standalone burner. All the live shows were botched. Instead of just a "pop," there were places where little snippets of the .wav repeated at the beginning and end of tracks. In some cases, notes were missing due to these "hiccups." That, to me, is unacceptable. To the guy who burned them, the glitches didn't bother him at all. I wound up re-trading to get clean copies of all the live shows he sent me but one. On one of them, I extracted everything and fixed it by hand. A waste of my time.

    So, if you are conscientious about how you do your burns, you make the people who don't want to hear pops between tracks happy and the people who don't care happy. If you're half-assed about it, you wind up irritating the people who do care.

    With EAC being totally free, I think it should be a no brainer. Thoughts?

    Also, because this is a tree designed to get this show out to everyone who wants it as quickly as possible, I'd like the show to be consistently correct and the same for every person who gets it. So, for future trading purposes, one person doesn't have version 1.0a with pops while someone else has the re-EQed and pop-free 1.0b version.

    Further, because this is the first King's X tree ever, I want to set some standards and develop a good precedent for future trees. Assuming we have any after this mess.

    Anyone out there who's worked on a lot of trees care to chime in with the voice of experience?
     
  7. Sckott

    Sckott Hand Tighten Only.

    Location:
    South Plymouth, Ma
    Generally, I agree with you, Patrick. Unfortunately, I say just milk your trading circle and correct things yourself. You realize a lot of people wish they could, and a lot of people don't.

    This may come as a suprise to you, and you (and Luke especially) will HATE this terminology, but even though a lot of SH'ers can do it, you're still the upper 1%'er. You have NO IDEA how many people flounder just getting those CDRs done at all. It doesn't help the situation, but I say just enjoy the fact that you have the tools to do it, and so be it.

    "I don't make monkeys, I just train em!" - Pee Wee Herman
     
  8. Craig

    Craig (unspecified) Staff

    Location:
    North of Seattle
    Sckott's right. Just as in the rest of life there is wide range of what is acceptable when it comes to making CDR's. I've seen several trading lists get to into looooong discussions about trading/treeing Track-At-Once discs, with both extremes and everyone in between represented. If I get something I can fix, I fix it before passing it on. If not I junk it and try to get a better copy later. If it seems like a defective copy I'll question the person I got the disc from, otherwise I just cut my losses.

    Just remember it takes all kinds to make the world go 'round. I know a fellow that was pleased as punch with the CDR copy he'd made of a friend's Sgt Pepper disc. It was TAO complete with two-second gaps. OUCH!!!, but it didn't bother him at all.

    -Craig
     
  9. Patrick M

    Patrick M Subgenius Thread Starter

    Location:
    US
    Details

    OK, turns out the pops are on the 'original' seed discs. So even the copies the person at the top of the tree has have these problems. He just didn't notice till I pointed them out. Hmmm.

    Also, the guy burning copies for the branches used EZ CD Creator Platinum to image to disk. He's also blaming the pops on "a power surge on the laser during the burn."

    So help me out here: If I just go in there and delete the offensive sections so the .wavs all butt up together nicely, will that make everything peachy?

    Or is there a better/easier way?
     
  10. FabFourFan

    FabFourFan Senior Member

    Location:
    Philadelphia
    Just an odd thought -

    If the CD was (at some point) sourced from separate tracks,
    and if a leading track does not completely fill out the last sector
    (= 1/75th sec = 2352 bytes = 588 samples/track x 2 bytes/sample x 2 tracks),
    then there will _not_ be a clean transition to the next track.

    When this happens, there will be between 1 and 587 samples which contain all 0's,
    before the trailing track starts providing its data.

    This 'surprise' drop to 0 can make quite an ugly 'snap' between tracks!
    Which is why every CD track has to contain a multiple of 2352 bytes of data!!!

    What confuses me in your particular case is why the drop to 0 seems to be for such a short time,
    and why the pop sounds so minor? Was it some other 'joining' mistake??


    I hope this helps!
     
  11. Sckott

    Sckott Hand Tighten Only.

    Location:
    South Plymouth, Ma
    Re: Details

    Nope. Hey, but you have Cool Edit, and any of their flavors will do smoothing. The trick is, just try to connect the two by approximating where they'll connect, and Cool Edit will make everything smooth together. If you make a boo-boo, Ctrl-z is undo.

    Then, if you used EAC, you need only to save the wav after you've knocked out all the bumps. Save and burn. Done deal. Good luck.
     
  12. Grant

    Grant Life is a rock, but the radio rolled me!

    All in all, I believe it should be the responsibility of the seed to do the hard work, to provide the best source material for distrubution.

    You have to have someone who cares at the top. It is appalling what some people regard as quality. It is worse that some people has no regard for it whatsoever! I guess they can look past the flaws and get to the music. But I can't see gaps in a concert recording!
     
  13. Patrick M

    Patrick M Subgenius Thread Starter

    Location:
    US
    Grant, how do I remove DC bias? I know it's an option with a fade, amplify, and normalize, but I don't want to fade, amplify, or normalize. TIA.
     
  14. Grant

    Grant Life is a rock, but the radio rolled me!

    There should be a preset there called something like "Center Wave". That'll do it. Just make sure you do it before you do anything else.

    I also suggest you convert the files to 32-bit float first before ANYTHING, then when done, dither back down to 16-bit, maybe with noise shaping to move the dither noise to a more desirable area of the frequency spectrum.

    Cheers
     
  15. Patrick M

    Patrick M Subgenius Thread Starter

    Location:
    US
    OK, two more things and I'll drop this thread.

    I ran statistics on the entire .wav file (i.e., all of one disk) and the DC Offset was 0%. Ergo, I assume I don't need to do any .wav centering. On some of the individual files (i.e., tracks), the DC Offset is -0.001%, which I assume is essentially irrelevant.

    I'm curious: if you take a 16/44.1 file and open it as 32/44.1, do your edits, then save it back to 16/44.1 *without* dithering, do you actually lose any information? I assume not.

    Oh yeah, and if anyone wants to help me with a phase problem on this CD, I'll start a new thread. :D
     
  16. Grant

    Grant Life is a rock, but the radio rolled me!

    Anytime you go down to 16-bit you always want to use dithering to reduce low-level distortion. The distortion sounds really grainy and id very audiable! Dither is random white noise but much preferable to the distortion I just described. That is why we can use noise shaping, to move that dither noise around to less sensitive areas of the musical spectrum.
     
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine