I am trying to understand 'distortions' introduced by poor tracking so that I can avoid them and get usable tracks. The way I see it is if a person can't listen to your song on their system without it distorting then how good the song or mix is is irrelevant since a person never gets to hear it.

This track was recorded in a church for the favorable effects of the sound of the room, but maybe it had some acoustical problems of its own as well.

This was recorded with two mics and one is panned hard to either side. No additional processing was done to either track. There is a large amount of hiss in the right channel which is due to me having to use a different mic than I planned to there (because someone else showed up without the mic), so try to ignore that.

Ok, I notice that when I burn this onto a CD, speakers will tend to 'distort', for lack of a better word, at certain spots. You can notice this particularly when he plays the harmonics. I just want to understand what is causing this and if it can be dealt with in a track like this that has already been recorded and more importantly, how to avoid it when tracking in the future.

A couple of people elsewhere told me that the track was just overly dynamic and needed compressed. One guy eq'd the track and compressed it for me, but I then burnt his version and the original onto a CD and the original handled better!

The worst sounding spots you will notie don't occur at the loudest spots in the recording. The thing doesn't have too much bass overall and so that's not the problem either. I am thinking it might have to do with the room enhancing certain fequencies too much?

I know this seems like a very basic thing and is probably obvious to most of you, but it's something I missed along the way and never learned. I figure going back and learning how to track to get a nice clean recording will help me finally get a good sounding finished mix. Some of the other songs that were to be on that same EP were pretty involved. Thanks.

What kind of mic did you use on this?? It's not so terrible that you can't listen to it, but you can definitely tell. Did you compress on the way in? I would throw a nice transparent compressor on each mic's channel, and have a second go at it. Nice sound by the way, and the hiss isn't as audible as i expected.

actually, now that i'm listening, it sounds like when you don't press the strings down hard enough, but this being a very good guitarist, i doubt that was the case, although you never know. Just a suggestion.

What is annoying is the sound of the bad pressed notes ( buzzing ), at least for me.

That's what i'm talking about. I don't hear any low end distortion. I am on computer speakers though.

Yep, even in that long speech of my first post, I forgot to include a disclaimer about the guitarist's performance. Usually people can't deal with his extra long 'artistic' pauses. Yes, there are some poorly fretted notes as well. I don't believe he had adequately prepepared, and in fact neither of us caught the fact that he forgot to play the intro he had made for the song! I wish this was the only cause of the buzzing I experiencd from TV and cheap CD player speakers, but I doubt it is.

They'd be a lot more likely to distort then(meaning you would be more likely to hear what I am talking about)... It seems nothing will distort my studio monitors.

I guess I need a better understanding of what causes speakers to distort as well. I have a paranoia that everything I mix/record distorts quicker (that is at lower volumes) than everyone else's stuff, even though I watch the overall eq and dynamics(though this particular clip is pretty dynamic). What amazes me more is to hear it more in something as simple as this vs. a full mix. Maybe it's just that you just notice it more when fewer things are playing?...

I'll post another clip from this session soon.

Thank you for listening.

Sores,

Thanks for the heads-up email, and I'm glad to comment.

> I notice that when I burn this onto a CD, speakers will tend to 'distort', for lack of a better word, at certain spots. You can notice this particularly when he plays the harmonics. <

If this is an acoustics issue, it's probably the room you're playing it back in because it sounds fine here. (Aside from the stuff you disclaimed above.) I do hear little crackles and other artifacts, but it sounds to me like those are probably due to rustling clothes or whatever. All of the harmonics sound clean, and looking at the waveform I don't see any obvious clipping or anything.

> I guess I need a better understanding of what causes speakers to distort as well. I have a paranoia that everything I mix/record distorts quicker (that is at lower volumes) than everyone else's stuff.<

This often occurs on material that has too much very low bass, like when someone boosts 50 Hz by 10 dB or more in an attempt to gain "fullness." But again, I hear none of that here, and the lowest note on an acoustic guitar is only 82.4 Hz anyway. What kind of speakers do you have? And how about the room you're listening in? If your room is resonating it can create the kind of effect you're describing. As an easy test you can play the track through headphones to remove the room and speakers from the chain.

--Ethan

Thanks for the reply Ethan.

It's good to know that there is really nothing wrong in the actual recording.

The important thing is how my recordings compare to others. If they all exhibited the same effects, I would probably attribute it to the room it is being played back in like you suggest. For this particular song, I was comparing to the Trans-siberian Orchestra's version of the same song on TV speakers in a dining room - a definite low end system, but maybe comparable to where others might be listening. I didn't notice near as much distortion with the other song. I do know that the other song does not have as loud of peaks and appeared to be compressed to a high degree(anyone know what was done on that song or have the song to listen to?) and it also uses a steel string guitar rather than a nylon string.

I also have played this and other songs back on a cheap little CD player in various locations. Now most everything distorts this playback device some but I was just thinking my recordings seemed to distort easier at lower volumes. Heavily limited modern recordings like Mushroom Head' s XIII distort pretty badly on this device but something like Elton John's Goodbye Yellow Brick Road can be turned up the whole way with minimal distortion.

I think that the clips in my two other related threads contain some unwanted room effects in the actual recording because you can see peaks in the files...

Thanks.

Do a frequency analysis and you'll see why right off the bat ....

the harmonic peaks are stacking up well bellow 87 Hz
you still have considerable levels at 46.7 to 31.2 Hz
you can EQ it out or bring into level with the rest of mix

great playing, really. get thoes fret notes down though. cut your nails on your fretting hand and hug that thing! nice tone though. I love a good classical guitar sound, but to me that would sound NATURAL. this sounds RECORDED. jeez, I hope that made sense. I want to hear the room in the recording, in other words. personally, im a huge fan of a little stereo delay. The john william bach suites have incredible classical nylon tone. just SIMPLE. its just a stereo configuration, and you can hear everything. now, another thing, this is being very technical, but I enjoy the sound of position sliding and stuff. not sure if there was any (I doubt in this peice) but your not getting 'that' sound. anyway, nice playin, and clean recording

Eyes,

> something like Elton John's Goodbye Yellow Brick Road can be turned up the whole way with minimal distortion. <

Yes, and to me that's one of the defining differences between amateur and pro recordings. I'm more in the amateur camp with mixing, but the "turn it up and see if it distorts" test is one good lesson I have learned to always use.

--Ethan

No, I'm not following you, actually. First you say you like a more 'natural' guitar sound and then you recommend using something like a delay. The only 'effects' added here are those of the mics themselves and the tubes in the preamp. You really can't hear the room sound in there? I can. If it wasn't for the hiss problem that I explained, I could also bring up the room some with some compression, if it was desired, but I thought if anything it was leaning more toward having too much already, compared to most things I hear. I don't have any good references for classical guitar solos, however, and I will try to check out the recordings you mentioned. I think you can hear some that string noise you want in there, and you can also hear my assistant opening the door prematurely toward the end of the piece Again, I did not play this but was the engineer, but I have passed the info on to the guitarist.

It's possible that the frequency analyzer in CEP isn't very accurate. I do make frequent use of it - I know I overuse it. What frequency analyzer would you recommend? I don't observe the behavior you describe. I notice a lot more energy around 200 Hz and less highs when the harmonics are played. When comparing to the guitar in Pink Floyd's "Goodbye Blue Sky", their's actually seems to have more energy below 30Hz, though ours may have slightly more from 40-80.

Exactly, so what are the pros then doing different? Or what would you do if he you did this test and it was distorting significantly? You think it's more in the mixing stage then? Further, on a song like this, where there is no mixing to be done, it can be handled in the mastering stage?

Thanks to all.

If someone could check out the other two related threads, we might be able to get a more complete picture of things.

I use the Adobe Audition 1.5 in the WinBox and Audio Analyzer in Linux

I'm posting a screen shot for you with some holds it's a log scale in case the numbers are a little hard to read... just save it and open in an image editor... windows or mac can open Gif native so you don't need any special software....

On really crappy speakers low volume distortion sets in at between 30 -20 Hz if the levels are right.... One of the trademarks of those Floyd and Elton John LPs (No the CDs because they have been remastered) was that you couldn't get any punch or bass unless you craked it up really loud. At which point the cones are vibrating fast enough that they can not reproduce the low end unless they have superiror responce times (or are really ridgid high watt types - not paper cones)

At low volume there is not enough current to cause distortion ect ...

This why so many HiFi in days had EQ to adjust bass ... and that gave way to solid state bass expanders and enhancers ect to push the low end up ... because it masked on purpose so that it would cause distortion in cheap speakers... and that's also the reason every comercial studio should have a pair or crap speakers to make sure that the track is audible to mass consumers ... ect ...

I hope it makes sence... English is not my primary language so excuse any mistakes in spelling or symantics, but I think it will become clear if you think about it. Anyway this is what I have been told. If anyone has more info or corrections I will happy to learn something new.


.Z

Try this version of your clip and see if it distort as much....

It's always better to have a low cut and hight cut when you record..... this would limit this kind of stuff.... This is just a instrument track so you can attack it but if there were more it would be harder....

If you record it right you save yourself much time later cleanining...

Also if you listen now you can hear the pick and the players fingers on the strings.... I cleaned up each channel by itself and now you can see the different frequency responce from the 2 different mic it more than I would have tought but not so extereme it actually give the clip some warmth...

ciao


Z

hi eyesore,

I ran your file of "Silent Night" through the SPL Transient Designer and the PsyQ devices in CreamWare Scope and tracked it using the Scope's VDAT. I then converted the file back to mp-3 using Sound Forge 6. Will it solve your problem? I don't know. Will it sound sound better to you? I don't know. I do know it sounds a little different.

The reason this file was here is because I am unable to upload files to Gearslutz, and I wanted to give you an idea of what the Transient Designer can do.

Let me know what you think. I took the file down after the weekend. My only attempt at posting a file here got 1 response: my own when I 'bumped' it :-), so I'm glad to see somebody is doing well in getting feedback.

best,

John

(edited 6-27 to reflect fact that the re-worked file noted above has now been removed from my site)

ES,

> so what are the pros then doing different? <

I don't know what a pro would do differently with your guitar track. But as I explained above, it's a common problem to have too much deep bass in a mix. And why is this a common problem? Because the low end in most home studios is way off because the rooms are small and don't have bass traps. heh

--Ethan

The thing is in this particular case, I purposely recorded in a larger space to avoid the phenomenon you mention above. I took the two tracks I recorded and panned one hard right and one hard left and mixed it down in CEP. Other than that there was no processing done that could impart any additional small room effects.

But you said in your first post that you don't perceive any low end problems in this particular clip anyay, so none of this would matter in the instance of this particular file. I tend also to think whatever I heard was not due to deep bass but maybe in the 200Hz-400Hz region if anything bass related(but it might not be bass related..). It seems that is where I have the greatest concentration and those frequencies are even more prevalent when he strikes the harmonics.

I do agree that a great majority of people do experience low bass problems as you describe that are due to mixing in small untreated rooms. My first mixes were all bass, and lots of songs I have downloaded from here contained excessive bass. The perplexing thing for me is that I could have a mix containing a lower level of bass more comparable to commercial stuff, yet mine still seems to distort a subwoofer system more than some of the stuff (with more bass)I downloaded from fellow gearslutz.

I guess more than the question about a pro I wanted to hear your answer to the second question, Ethan:
Also, I read some of your acoustic information in your acoustics forum last night, and one thing I noticed you mention that could be a problem in a larger room such as this is low frequency reverb....

The bottom line is, if I went and tracked some pop rock type songs in a nice auditorium, am I still going to run into acoustic problems? This is what I planned on trying to do for my next project. One thing that I realized is that the room will probably have a higher reverb time without an audience? Or am I really better off just treating a small room and tracking there?

Thanks, Ethan, I do appreciate it.

I am starting to get confused on what the question is...

for some reall basic into about microphone and signal check out
http://emusician.com/sod/guitar_recording_techniques/

for some easy reading about mixing and mastering and if you want to see what the "pro are saying about mixing and mastering..." here are two other links

http://emusician.com/mag/emusic_myst...ing/index.html
http://emusician.com/tutorials/emusi...ing/index.html


In the end, there is nothging wrong with your recording as you made it... it is accurate for how you set it up... not all speakers is capable or reproducing sounds the way they really are. If you want your mixes to play on most everything then you have to work at getting rid of frequencies that the cheaper stull can not make... or reduce the level of those low freq transients. Check out the Effects>Filters in Cool Edit Pro and look for fast fourier transform (FFT) this can help you attenuate some of the lower stuff without killing the rest of the mix.... of course some eq, ect ... will have to be used also to keep your sound the same.

What I don't know is are you more interested in microphone and setting up for ins (with allowing characteristic for room size and acoustical characteristic of a room,) or in frequency filtering, (which is more for final mastering).... They are two different questions and they work at cross purpose.

I don't know anymore to say ...

Good Luck

Zabour,

I tried to reply earlier before your last post, but I was having trouble with both my computers. I downloaded both yours and John's files. I haven't been able to burn them onto CD and compare on other systems yet. From what you said, and judging from the picture you posted, maybe I need to upgrade my CEP 2.0 to the version of Audition that you have. In the frequency analysis in my version of CEP, the curves for your version of the song and mine both look the same! I am posting a screen shot of it below. Can you tell me what exactly you did when you processed the file? Also, what does each color represent in the picture you posted? Like I said, I didn't get to listen very much before the computer started acting up, but the biggest thing I think I noticed was decreased room sound in your version.

In my picture, red is the Pink Floyd guitar, purple is yours and/or the original silent night mix I posted, and green is the original silent night file at the point where the first set of harmonics is played.
You will notice that the Floyd example contains the most low bass, and that the harmonics have large peaks at 220 and near to 400Hz.

I can't get work done, so I am going back to the beginning and try and make sure I am doing everything ok. Once I am convinced I am not introducing artifacts that doom a project at the tracking stage, then I will worry about mixing and then finally mastering. I am at the point right now where I am afraid to track for fear that I am always bound to get standing waves, or comb filtering or something that will ruin the work.

Thanks for your help.

the recording isn't so bad. no serious distortion. for some notes, the string hits the fret, that's causing a short minor rattle.

but it is true that without compression+limiting many smaller playback devices can't keep up with the natural dynamics. they could play pop music fairly loud, but have difficulties with solists. this is because the speaker needs a high sine-wave power.

the noise is quite bad. also the stereo balance and phase has some difficulty.
you had to go with these mics so it's what it is.

the transient designer is great but easy to overdo.

I took a different approach:
30Hz hipass, a bit EQ, denoiser, stereo symmetry adjuster, expander tuned to upper mid and HF range, final EQ, security limiter. but this took away the church a bit..

Ethan revealed part of my 'secret' in his first reply heh :

Ask and they will come.

I downloaded your version of my file as well as your song and I will post in your thread soon.

Yeah I got lots of replies and I really do appreciate them all, but not sure what we figured out other than I am possibly crazy and there is nothing wrong with my tracking and also that maybe my frequency analysis software isn't accurate and that I put too much faith in it which I always kind of thought a possibilty too....Thanks.

Maybe I made an error and sent the same file back.... I saved both as same name before so I am using different name (I messed up I'm so sorry to have caused any agrivation.)

usuall proceedure I use to remove subsonic rumble and VFL from a track is to split to mono and chain...

Then adjust phase (CEP and Audtion show cross talk as the wave from in between the two channels) a little is okay and tends to make the sound a little larger than it is but too much can null out some frequances.

Then a high pass filter usually remove or reduce anything under 24 Hz since the sound of the room was important I only reduced in you clip.

Next I use a FFT filter to mould sound for generic speakers, you can get a great deal of info on spkeakers and their profiles from searching on internet (nothing beat a good listen to on a pair of bad speakers tough). Basically, they do good from pop songs.... Good on mid-high A little worse on mid and mid-low, and Terrible on HF, LF and ULF. Min responce is usually between 24 Hz and 30 Hz. So a gentle curve up from 24 Hz (at -inf) to 36 (at -90) and then more aggresive curve to 50Hz (at -2 db) to 63Hz at (-0.1 db) you can leave the rest flat or adjust to suit your ear. If you clip has enough in the mid and mid high to keep the speaker cone vibrating quickly your bass will not come out on crappy speakers, which would reproduce it as distortion anyway so you will be safe then.


Here is the screen shot as I get in in Audition an a reworked file (I erase the one I had before so I start again - but I did not spend as much time.

This is at 1:00:121 which is peak harmonic in the end
The Red (1) : is the main one from you.
The Green (2) : is the one I edited

This is at 56:361 the quietest point in the mix, where you can see the different microphones and the chracteristics of the room a little.
The Blue (3) is from the main one from you
The wav form under it is from the filered one after the phase was taken away... look at the big difference in what was recorded!!! I will guess that the left mic was closer to the sound hole (but notice how the room harmonics are almost identical (this is the stuff under 100 Hz) and since Bass wave travel slower you can see them chasing each other on the graph as the notes in the room arrive to mic with dopler shift in frequance.)

As for the tracking (this is what I use for classical guitar it works okay but I am sure there is better ways.) try 3 to 4 mic a matched pair as x pattern (invese phased) 1/2 diameter of sound hole to each side, and one at pointed at the sweetspot of fingerboard. The player has to sit still for best result you can use one more HQ condenser mic on center and adjust the distance from the player to get more or less of the room. I usally use hardware bandpass and drop anything bellow 43 Hz which apromximative 1/2 of the lowest note a guitar can make (this subsonic rumble because the guitar body will not vibrate that slow and the probability that the wave form is to make out of the sound hole the sound hole is so so low as to negigiable.) If you want to have a DI from a ribon microphone for reference and mixing that is good too ...

Then you adjust the mix as you like and if you don't record any ULF you don't have to worry about them latter (ULF is more felt than heard anyway and when was the last time a classical guitar hit you like a drum... for me I don't recall )

For the Pink Floyd look at the frequencies bellow 100Hz especially between 70Hz and 20Hz this is what is most likley to cause distortion (although you always hear it with other notes the LFs to high ULFs are your big problems for reproducing on low cost speaker. Looking at that picture I might try some changes on my FFT pass to remove more of the frequances from 70Hz to 40Hz

I will make sure the correct file is attached this time....

In the end it is you who will have to decide what is acceptable and not because most people do not have the equipment to hear real sound (but it is always nice to try to get as much of it as possible before destroying most of it - that's the diff between tracking and mastering ... you spend all your time to get super real sound and the final masterer will spend all his time to chop off the parts that are too real for vulgar equipment amongst other things.

Eyes,

> you said in your first post that you don't perceive any low end problems in this particular clip anyay, so none of this would matter in the instance of this particular file. I tend also to think whatever I heard was not due to deep bass but maybe in the 200Hz-400Hz region <

Exactly. And I've been talking about at least three different things here. heh

1. The level and bass content in your recording is fine, so if it sounds wierd in your room it's probably your room.

2. Many people create mixes that distort small speakers at too low a level because their rooms are terrible and they can't hear how much deep bass they actually have. This is a mix issue, aside from the original recording.

3. Because of 2 above, inexperienced mix engineers often add too much deep bass when higher bass would be more appropriate. This too is a mix issue.

> if I went and tracked some pop rock type songs in a nice auditorium, am I still going to run into acoustic problems? <

It depends entirely on the auditorium, and also how far away you have the microphones. You can certainly get a good recording of acoustic instruments in a much smaller space. My live room (all in one room really) is 34 by 18 by 12 feet high at the peak, and I can record pretty much anything and not suffer from "small room" anomalies.

--Ethan

Zabour,
I tried downloading your files, but my computer keeps getting disconnected when I do. I also tried to download the file in one of the threads you started but I can't get it either. This happens with some files from this site. I don't know why. My computers are both junk.

Take your time... none of us is an a hurry and on something as complicated as this question we should all take our time.... There is a lot of science behind your question... I was looking for formulae and data on how cumulative accoustic shifts add and attenuate on different loud speakers, but I have not really had the time to do a good search... and most of the stuff I find concerns really high end loud speaker systems.....

If anyone has a lead or can point for me some referance material I would apreciate it very much.... This becoming a very interesting subject, and not one I had tought about a great deal. So, that I am thouroughly confused (and I am not sure if I always believed is accurate or bunk) I want to explore the physics and the math a little so that it will make sence again.

Good Night

I'm not sure what you mean here.

Can you explain? I know that phase is one thing that constantly has given me trouble and I do a lot of stereo micing. I know there is significant cancellation between the two channels in this song when going to mono, but I don't believe that it is possible not to have some when micing with a spaced pair. Phase is frequency dependent and if some frequencies are in phase others are out in a spaced pair setup, correct? The idea is just to listen and adjust to where the cancellation is minimal and in the least important frequencies or even frequencies you want to cancel, right? That's why I never understood people talking about aligning the tracks in a DAW by moving one by a few samples - the two will never truly be in phase.

You can check out a related thread where I miced a choir with a spaced pair out in front in that same church. It can be found here:
https://www.gearslutz.com/board/work-in-progress-advice-requested-show-and-tell-artist-showcase-mix-offs/35782-tracking-help-2-choir-church.html
I have no responses there right now. See if the phase relationship there sounds better or worse to you. See if you do notice room problems there since the mics were a little further from the source. Thanks.

But do you typically add artificial ambience to your recordings?

Thanks.

>Can you explain?
it is kind of hanging in my right ear. these stereo adjusters (balance, symmetry, etc..) can do a bit about it.
sample shifting is used a a correction for latency problems, and problems when you would want to correct the distance of a microphone from the source.
there is a free plugin package SS106.zip by "sweetboy", that has some effects for this. I found the stereostretcher very interesting.

Eye,

> do you typically add artificial ambience to your recordings? <

Sometimes, in the same situations anyone else would in any room. I mean, even the best studios have reverb units!

--Ethan