Home |
Search |
Today's Posts |
#1
|
|||
|
|||
Recording Question(s) from a Beginner
hello, I'm new to recording, I have a Boss GT6 that I run into my
audiophile 2496 soundcard, been messing with it for a month now and cool edit pro, I manage to get some pretty clear recordings out of it most of the time.. i have a lot of trouble with guitar distortion, often they sound muddy and distant.. my questions are as follows: 1- what are the best drum samples available for purchase? grouped as full kits, wavs. i've heard battery mentionned a lot? 2- i read an article by billy howerdel of a perfect circle where he mentionned that all the effects on their latest album were done using protools software, i find that remarkable because the album sounds like a million bucks. can someone give me a brief explanation of how this works? where can one find professional software guitar distortion/overdrive effects? is there a site that has samples? 3- if i record a guitar track in cool edit, i usually keep the levels so that nothing ever goes past -9db, 0db being the point where there would be clipping or crackling.. now if i have my one track thats at -9db, i can hard limit it and it sounds like this: http://myshoeshurt0.tripod.com/1.mp3 BUT, if i record another track, or a couple other tracks, all at -9db or lower, and mix them together, the mixdown track is now nearly maxed out at -1db.. so when i hard limit it, its not as loud and present as the original single track: http://myshoeshurt0.tripod.com/2.mp3 is this normal? because the more guitar tracks i record, the lower the volume when i mix them together, its like they get stacked on top of each other.. what i don't get is how do people record 20-30 guitar overdubs without the final product sounding like its playing out of a $10 stereo? it just leads me to believe i'm doing something wrong because every track i add lowers the volume of the final mix otherwise the sound gets distorted.. if i put 10 tracks in i have to lower the dbs on each track so low to avoid clipping and crackling that the final product is -noticably- less audible and -noticably- more distant.. any tips would be greatly appreciated. thanks |
#2
|
|||
|
|||
Recording Question(s) from a Beginner
In article , Vine wrote:
1- what are the best drum samples available for purchase? Best ones I've ever heard are in my Roland XV. But this might be related to the fact that when I first heard them, they were being played by an awesome drummer playing on his awesome midi drum kit. 2- i read an article by billy howerdel of a perfect circle where he mentionned that all the effects on their latest album were done using protools software, i find that remarkable because the album sounds like a million bucks. Lots of people use protools, even in studios with a million bucks invested. Did Billy claim that the album was done *just* using protools, or using protools and a studio's sweeeeet (5 e's) collection of outboard fx, $20,000 mikes, and some expensive engineering talent? 3- if i record a guitar track in cool edit, i usually keep the levels so that nothing ever goes past -9db, 0db being the point where there would be clipping or crackling.. Why? You have a 24 bit sound card. That's 48dB of headroom over the best 16 bit ADC. You don't have a -120dB noise floor. Why risk clip? You can have way low record levels and not really lose snr or dynamic range. Don't be tempted to treat digital peak meters like VU meters. Keep your levels low, and think in terms of average level. There is a hell of a lot of room in 24 bit. You don't really need to be anywhere near the top of the meter. just leads me to believe i'm doing something wrong You're trying to get more "loudness" into the first stage of your recording, and that's raising your average signal. You're not normalizing tracks are you? Or using a compressor? |
#3
|
|||
|
|||
Recording Question(s) from a Beginner
Vine wrote:
3- if i record a guitar track in cool edit, i usually keep the levels so that nothing ever goes past -9db, 0db being the point where there would be clipping or crackling.. now if i have my one track thats at -9db, i can hard limit it and it sounds like this: http://myshoeshurt0.tripod.com/1.mp3 BUT, if i record another track, or a couple other tracks, all at -9db or lower, and mix them together, the mixdown track is now nearly maxed out at -1db.. so when i hard limit it, its not as loud and present as the original single track: http://myshoeshurt0.tripod.com/2.mp3 The total is equal to the sum of its parts. If you mix two tracks each maxing out at -9 dB, you will at least occasionally get peaks of -3. That's because two equal strength signals add to a signal that peaks 6 dB higher than either of them. As they say, do the math. Add a third or fourth track and you might go over the top. is this normal? The lazy solution to "maxing out" mixdowns in Audition/Cool Edit is to mix down in 32 bits mode. In multitrack settings, just turn on 32 bits for all three items where it relates. However, depending on which version of Cool Edit you have, these options may not be available. Check it out. The more-work solution to "maxing out" mixdowns in Audition/Cool Edit is to set lower operating levels for each track. There seems to be phobia to running tracks peaking at -20 dB and below, probably because at this point the waves become so small that you can't see them very well in multitrack view. I don't know what the solution to this situation is. Maybe some we'll have the option to start getting used to viewing waves in dBs. Maybe we'll just have bigger displays. In 32 bit mode, CoolEdit is almost impossible to overload. The dynamic range is literally 100's of dBs. Once your 32 bit mixdown is done you'll have to attenuate or normalize before you covert the mixdown to 16 bits, to avoid clipping. However, you'll know what the true maximum levels are, so you can normalize or attenuate accordingly. |
#4
|
|||
|
|||
Recording Question(s) from a Beginner
Best ones I've ever heard are in my Roland XV. But this might be
related to the fact that when I first heard them, they were being played by an awesome drummer playing on his awesome midi drum kit. that doesn't help Lots of people use protools, even in studios with a million bucks invested. Did Billy claim that the album was done *just* using protools, or using protools and a studio's sweeeeet (5 e's) collection of outboard fx, $20,000 mikes, and some expensive engineering talent? said all the guitar effects were done with protools software and then he had a hell of time recreating them with hardware for the tour. so how i find professional guitar distortion software effects? Why? You have a 24 bit sound card. That's 48dB of headroom over the best 16 bit ADC. You don't have a -120dB noise floor. Why risk clip? You can have way low record levels and not really lose snr or dynamic range. Don't be tempted to treat digital peak meters like VU meters. Keep your levels low, and think in terms of average level. There is a hell of a lot of room in 24 bit. You don't really need to be anywhere near the top of the meter. i didn't understand any of this.. can you dumb it down for me? You're trying to get more "loudness" into the first stage of your recording, and that's raising your average signal. You're not normalizing tracks are you? Or using a compressor? nope, like i said beginner what should i be doing? |
#5
|
|||
|
|||
Recording Question(s) from a Beginner
The total is equal to the sum of its parts. If you mix two tracks each
maxing out at -9 dB, you will at least occasionally get peaks of -3. That's because two equal strength signals add to a signal that peaks 6 dB higher than either of them. As they say, do the math. Add a third or fourth track and you might go over the top. so how do you record 30 guitar overdubs and manage to have the final product sound normal (volume wise)? The lazy solution to "maxing out" mixdowns in Audition/Cool Edit is to mix down in 32 bits mode. In multitrack settings, just turn on 32 bits for all three items where it relates. However, depending on which version of Cool Edit you have, these options may not be available. Check it out. ok just did that, i rerecorded the whole thing, mixed it down, everything is the same. it changed nothing The more-work solution to "maxing out" mixdowns in Audition/Cool Edit is to set lower operating levels for each track. There seems to be phobia to running tracks peaking at -20 dB and below, probably because at this point the waves become so small that you can't see them very well in multitrack view. I don't know what the solution to this situation is. Maybe some we'll have the option to start getting used to viewing waves in dBs. Maybe we'll just have bigger displays. but i i've done this, i've recorded all my tracks at low volume so that i never risk clipping and have to lower volumes in the multitrack, but it doesn't make a difference because when i mixdown i can still only get the track to sound clear up to a reduced volume level inversely proprotional to the number of tracks recorded. so if i recorded 30 tracks, the mixdown file would probably be three times less loud than i could make a single guitar track.. and that would be too low. whether i record a track low at -18 db and hard limit it 18db or record a track at -9db and hard limit it 9db, the end result is the same. the level i record it at changes nothing, i could record one track at -1 db, and then if i add another track i'll just have to lower it by 3db in the multitrack, add another track, lower it more, etc.. even you end up lowering beyond a reasonable point. In 32 bit mode, CoolEdit is almost impossible to overload. The dynamic range is literally 100's of dBs. Once your 32 bit mixdown is done you'll have to attenuate or normalize before you covert the mixdown to 16 bits, to avoid clipping. However, you'll know what the true maximum levels are, so you can normalize or attenuate accordingly. it changed nothing for me, it sounded exactly the same as the 16bit, it looked the same (visually) as the 16 bit, the markers are the same, the maxes are the same, i could increase the volume no more than i could the 16 bit, and when i normalized to 100% it cut the volume in half. half of what you (may) have heard in 2.mp3.. |
#6
|
|||
|
|||
Recording Question(s) from a Beginner
Vine wrote:
The total is equal to the sum of its parts. If you mix two tracks each maxing out at -9 dB, you will at least occasionally get peaks of -3. That's because two equal strength signals add to a signal that peaks 6 dB higher than either of them. As they say, do the math. Add a third or fourth track and you might go over the top. so how do you record 30 guitar overdubs and manage to have the final product sound normal (volume wise)? Very carefully... The lazy solution to "maxing out" mixdowns in Audition/Cool Edit is to mix down in 32 bits mode. In multitrack settings, just turn on 32 bits for all three items where it relates. However, depending on which version of Cool Edit you have, these options may not be available. Check it out. ok just did that, i rerecorded the whole thing, mixed it down, everything is the same. it changed nothing I'm betting that you based your judgment on either monitoring of the 32 bit file via your audio interface, or via a 16 bit file that wasn't first normalized in the 32 bit domain. Audio interfaces as a rule don't handle 32 bit signals that are greater than full scale (i.e. 1.00) very well. Here's the exact order of operations: (1) Mix down to a 32 bit (floating point) mixdown file. (2) Normalize or other wise attenuate the mixdown file so that it doesn't exceed FS. I typically normalize to 1 dB below FS. (3) Playback via audio interface or convert 32 bit normalized mixdown file to 15 bits. The more-work solution to "maxing out" mixdowns in Audition/Cool Edit is to set lower operating levels for each track. There seems to be phobia to running tracks peaking at -20 dB and below, probably because at this point the waves become so small that you can't see them very well in multitrack view. I don't know what the solution to this situation is. Maybe some we'll have the option to start getting used to viewing waves in dBs. Maybe we'll just have bigger displays. but i i've done this, i've recorded all my tracks at low volume so that i never risk clipping and have to lower volumes in the multitrack, but it doesn't make a difference because when i mixdown i can still only get the track to sound clear up to a reduced volume level inversely proprotional to the number of tracks recorded. so if i recorded 30 tracks, the mixdown file would probably be three times less loud than i could make a single guitar track.. and that would be too low. So then bring it up to say, 1 dB below full scale. whether i record a track low at -18 db and hard limit it 18db or record a track at -9db and hard limit it 9db, the end result is the same. the level i record it at changes nothing, i could record one track at -1 db, and then if i add another track i'll just have to lower it by 3db in the multitrack, add another track, lower it more, etc.. even you end up lowering beyond a reasonable point. I think that's inherent in the process. The more voices you add, the smaller the contribution of each voice, given that overall loudness stays the same. In 32 bit mode, CoolEdit is almost impossible to overload. The dynamic range is literally 100's of dBs. Once your 32 bit mixdown is done you'll have to attenuate or normalize before you covert the mixdown to 16 bits, to avoid clipping. However, you'll know what the true maximum levels are, so you can normalize or attenuate accordingly. it changed nothing for me, it sounded exactly the same as the 16bit, it looked the same (visually) as the 16 bit, the markers are the same, the maxes are the same, i could increase the volume no more than i could the 16 bit, and when i normalized to 100% it cut the volume in half. half of what you (may) have heard in 2.mp3.. That's inherent in the process. In the end, everything can sum to no more than 1.0. With floating point it is acceptable to have intermediate results that are 1.0, but before you burn the CD or route it out your audio interface, nothing can go 1.0 without the certainty of clipping. |
#7
|
|||
|
|||
Recording Question(s) from a Beginner
In article ,
Vine wrote: i didn't understand any of this.. can you dumb it down for me? Yeah. Your input levels are too high. There's no reason you need them at the top of the scale. Work at -16dB or so. -48dB on a 24 bit recorder is equivalent to 0 on a 16 bit recorder. Do you, by any chance, have an analog tape recorder with an actual VU meter? If you were to put that in your signal path, it would show you the difference between analog VU, which is a measure of average loudness, and digital peak, which is a whole nother animal. Work for a while with way lower levels, and you'll get a good idea of what's going on. Even if you're recording heavy metal guitar tracks or a punk rock band, it's a mistake to record near the top of the 24 bit scale. If you really want to end up with one of those awful finished "products" with all the dynamic range in a single bit, you can f*k that up at the last step. nope, like i said beginner Unlearn anything you know about using a tape recorder. |
#8
|
|||
|
|||
Recording Question(s) from a Beginner
nevermind, you guys don't get it, what difference does it make what level
i record at? in the sense that how does that change my potential final mix maximum volume? if i record a track at -9db, i can hard limit it to its max volume +9db, if i record the same thing at -18db and hard limit it to its max +18db, it'll sound -exactly- the same.. i did it, sounds the same, i made a new file in cool edit, one 16 bit and one 32 bit, recorded the same thing, made each one as loud as possible without dsitorting the sound, verdict: exactly the same. i normalized to 100% and all that did was cut the volume in half, the exact opposite of what i'm trying to accomplish now if i record 30 guitar tracks and try to mix them together, it'll sound like crap, all i want to know is how do bands record 30 layers of rhythm guitar and have it sound like a huge LOUD wall of sound five inches from your face instead of it sounding distant and barely audible? the more tracks i record, the lower the volume of each individual track has to be to avoid crackling and distortion.. at ten tracks it becomes futile. also, do any of you know where to find software guitar distortion/overdrive tones, ie protools, and where one could hear samples, etc.? |
#9
|
|||
|
|||
Recording Question(s) from a Beginner
nevermind, you guys don't get it, what difference does it make what level
i record at? in the sense that how does that change my potential final mix maximum volume? if i record a track at -9db, i can hard limit it to its max volume +9db, if i record the same thing at -18db and hard limit it to its max +18db, it'll sound -exactly- the same.. i did it, sounds the same, i made a new file in cool edit, one 16 bit and one 32 bit, recorded the same thing, made each one as loud as possible without dsitorting the sound, verdict: exactly the same. i normalized to 100% and all that did was cut the volume in half, the exact opposite of what i'm trying to accomplish now if i record 30 guitar tracks and try to mix them together, it'll sound like crap, all i want to know is how do bands record 30 layers of rhythm guitar and have it sound like a huge LOUD wall of sound five inches from your face instead of it sounding distant and barely audible? the more tracks i record, the lower the volume of each individual track has to be to avoid crackling and distortion.. at ten tracks it becomes futile. also, do any of you know where to find software guitar distortion/overdrive tones, ie protools, and where one could hear samples, etc.? |
#11
|
|||
|
|||
Recording Question(s) from a Beginner
First of all I didn't download your mp3 as it would take about 10
minutes. It's helpful to have smaller samples. I only need to listen for about 20 seconds to know what is going on. they are small, they're like 10 seconds and 500k each I know your GT6 unit very well and the Satch setting sounds HUGE in STERIO. whats the satch setting? satriani? i don't see it You mentioned crackles. This almost maks me think of completely off the wall issues such as driver problems, card incompatability, slow processor speed and who the hell knows what else. nah, it happens when i make my tracks louder than they should be i guess.. Ten tracks should be easy to do as long as your instruments have thier own place in the sonic mix. Each instrument occupies a differant frequency in the mix. but when you do overdubs of the same instrument you now have two of the same trying to squeeze into the same piece of the pie, is there a trick to getting them to fit together without affecting the volume of the overall mix? some eqing trick or something.. Are you using the headphone setting on your GT6? i do even though they all sound the same, i use the digital out.. Dont forget to try recording things in sterio. It'll sound bigger. The grunge distortion is quite a good in your face setting patch. where is the grunge distortion, i read the list i can't find it? i have such trouble getting a big, deep, thick, in your face distortion that doesn't sound muddy or distant.. its either distant and muddy or bright and fuzzy.. any ideas? preamp/overdrive combos, settings, etc.. ? i wish there was a site online where people put up samples of their favorite tones and the settings to make them.. |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
recording gear questions | Pro Audio | |||
Why all the bad recordings | High End Audio | |||
Recording Live Music: Questions Of Price, Quality & Portabiltiy | Pro Audio | |||
Questions about computer based recording | Pro Audio | |||
a couple of questions about recording on PC | Pro Audio |