Need help: using a DAW EQ tool for balancing volume and analysis of EQ

    I could use some help... yes, that kind, but also this kind! (I don't want to keep bugging @schreckmusic in PMs... LOL!)


    Basically, I am frustrated with my attempts at being able to balance volume across rigs & across performances subjectively using just my ears, so I though I would try this approach since my manual comparison/tweaking wasn't working as I thought it should.


    1) Create a looper recording. (In the case, playing an E major, first position, strummed hard).
    2) Record multiple tracks into my DAW (Reaper) one at a time, each time changing the amp profile. I am going from my Monitor Out (since that is what drives my on stage sound support) into a Scarlett 2i2, into Reaper.
    3) Use a plug in tool (Voxengo Span) to look at each track, and (hopefully) analyze the peak volume and use that as a reference for adjusting the rig volumes to match.
    4) Secondary, but also hopeful that I could understand my EQ curves for future reference.


    So that is what I did. Below you can see 6 results including Peak dB, EQ curve at Peak and a secondary EQ that is supposed to represent the average of the entire track including decay. I used the stock Crunch profile and my current 'go to' rig based on M. Britt's AC30 65 profiles. (actual profiles not listed, I can provide that later if its at all useful.) The screen captures are just that - I took them at approximately the same time while the sample was playing but not exactly and they are not actually exported from the tool.


    Again, the goal is to volume balance all rigs. You can see that to my surprise, the Crunch rig comes out with the lowest volume at -18.5 dB, and my not so accurate attempts at balancing the others. I realize that I may want to increase the volume slightly with the gain at some point for sitting in the mix, but for now I am attempting to come up with a 'reasonably fool-proof method' & repeatable process for volume balancing. Rigs go from clean (AC30 1) to what is for me, 'high gain' (AC30 5).


    So, with that said, a couple of questions:


    1) Is this a reasonably sound approach? Use a consistent sample, compare to the stock rig, tweak the volume on the target rigs until I can get them to peak at the same dB (-18 dB here). I suppose I could also mic my Tech 21 Power Engine if that would be better than going directly into the DAW.
    2) What else can I learn from the EQ? Anything stand out to you when you look at it?


    Thanks in advance!!!



    Crunch - reference
    -18.5 dB Peak Volume
    [Blocked Image: http://i1368.photobucket.com/albums/ag198/flyingheelhook/Crunch%20for%20reference_zpsxodtk7xr.jpg]


    AC 30 1
    -12.5 dB Peak Volume
    [Blocked Image: http://i1368.photobucket.com/albums/ag198/flyingheelhook/AC30%201_zps13veitkg.jpg]


    AC 30 2
    -13 dB Peak Volume
    [Blocked Image: http://i1368.photobucket.com/albums/ag198/flyingheelhook/AC30%202_zpsd3haxgmv.jpg]


    AC 30 3
    -16.5 dB Peak Volume
    [Blocked Image: http://i1368.photobucket.com/albums/ag198/flyingheelhook/AC30%203_zpsebn0pov4.jpg]


    AC 30 4
    -15 dB Peak Volume
    [Blocked Image: http://i1368.photobucket.com/albums/ag198/flyingheelhook/AC30%204_zpsyevhpp68.jpg]


    AC 30 5
    -15.5 dB Peak Volume
    [Blocked Image: http://i1368.photobucket.com/albums/ag198/flyingheelhook/AC30%205_zpsudzyogz9.jpg]

    I do exactly the same thing for my sounds. Especially when it comes to deciding on my EQ settings. Has always worked well for me and I have played many venues of various sizes, with the best and worst of sound engineers at the helm. All of them, without exception, when asked about the guitar sound and rig balance said they had no problems whatsoever. This was the case for both miced cabs and running direct from the KPA.


    Great approach and I will continue to use it.

    “When the power of love overcomes the love of power the world will know peace.”

    Like 1
    Quote from SanctumSolaris

    I do exactly the same thing for my sounds. Especially when it comes to deciding on my EQ settings. Has always worked well for me and I have played many venues of various sizes, with the best and worst of sound engineers at the helm. All of them, without exception, when asked about the guitar sound and rig balance said they had no problems whatsoever. This was the case for both miced cabs and running direct from the KPA.


    Great approach and I will continue to use it.

    Do you slightly increase the volume with the increase in gain or keep all rigs basically the same? Do you shape your EQ curve differently than mine?

    I set my main rhythm preset to just as it hits yellow on the meter in the DAW. Then my lead only sounds boosted because I invert the EQ from a mid scoop to a mid boost, (ala Randy Rhoads). It usually does boost the lead into the yellow. My clean sound is about 4db lower than the rhythm preset. I only use the clean on two songs, but we like that when the change occurs there is a dynamic drop. Below is a list of the two EQ settings I use. Toying around with using the Studio EQ preset for rhythm. It works for that purpose, but I will need to adjust it for the lead sound. Also, my EQ for rhythm and clean are the same.


    RHYTHM and CLEAN



    Studio Equalizer

    Low Shelf High Shelf Low Mid High Mid
    0.0 at 120Hz 0.0 at 12004.8Hz -3.1 at 397Hz -3.1 at 800.7Hz
    Low Cut High Cut Q-Factor at 5.000 Q-Factor at 5.000
    52.0Hz 12004.8Hz Volume 0.00
    Mix 100% Ducking 0.00


    LEAD



    Studio Equalizer

    Low Shelf High Shelf Low Mid High Mid
    -0.3 at 150Hz -0.3 at 12004.8Hz 2.1 at 900.4Hz 2.5 at 3001.2Hz
    Low Cut High Cut Q-Factor at 0.670 Q-Factor at 0.670
    80Hz 16009.8Hz Volume 0.00
    Mix 100% Ducking 0.00


    STUDIO EQ RHYTHM PRESET


    Studio Equalizer

    Low Shelf High Shelf Low Mid High Mid
    1.3 at 200.8Hz 2.7 at 10006.9Hz 1.4 at 501.9Hz 4.0 at 3012.2Hz
    Low Cut High Cut Q-Factor at 0.707 Q-Factor at 0.707
    20.6Hz 33488.1Hz Volume 0.00
    Mix 100% Ducking 0.00

    “When the power of love overcomes the love of power the world will know peace.”

    Like 1

    For your EQ comparisons, I'd suggest running a white noise loop instead of some guitar chords. This will give you the EQ picture for all frequencies, not just the notes in the E-chord and their harmonics. You can download a free white noise loop, put it on a track and loop it. Reamp the noise into the Kemper as if it were a DI guitar track and monitor them in SPAN.


    For level matching, it's a little more complicated due to Fletcher/Munson. Looking at the main meters won't necessarily give the apparent loudness because some sounds will have more energy in frequencies that the ear is less sensitive to, such as bass heavy sounds. It might be better to limit your level matching to 200-1000Hz where the ear is more sensitive but where some fundamentals still exist. You can do this in SPAN by either looking at the spectrum, or putting a HP and LP on the noise loop and using the main meters. I'm sure there is some AES paper somewhere that covers matching perceived levels, but this would be my guess without doing the research.


    And finally, it might be easier and faster to set your levels in real time. Run your noise loop, reamped into the Kemper, and put SPAN on the return input with real time input monitoring. Then run your reference sound that you want to match levels to and note the levels. Then change sounds on the Kemper and set your patch level to match the reference and save it. This way you can scroll through your sounds without recording anything.

    I hate emojis, but I hate being misunderstood more. :)

    Like 2

    Also, you might want to band pass the white noise to guitar fundamental frequencies regardless, 82 to 1200Hz so your harmonic distortion isn't over-represented in the signal returning from the Kemper. For example if unfiltered, 2000K would be represented by both the harmonic distortion from normal guitar frequencies AND from fundamentals in the noise that aren't significantly present in a dry guitar signal.

    I hate emojis, but I hate being misunderstood more. :)

    Like 1
    Quote from flyingheelhook

    So, with that said, a couple of questions:


    1) Is this a reasonably sound approach? Use a consistent sample, compare to the stock rig, tweak the volume on the target rigs until I can get them to peak at the same dB (-18 dB here). I suppose I could also mic my Tech 21 Power Engine if that would be better than going directly into the DAW.

    I'd say that if perceptual volume is more-important to you, and let's face it, levels are always set when recording so IMHO aren't worth bothering about in that regard, I'd not use a conventional metering paradigm.


    Enter the ITU-R BS.1770 standard. To quote SOS:
    This is a technical recommendation which defines a 'loudness metering' algorithm (see side boxes) which provides a standardised, objective and reliable means of measuring, comparing and adjusting programme loudness.

    Note that this is about loudness as perceived by human beings, and not simply 'level'. In other words, the ITU-R 'Loudness Meter' algorithm is essentially an 'electronic ear' that perceives audio loudness in much the same way as the average human listener, and it has proved to be very accurate and very reliable.


    The excellent SOS article by Hugh Robjohns:


    The End Of The Loudness War?


    Interestingly, the iTunes' Sound Check levelling algorithm attempts to achieve the same thing, but it's a proprietary algorithm developed by Apple in the early noughties.


    Unfortunately Hugh doesn't, AFAICT, specify any plugins by name, but I'm sure there'd be many out there by now that offer this as a readout option along with the usual suspects such as peak, RMS, the various K-System flavours and so on. One of these plugins is what you want, 'Hookster, IMHO.


    EDIT:
    Here's an excellent TC Electronics article on the subject of loudness. Be sure to check out the paragraphs under the heading, "LKFS, LUFS & LU":


    Loudness Explained - Loudness | TC Electronic

    Doesn't pro tools have a plug in to do this automatically? Sorry of its not helping
    thats the only thing i can thing of. Is the prob just REAPER? hmmm...



    Ash

    Have a beer and don't sneer. -CJ. Two non powered Kempers -Two mission stereo FRFR Cabs - Ditto X4 -TC electronic Mimiq.

    Quote from Michael_dk

    I'll also add that it would be more beneficial to look at the RMS level rather than the peak level for your purpose.

    This. But if you are also interested in peak levels, the auto gain plugin by hornet plugins is useful for what you are after. It will auto level any input signal typically to -18dfs, but that reference level can be adjusted. You could use this for a more accurate way of setting the same (peak) levels across profiles.

    I think perceptual levels are his primary concern here. I could be wrong of course.


    IMHO, the volume balancing, when carried out using perceptual measurements (LKFS / LUFS / LU units), won't incur huge variations in peak levels, at least, not severe enough to bother FOH engineers.


    As long as he uses his loudest-peak Rig (label one accordingly, 'Hookster, even if it's one you don't use for any other purpose; it only needs to be a tad hotter than the rest of your Rigs) as a reference for setting the desk's input trim (at soundcheck - "Here's the most juice I'll be sending ya, Mr. FOH engineer"), all should be fine, 'cause perceptually they're all gonna fall into the same ballpark, which in turn will be well-under potentially-clip-inducing levels.

    Just curious Schreck, why would you go for pink noise? Is that to avoid having to put a band-pass filter on the white noise? The pink might be better for judging loudness by ear by favoring the lower frequencies, but for assessing the EQ response by eye, I think white noise might be better, no?

    I hate emojis, but I hate being misunderstood more. :)

    Quote from Klappy

    Just curious Schreck, why would you go for pink noise? Is that to avoid having to put a band-pass filter on the white noise? The pink might be better for judging loudness by ear by favoring the lower frequencies, but for assessing the EQ response by eye, I think white noise might be better, no?

    I do not have a scientific answer as to why I use pink noise as apposed to white. In doing different tests using both white or pink, I've always found that pink gave me better results as to what the frequency response was.


    The plugin I use is http://www.waves.com/plugins/emo-generator Granted I've only been using it for about a week or so.

    Yeah, it's definitely-technical, AJ. For most, levelling by ear will be enough.


    Some, of course, will seek to further reduce the margins between the loudness of their Rigs. Using perceived-loudness metering could conceivably speed up the process of levelling large numbers of Rigs 'though.

    Hmm, intereting read, but I have some thoughts:
    * you are looking at the peak level reading, but its my understanding the RMS value is more important when it comes to loudness
    * surely, playing a big open E major chord isnt the best way to compare a clean and a heavy distorted signal in terms of loudness. Be aware that a distorted amp acts as a compressor, and the more distorted you play the more you lean toward monophonic playing (=less input volume). So not sure what you would be achieving in a real world scenario, when you would for example play a solo on a clean patch...
    * Also, sending the soundtech the loudest patch you have for setting trims has a pitfall: If the sound engineer tends to trim that patch to "0dB", he will find himself pushing your average "playing rythm in the carpet" patch to +6dB or more, finding himself to be unable to lift your volume when you would solo using that same rythm patch (been there, done that). Better to set 0dB on an average patch, and then double check you digital console's input isn't clipping at +18dB on that very loud patch.
    * Also, sometimes you have lead, sometimes you have not. Offcourse you can control this volume difference with a pure booster or something else
    * you're talking abot EQ as if it were part of your tone, and it sure is, but it's also a way to balance the mix: you would cut the guitar in de 1-2k region to make place for the vocals, and so on. So your ideal tone as a solo instrument isn't necesarily your ideal tone in a band.


    In the end there are too many variables. What counts is that as a band, you are able to play in balance. The sound engineer should, in theory, be able to go home after soundcheck. So, set your patches that during rehearsal everything sounds as it should on the master. Be not afraid to judge the volume of a different bandmember. Ask someone else to listen in. Record the whole thing and listen to it without changing the mixdown volumes: where do you need more volume, where do you need less? My opinion is that thats the only real foolproof method to take into account all of the above?

    OMG. You guys are effing awesome and also mostly talking waaaaaayyy over my head!!! LOL! 8|


    But, I think you are definitely pushing me in the right direction and I probably gave too much information in my original post.


    Here is what I am after:


    PRIMARY: A *simple*, repeatable process for volume balancing rigs. Monkey Man is spot on - MOSTLY I am aiming for perceptual volume balancing as 90% of what I do is live work, and the little I do for recording is mainly just to lay down rough draft ideas. I have a ton of profiles I want to set up in performances and every time I sit down to do it I get frustrated by both trying to manually go back and forth to balance the volumes across performances and then finding that there is too much variation when I try to use them live.


    SECONDARY: A better understanding of how to understand what the EQ says so I can use it to my advantage. Lets forget about that one for now as I think I need a lot of reading to fill in the gaps in my understanding just so I can understand what you guys are saying. :/


    So, this, barring the white vs. pink discussion, seems like the simplest way to get me closer:


    Quote from Klappy

    For your EQ comparisons, I'd suggest running a white noise loop instead of some guitar chords. This will give you the EQ picture for all frequencies, not just the notes in the E-chord and their harmonics. You can download a free white noise loop, put it on a track and loop it. Reamp the noise into the Kemper as if it were a DI guitar track and monitor them in SPAN.


    For level matching, it's a little more complicated due to Fletcher/Munson. Looking at the main meters won't necessarily give the apparent loudness because some sounds will have more energy in frequencies that the ear is less sensitive to, such as bass heavy sounds. It might be better to limit your level matching to 200-1000Hz where the ear is more sensitive but where some fundamentals still exist. You can do this in SPAN by either looking at the spectrum, or putting a HP and LP on the noise loop and using the main meters. I'm sure there is some AES paper somewhere that covers matching perceived levels, but this would be my guess without doing the research.


    And finally, it might be easier and faster to set your levels in real time. Run your noise loop, reamped into the Kemper, and put SPAN on the return input with real time input monitoring. Then run your reference sound that you want to match levels to and note the levels. Then change sounds on the Kemper and set your patch level to match the reference and save it. This way you can scroll through your sounds without recording anything.


    It definitely makes more sense to use the white/pink noise instead of just a cord. My only concern is how to accomplish it as my Scarlet 2i2 doesn't have SPDF - but I am going to assume for my purposes that is ok. I hope to be able to test this tonight or tomorrow at least dry run.


    Quote from Michael_dk

    I'll also add that it would be more beneficial to look at the RMS level rather than the peak level for your purpose.


    I googled RMS and that just went way over my head. BUT, the fact that my plug in does measure it and give me a number does at least give me a metric that I can set a standard (or even multiple standards like Clean, Grit, Gain, Lead, etc. once I have a standard for them) to match using that tool. I DO understand that different tones and different uses (i.e. Lead, Rhythm) require different volumes, but I want to set one standard for now - once I understand and can repeat that standard easily I can look at the others.




    Quote from Monkey_Man

    Yeah, it's definitely-technical, AJ. For most, levelling by ear will be enough.


    Some, of course, will seek to further reduce the margins between the loudness of their Rigs. Using perceived-loudness metering could conceivably speed up the process of levelling large numbers of Rigs 'though.


    I am definitely going to look into this as that would seem to be the most suitable solution if its works and I can set it up properly.

    Glad to be of service, 'Hookster!


    Yeah mate, if you can find a meter that offers a perceived-volume option, it'd be a simple matter of settling on a level you like, running a DI loop through the Kemper and knocking yourself out.


    There'd be no need to bother with fancy, schmancy filtering, white or pink noise or anything else.

    Quote from flyingheelhook

    It definitely makes more sense to use the white/pink noise instead of just a cord. My only concern is how to accomplish it as my Scarlet 2i2 doesn't have SPDF - but I am going to assume for my purposes that is ok. I hope to be able to test this tonight or tomorrow at least dry run.

    Yes, Analog reamping will be fine.

    I hate emojis, but I hate being misunderstood more. :)