HarBal. Useful? Alternatives?

Hi there!

I'm mastering a compilation cd soon, containing songs from local bands. I recorded some of the songs but most of it is recorded in different studio's, so there will be a lot of different sounding recordings on it, though it's all in the same genre (metal). Luckily most songs the bands send me are unmastered (not already squished) mixes.

Now I'm not a specialized mastering studio but still they asked me to do it. The reason being that I'm a lot cheaper than a professional studio... It will be a good opportunity to get my mastering skills to a higher level (I hope).

I did master a lot of my own mixes with Waves Q10, C4, C1 and L2 in the past and got pretty good results but I feel I could be much better so I started to do some more research. I then stumbled upon HarBal!

It seems like the perfect tool to get the different productions on the compilation sound more alike quick and easy.

I'm just wondering what you guys think of HarBal, and if you maybe know better alternatives.

Please allow me to note two things:

1.Audio mastering needs your ears, not your eyes.

2.There is no need to make the different productions on this compilation sound more alike. Hey, they are different for a reason. You don't want them all to sound "the same". This would not be a compilation of different bands, but an album of one artist.

3.If you want to try something different than Waves, may I suggest Voxengo?

Originally posted by lowbasslowbass Please allow me to note two things: 1.Audio mastering needs your ears, not your eyes. 2.There is no need to make the different productions on this compilation sound more alike. Hey, they are different for a reason. You don't want them all to sound "the same". This would not be a compilation of different bands, but an album of one artist. 3.If you want to try something different than Waves, may I suggest Voxengo ?

Yes, I know I should use my ears. I was just wondering if Har-Bal makes it easier, especially for someone who's main specialty isn't mastering (like me!).

Since I posted this I read a lot of reviews and they're all very positive about Har-Bal.

In fact: I just played around with Har-Bal on a mix which I thought was pretty good and was able to make some really nice improvements! Too bad the demo version is limited to 8 bit...

I guess making the different songs on the compilation sound exactely alike will be impossible anyways...

I'll check out the Voxengo stuff! Thanx!

Voxengo Soniformer looks kinda like an alternative to Har Bal. Just an observation.

I use Har-bal as a spectrum analyser. I don't use it as an eq - although it's a very good eq. I prefer to use the information I see to help me go back and mix better.

Products that simply analyse the spectral curve from one mix, and then create an eq to ram another mix into the same shape, are a recipe for disaster.

Mixes should be unique. Only your ears can tell if a particular frequency band sounds good or nasty. No point automatically boosting a particular band willy nilly just to match another curve. Who knows what you might be boosting. Also, no use automatically cutting out stuff - it might be the good stuff.

Spectrum analysis is a useful guide - I consider it a second opinion.

Leave a mastering engineer to put your whole album together - it's money well spent.

I use HarBal a lot for mastering and I like it. I also use it as a spectrum analyser. I also use it on individual tracks at times!

Currently, HarBal does not automatically create an eq curve and "ram" it into your mix. You can use the curve from any mix and then you have to adjust your mix manually.

And yes, you have to use your ears! It takes quite a few tries to create a good curve for a particular tune and then listen to it on different systems.

60Hurts! I agree. The people that diss HarBal are the people that haven't used HarBal, relying on wisdom and soundbytes from forums, etc. In other words, second hand info.

HarBal is a great tool. It requires your good taste to use it well. It is also, as Kiwi pointed out, a great eq.

I would much rather send my stuff to a good ME than do it myself but... if I've got to do it myself, you bet I'm using HarBal to help me along, regardless of any "use you ears, not your eyes", second hand logic.

I use the Waves Phase Linear Multi Band Compressor, process.

Then HarBal, process.

Then L2. Process down to 16.

Regarding making different artists fit together well for a compilation... that's a matter of taste and certainly not a rule. If something is sticking out, definately fix things to be more in line. I wouldn't go overboard in this direction, but I would be mastering each track with my ears on the matter.

I'm getting the impression a lot of people are dissin' HarBal because they think it's a n00b tool but the more I'm using the demo and reading about it the more I realize it's a serious and very useful tool.

In fact, I'm ordering it tomorrow! I'm a believer ! .

Lee, the tutorial on HarBal's site puts the linMB after harbalizing the waves. I'm wondering why you use linMB first. Any special reason?

Har-Bal2 is at beta stage, I believe, and it does have the curve matching. I see so many people dissing Harbal for not having this feature, so they seem to be bowing to pressure.

I wasn't dissing Har-bal myself, if anyone took that. And I wasn't talking about Har-bal when I commented about products that match another eq curve.

Personally - I use it as a spectrum analyser. Then, if I identify a weakness at - for example - 80Hz - I go back and re-mix. Why don't I just "Har-Bal-ize" it with their eq??? ... Because I don't have to - I have full access to my own mix, so why slaughter it with an overall eq, when I can re-mix better?

For example - I might decide to boost the kick. Or the bass. Just eq'ing the whole mix doesn't give me that option. I strongly believe that no effects - whatsoever - should be applied to the master buss for the mix that is given to the mastering engineer. Otherwise, he might have to undo what you've done. He only has two tracks to work with - let him have two virgin tracks.

If you are mastering someone else work - sure, eq may be necessary.

Hmmm...use your ears - yes I see what you mean!

Or for the VST audioist with a little voyeur in their blood Voxengo GlissEQ and SPAN work nicely too. If you have a pile of tunes to put on a CD and need them to be somewhat in the same sonic ballpark export a GlissEQ reference curve(s) from any of the other tracks to any instance of GlissEQ to "see" if there is any really glaring problem, like one song is really hot at 2KHz or something, or another tracks has some crap happening at 10Hz or down there. Set the spectrum averaging to kinda slow, like 4 or even slower at 2 so that overall you could say this song is hot, or this song is really gonna be warm...same thing you do as when you listen. Why use spectrums over listening? You don't, it's an auxillary tool to listening - sometimes I set up a track at 2:30 in the morning - there's another reason. Here's one too - I'm trying to set up some harmonic distortion using a 1KHz tone - anyone want to hear that stuff?

I've got Har-Bal also but am waiting on the new averaging algos Paavo is coming out with - should be exciting! I'm not really into static spectrums - if I can't make it dance to the music then it doesn't mean much to me! Remember the Heathkit color organs? Haha - now we're talkin

ED: Oops - I forgot to mention GlissEQ uses spectrum overlay that work in real-time and are fast enough to compare useful information concerning the overall general frequency differences (bass, mids, highs). THen SPAN is cool to hook up to the master buss so you can set your K-14 loudness (I'm into K-14 personally that is - I'm a recovering K-12!) correctly for each song.

Yeah - I use Span too. Aleksy put in that K-scale metering because I heckled him a bit on KVR. Voxengo stuff rocks!

You did good then! Before that I was headed blindly towards K-6, now I'm a firm K-14 believer (thunder & lightening please!). Following the K-scale of contemporary commercial pop music, mastered and produced by folks that I know know better - and they know I know it! ...is only serving Satan...what the hell does that producer know, hehe

Oops I just noticed Satan is an anagram for Santa - coincedence?

I'm listening to "Close to the Edge" by Yes right now for some fun, it's about K-14 peaking at about -1dbfs on average and -0.1dbfs tops - how do we get back there?

Originally posted by kylen I'm listening to "Close to the Edge" by Yes right now for some fun, it's about K-14 peaking at about -1dbfs on average and -0.1dbfs tops - how do we get back there?

I don't understand how this scale tells you anything about perceived loudness. A peak of 0.1 is an inperceptible fraction of a dB away from digital zero.

And what do peaks tell you anyway? You could easily have a track with an RMS level of -25 that peaks at -0dBfs. That doesn't mean it's mastered loud--it would be quieter than any rock CD mastered in the past 15 years. Does the K system take RMS levels into account?

Yes!

I didn't want to ruin my contextual reply, hehe

K-scale is explained on Bob Katz website and in his "Mastering Audio the art and the science" book for those of us that need to read such things...I have mine right next to my DAW for inspiration!

http://www.digido.com/portal/pmodule_id=11/pmdmode=fullscreen/pageadder_page_id=59

I've used it - I'm not impressed. Admittedly - For mixes that truly sound "not good" it made for a fairly quick "get it halfway decent" pass. For material that sounded good on the other hand...

You use MBC as a default?

Originally posted by MASSIVE Master I've used it - I'm not impressed. Admittedly - For mixes that truly sound "not good" it made for a fairly quick "get it halfway decent" pass. For material that sounded good on the other hand... You use MBC as a default ?

Hey Massive... I would hope you wouldn't be impressed! That's your gig, and I'm sure you do a much better job.

Well, I admit I don't know what I'm doing... I'm not a mastering guy. And no, I don't use MBC as a default, but it does bring a cohesive sound to my mixes if used with taste. Most times, the MB stays in the closet.

I've got ears and I use them.

Pasz - I use the MBC before HarBal because I like the way I can contour the overall mix with the different bands. I gives HarBal less to do...

Official Statement - I AM NOT A MASTERING GUY! But you knew that. If you've got to get things sounding right, and don't have the resources for a professional mastering job, learn how to use HarBal. :)

I love HarBal.

You definitely need to listen to your EQ adjustment. I tried to match to a reference file to a tee once and it did not sound good. I admit I don't have good ears so I find visual cues very helpful. If nothing else, to give me a good starting point. What I also like about HarBal is EQ changes are adjusted for the volume change EQing may create (not sure I'm explaining that well) so my ears are hearing the EQ change results and not just cause it made the whole mix a touch louder.

I've sent work to an ME and definitely worth it but I can't do that all the time. I tend to do a lot of demo-type work so paying an ME is not practical but it's still nice to have a more balanced mix in the end.

If you want a quick sample, check out "Delusional" in my signature. I actually ran it through L1 and then Harbal (the reverse of the normal way I do it) and I think it sounds pretty balanced anyway.

Personally, I've worked with realtime spectrum analyzers and it's not what I need. I need what Harbal does, tell me the cumulative frequency spectrum of the overall mix.

Originally posted by Phil O'Keefe Hey Lee, are you a mastering guy?

Why I oughta!