Jump to content

Harbal EQ 32 bit Sonar 64 bit


Recommended Posts

  • Members

Hey

I've just seen the Harbal mastering EQ in the lastest Sound On Sound magazine...

 

From what I read it sounds amazing.

 

Over the last 4 weeks I've just been evalulating SONAR 5 and I'm pretty sure I'm going to get it ASAP... I want that 64 bit mix engine to preserve the clarity and dimension of my mixes.

 

OK... so I'm not a total idiot... I know that as this stage my mixes will end up at best as 16bit Cd's or possibly as MP3s...

 

But I think you need to keep your signal as high of a resolution as you can before making a mix and mastering it.

 

I now plan to mix AND master in SONAR 5 utilizing the 64 bit architecture to the fulllest advantage.

 

The Harbal EQ site says:

Sampling rates are any; bit depths include 24 and 32 bit; FIR implemented using 32 bit fixed point arithmetic with a theoretical resolution of 20 bits.

 

Does this mean that if I apply the Harbal EQ at any stage to a 32bit mix or master that I have done in SONAR 5 utlizing 64bit processing, that I am taking a 'backwards' step in my process??

 

Paspallum

Link to comment
Share on other sites

  • Members

It almost sounds like you think Har-Bal is a plugin. It is a stand alone program.

You load an already mixed song into it.

I record and mix at 24 bit and then load into Har-Bal, then later use other 24 bit processing programs.

After all that I just reduce to 16 bit or MP3.

Can you mix and output a 64 bit file in Sonar?

Link to comment
Share on other sites

  • Members

First of all, I'm a huge HarBal fan. I used it to "master" the audio in my NAMM performance, which came from a camcorder mic...the quality was terrible until it got HarBalized. I mean, it still doesn't sound like it was recorded in a quarter million dollar studio, but it doesn't sound like it was recorded with tin cans and a string, either :)

 

As to the bit resolution, yes, HarBal is stand-alone so you could export your Sonar mix as 32 bits and then bring it into HarBal. Will you hear a difference? I doubt it. The 64 bits is really useful for preserving headroom, and for "in the box" summing. But once it becomes a stereo file, I doubt you'll lose much by going from 64 to 32 bits. And you'll gain a lot by using HarBal.

 

Just remember HarBal is not a magic bullet; you need to know what you're doing. The way I sum it up is by saying that if you already know how to master, HarBal will help you make better masters. If you don't know how to master, HarBal will help get you partway there but you still need ears.

Link to comment
Share on other sites

  • Moderators

Originally posted by Anderton

First of all, I'm a huge HarBal fan. I used it to "master" the audio in my NAMM performance, which came from a camcorder mic...the quality was terrible until it got HarBalized. I mean, it still doesn't sound like it was recorded in a quarter million dollar studio, but it doesn't sound like it was recorded with tin cans and a string, either
:)

 

I just did the same thing with a friends live stereo recording. We all know how those sound, fairly good quality but it's a stereo mic placed semi-randomly in the room.

 

I used HarBal to quelch the out of control low mid room resonances and balance the spectrum a bit then used Sound Forge's Acoustic Mirror to put it in a nicer room and... not too bad.

Link to comment
Share on other sites

  • Members

I've been reading threads on HarBal ever since it first showed up. It was generally showered with offal in the early-on threads not least for the extravagant claims the company was making for the product.

 

Slowly, cooler heads at least partially prevailed until there seems to be a sort-of consensus that it is what Craig says it is - no magic bullet but very useful and rather astounding considering it's just a $95 USD piece of software.

 

But loud sniffs and rudenesses are still heard all around the pro studios whenever it's mentioned. Ok, if it's for amatuers, I can live with that.

 

The website still leans toward hyperbole hyper-hype, but I'm thinking that I might try the thing - at that price I could live with it even if it only did 1/3 of what it claims. Being a lowly home-studio amatuer.....

 

nat whilk ii

Link to comment
Share on other sites

  • Members

>

 

Sonar does have a 64-bit audio engine, even when running with 32-bit machines and operating systems. It uses the 64 bits for double-precision calculations. Of course it also runs with 64 bit operating systems as well.

Link to comment
Share on other sites

  • Members

>

 

Well, I just got nominated for a mastering award for work I did on an acoustic a capella recording and used HarBal, so let people say what they want...but I do agree that the initial hype was deafening. In fact, I asked Mitch Gallagher at EQ to review HarBal so I could expose it for being the "snake oil" it obviously was...until I actually worked with the program, and was astonished at how effective it could be. I was able to solve some mastering issues that I hadn't been able to solve with other EQs. Since then I've been a fan.

Link to comment
Share on other sites

  • Members

 

Originally posted by Anderton

>


Sonar does have a 64-bit audio engine, even when running with 32-bit machines and operating systems. It uses the 64 bits for double-precision calculations. Of course it also runs with 64 bit operating systems as well.

 

 

Yes, that's fine, but the OP seems to think that Sonar's 64 bit archetecture means that it will be outputting 64 bit wave-files, which I'm pretty darn sure is wrong.

 

Please correct me if I'm making a fool of myself...

Link to comment
Share on other sites

  • Members

I'm not Craig but maybe I can slide this in:

 

Sonar can output in formats 'up to' 64 bit.

 

The double precision is generally considered to be for minimizing rounding error in doing internal processing, since most processing math results in higher resolution results. By keeping the internal resolution higher than the i/o resolution, you help to preserve more accuracy. Digital processing devices have long used higher internal resolution. In the old days, digital processing boxes would often perform their internal math at 18 or 20 bit resoltuion, even though the i/o was limited to 16 bit.

Link to comment
Share on other sites

  • Members

Kurfu

you say

 

I'm not going to call you that.

 

But NO I do not think SONAR can output 64bit files.

 

I do know it is the only DAW that currently sums a mix with 64bit precision...

 

I choose my mixes from SONAR to come out as 32 bit wave files....

 

I just wanted to know if I did a 32bit mix (utilizing SONARS 64 bit engine) and then applied Harbal (yes '60Hurts' I know it is not a plugin!) would It damage my mix?

 

(apparently Harbal are planning to release as a VST plugin sometime in the future?)

 

Well anyway Craig answered my question really well.

So Thanks to his advice... I can trust Harbals usefullness.

 

I rarely ever use EQ when mastering.

So I guess I'd master my mixes in SONAR... (some compression etc) and then I could use the harbal on that mix to make sure my EQ's are 'in line' with the proper frequencys... Then I could dither down (from Sonar again.. or Wavelab) to 16 bit for CD.

 

See recording is a full-on hobby for me... I make no money from it

I have some really nice monitors... but have been caught out on at least three ocassions in the last 5 years where I've done mixes that had a really 'woofy' sub-bass.. at a frequency that my monitors don't reproduce... (yeah they are nice monitors... very flat... nice stereo field.... NO user fatigue at ALL... but their downfall is that don't reproduce ALL of the lower frequencies -and I don't want to get into buying a sub speaker)

 

So these certain mixes which sounded fine in my studio 'embarrassed' me when I heard them on other peoples systems that had all that artifiical 'sub-bass' processing.... they were REALLY bass prominent and woofy.

 

See I love listening to R&B, Hip-hop and Rap... but that's not the type of music I make myself... my music is rock...folk..folk-rock...

 

Anything from oldshool metal to Neil young... but not anything with a 'thud...thud... huge 'hip-hop' bass thing going on

 

So I figured this Harbal might be able to check that issue for me.

From what craig says... I might have to get it.

 

thanks guys

 

Paspallum

Link to comment
Share on other sites

  • Members

Craig

I have a couple of questions for you regarding SONAR 5 (and YES relating to that wonderful 64bit engine!)

 

 

You know how in both CUBASE and SONAR 5 (and probably all DAWs) you can apply audio effects (plugins) over a full track as a 'process'? you don't put the effect in the Plugin signal chain or bin... and bounce it down in real time.... you just select the track or tracks and use the menu to process them with the desired plugin and it prossess like soundforge or wavelab would.

 

Does this still utilize SONARS 64 bit processing engine? As it is not going through a mixdown or bounce? so it is not actually routing through the 64 bit mixdown engine?

 

Sometimes I have standard plugins like the Antares Mic modeller or a Tape Saturation Sim. that I don't have to 'hear' (I have a standard setting) I just want to quickly 'apply it' without having to do a bounce and I'm just wondering if it would be the same sound quality as if I did bounce it?

 

ALSO

 

I read that with analog the loudest elements get squashed and tend to get mushy (compressed) while the softer parts remain clear (as long as they have been recorded well above the 'noise floor') but with Digital the loudest parts are clear and the softer bits get 'brittle' and 'murky'...

 

I'm wondering if Normalizing a recorded track (or every individual recorded track) in SONAR 5 would keep the lower harmonics of the track up at a higher resolution...resulting in a warmer and clearer piece of music??

 

My favourite recordings are really from the 1968-1975 period when the gear was clear and warm and just magic... I like the production on the later Beatles stuff... CSNY, Neil Young, '69 to '74 Elvis... Stuff like that....

 

Most (not All) stuff pre-1968 can sometimes be a bit 'hard', 'boxy' and basic... and well after 1975 until now... well it's all so different... some good.. some bad...(most good) so many options now....

 

But I think that keeping clear, defined harmonics in the music right through to CD stage is one thing that will help me get 'closer' to that sound I like. Will Normalizing help me attain this?

 

What do you think?

 

I know these might be 'dumb' questions.. but you only get clever asking what you don't know... not pretending that you know everything!

 

Paspallum

Link to comment
Share on other sites

  • Members

 

FIR implemented using 32 bit fixed point arithmetic with a theoretical resolution of 20 bits

 

 

That's what I don't like to see - 20 bits.

 

I bought Harbal, and use it as a spectrum analyser. (From that point of view, the 8 bit demo would probably suffice).

 

I choose not to eq my mixes with Harbal. Not that I think it's a bad eq, but I believe that spectral balance issues are best resolved in the mix, on an individual track basis.

 

E.g. - if you've mixed the kick too loud, but the bass is ok - why would you want to eq the stereo mix, affecting both kick and bass together, when you could simply lower the kick fader? Or change track eq settings?

 

I don't like the idea of running my good 32 bit mixes through a bunch of FIR filters that have an effective resolution of 20 bits.

Link to comment
Share on other sites

  • Members

 

Originally posted by Kiwiburger





E.g. - if you've mixed the kick too loud, but the bass is ok - why would you want to eq the stereo mix, affecting both kick and bass together, when you could simply lower the kick fader? Or change track eq settings?

 

 

I think the kind of answer you'll get to this question from the pros is that the mastering process is supposed to be a subtle polishing, smoothing sort of fix, not the kind of fix for targeting something like a single instrument.

 

Probably the only real answer to your question is found in the actual experience of mastering, where you can hear what the engineer is doing to the tracks, and how the tracks are being changed to fit into the context of the entire album.

 

But the real experts will probably chime in here, so here's Mr Amatuer signing off for now...

 

nat whilk ii

Link to comment
Share on other sites

  • Members

Kiwiburger

Thanks for your input.

This is what I was concerned about. (this 20bit thing)

 

(hey and we DO all know it'll end up at 20 bits eventually)

 

I agree with you.

 

In the old days (now too probably) someone would mix a song or an album to 'two track' tape... and a mastering enginner would have to fix all the problems with EQ and multiband compression.

 

I myself don't really use EQ over my mixes for the same reasons you mentioned... I just throw up a 'mastering' compressor plugin on my SONAR or CUBASE mix on my main bus and get an idea of how much compression is needed... then I'll adjust the pan, eq and volume of each individual instrument or track BEFORE it is summed to a stereo track.

 

I was just thinking that the Harbal could 'filter' out any frequencies that shouldn't be there that my monitors are not revealing...

 

Yeah I've got access to audio systems that will 'reveal' these frequncies... but I gotta burn a mix to CD... take that to whatever player I want to listen through... I thought teh Harbal might be a 'one-stop' saftey filter.

 

hey it's good to hear advice from both sides of the arguement!

 

Paspallum

Link to comment
Share on other sites

  • Members

 

I think the kind of answer you'll get to this question from the pros is that the mastering process is supposed to be a subtle polishing, smoothing sort of fix, not the kind of fix for targeting something like a single instrument.

If we are talking about serious commercial release quality, then you would be giving the mastering engineer a high resolution (24 bit 96khz, or whatever your project rate) file.

 

For serious work, I don't recommend any process over the 2 buss at all. Not even a limiter - you shouldnt be approaching zero: you should be leaving headroom for the M.E.

 

There is no justifiable reason for mutilating your 2 buss mix with cheap tools, when the M.E. has superior unaffordable-by-mere-mortals tools.

 

You don't want your M.E. to have to use those tools to un-do some mutiliation you've applied with cheap tools. Give him the untouched mix.

 

Because the M.E. only has two tracks to work with, obviously the M.E. has limited options. Very possibly there will be a problem with a specific instrument, and he will need to use every trick in the book to fix it - including Mid/Side processing and compression and eq tricks.

 

BUT - anyone using Harbal is doing home-style self-mastering on material that doesn't matter so much. Provided you keep your high resolution files with NOTHING on the 2 buss, in case the record company calls.

 

But if 'self mastering', you have access to the multitracks that a M.E. simply doesn't have. You have no excuses.

 

That's why I like to use Harbal as a tool to improve my mixing - not as a band-aid.

 

And whatever you do, don't Harbal the mix that you give to a M.E. But if you can use it non-destructively simply as a spectrum analyser, so you can improve your mix, I don't see any problem.

Link to comment
Share on other sites

  • Members

 

Originally posted by Kiwiburger

There is no justifiable reason for mutilating your 2 buss mix with cheap tools, when the M.E. has superior unaffordable-by-mere-mortals tools.


But if 'self mastering', you have access to the multitracks that a M.E. simply doesn't have. You have no excuses.


That's why I like to use Harbal as a tool to improve my mixing - not as a band-aid.

 

 

Having read your expanded viewpoint, I think you're right.

 

But I would like to try this when I get HarBal - do a spectral analysis of the original 2 tk mixdown. Note the details to fix. Then a)do the fixes track by track and remix to 2 tks, and b) let HarBal do it's magic thing on a copy of the original 2 tk mixdown. Then compare.

 

It would be interesting to see if, in certain situations, HarBal, strapped across the 2-buss, could actually outmaster my track-by-track manually remixed/fixed 2-tracks. At the very least HarBal might teach me a thing or two about my mixes.

 

nat whilk ii

Link to comment
Share on other sites

  • Members

The big thing to remember, if you are doing your own mastering, is that your ears are going to decieve you.

 

Unless you have spent squillions on the perfect acousticaly treated room, and high end speakers, and have accumlated years of experience ... you shouldn't be mastering your own work.

 

Assuming your room has defects, you mix will have been affected by these defects. If you then master in the same room, those defects are still there.

 

In otherwords - if you didn't get the mix right, you aren't going to fare any better trying to self master.

 

That's why you use Harbal - you get a graphical representation of the spectral balance, which will show if you are in the same ball park as commercial mixes for that genre.

 

If you are judging the outcome of this self mastering with your ears, you will fall back into the same traps that caused you to mix it wrong in the first place.

 

What I choose to do is simply mix as well as I can, comparing to reference tracks. I don't worry about loudness - just spectral balance.

 

I then export a mixdown, and analyse it with harbal. If it indicates a problem, I then go back and fix it in the mix. I have far more options that simply eq'ing the whole mix, with Harbal or anything else for that matter.

 

For self-mastering, all the remains is to make it as loud as the client wants. If I need wider stereo, I fix it in the mix. If I need to change eq, I fix it in the mix.

Link to comment
Share on other sites

  • Members

Kiwiburger....

 

I'd agree with every word that comes from your last post on mastering.

 

But some of us (me for sure) - (you??) do music as a hobby... I don't play 'playstation' or X-box' my 'task' is to try and master my own mixes.. and yeah it is nearly impossible ...fresh and 'un-emotional' ears (with GREAT gear and years of experience) are what is needed... but I don't have the money to send my 'private' mixes out for mastering..

 

Fear not Kiwi-burger! Just in case the 'record company ever does call' I always do a 32bit stereo mix with NO bus processing at all... then once I have that I can try OZONE or T-Racks or any of my Waves processors... and pretend I'm a mastering wizard! They never sound as good as 'industry standard' mastering jobs.. but they always sound a lot better than the 'un-mastered' mix!

 

I don't 'squash' them too much anyway.. I like an open mix with lots of space in it (if possible)

 

Thanks again for your reply!

Paspallum

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...