Mastering Limiter vs. Wave Hammer

Post » Thu Jan 13, 2011 1:32 pm

First of all, before anyone says do a search, I have placed the exact subject line in a forum search, then I removed the "vs." and did another search, then I did a Google search. I can't find this answer anywhere.

I am trying to use my Sound Forge to give me as good as possible fidelity for mix cds,most of which use multiple sources. Since all the nice mastering engineers have differing ideas about where the peaks should hit, I have wildly differing volume levels.

I'm not trying to compress the sound, I'm trying to listen without having to run to change the volume control at every song.

I've been working with the Mastering Limiter at settings Threshhold:-1.5dB, Margin:-1.5dB, Character 7.00 wiht the prevent inter-sample clips ticked. I get good results unless the recording levels are really low, in which case the -1.5 on the Threshhold doesn't do any good. But it seems like when I raise it to something around -10dB, the sound gets more compressed. Since I don't understand the help file (which is par for the course), I don't know if I'r really screwing something up.

Or is the Wave Hammer better to use? The best sound I've gotten so far is bypassing the compressor,then setting the Volume Maximizer to a threshhold of -7dB and output level at -1.5dB, with longer lookahead ticked and a release time of 1,000ms.

Could someone who understands the theory share their experience and intelligence with me? And if there is a better plugin in the SoundForge family - or in anything that Cakewalk had ever put out, like Boost 11 and the Sonitus stuff - could you let me know?

Thanks in advance for any good soul who provides much needed assistance.
User avatar
Antonio Gigliotta
Posts: 501
Joined: Fri Jul 06, 2007 7:39 am

Post » Thu Jan 13, 2011 2:10 pm

Firstly, the wavehammer and limiter are 2 different beasts with differing intentions.

It sounds like for the most part, you are getting good results and what you want with the mastering limiter (not suprising - iZotope make damn fine products). The wavehammer is a compressor and maximiser, different to a limiter. & I see you are bypassing the compression stage of it...

So you're getting good results except for the quite recordings - then why not boost those levels, and stick to a pattern you have working for you? Either normalise, or run through a compressor with gain boost to bolster up the quite recordings. It's hard to know without any idea what sort of dynamic and style the audio has.

And yes - many people certainly will tell you differing levels to aim for! I would have thought the RMS level would be what you what want to balance out between songs in your case - as your are aiming for more of an averaging of levels>?

An always popular and very highly rated plugin is another iZotope product - Ozone. It features what you get in the mastering bundle, and them some! Also includes a helpful maximiser and the presets will probably get you sorted out in little time.

That any help for a starter?
User avatar
Brandi Norton
Posts: 530
Joined: Fri Feb 09, 2007 3:24 pm

Post » Thu Jan 13, 2011 12:57 pm

Huge help. Thanks so much for taking the time.

Let me answer a few of your questions. The music is generally jazz, R&B or rock that still has some dynamic range. I tried using the RMS method with a few different programs and the results were too loud and sounded a little grating to my ears, as though certain midrange freqs were being pumped up instead of preserving the signal and just bringing all the levels into parity.

Since I don't understand the difference between any of these program's functions, I may have just made some critical errors along the way.

I'm still confused about some theory. I've tried some pretty extensive tests on the Wave Hammers Sound Maximizer and the Mastering Limiter on jazz and older, less compressed rock and found the Mastering Limiter to provide slightly more realistic tone with decays being the main area of difference. But the differences were slight.

Which is why I was hoping for a little guidance on a purely theoretical basis.

And the reason I ask is, I don't know the difference between any of these devices, purely in terms of what each does to the signal. If it isn't too big an imposition, I would love some guidance, either to a website, or if it's really simple and doesn't take too much time, perhaps the definitions here. I suspect there are several names for the same thing, but I just don't have the theory background to know how each manipulation differs.

As an example, a jazz person like Diana Krall will not want her music compressed, other than maybe just a hair to give it an anolog sound. So when a good mastering engineer sits down with the raw tapes, how will he or she bring each of the levels to a similar place?

I hate the sound of most radio rock today because of the extreme compression. I used to be a DJ and I can tell you every radio station I worked for, except one classical station, compressed the hell out of the signal so dial twiterrers would find our station amongst the sea of others. When you add the rock compression (aimed at MTV, maybe?) you get a dynamic range of about 1dB.

So I have some experience with compression, and Im sure you can use it sparingly to get the sound up. Rhino Records generally does a good job of remastering, and when you put the original and the Rhino up on Sound Forge, you see one reason is theyve boosted and compressed the signal. But their goal is to make the records sound better than before. My goal is to make them sound the same as before with each song being at the same level.

I know one solution would be to compress the signal then boost it. Is that what normalizing is, but all in one step? Using both ends of the Wave Hammer, I find I have to use a very gentle hand with the compressor, and sometimes don't even use it.

To go back to Kennymusicman's helpful reply, I now understand they are 2 different beasts with different intentions. But I don't know what the difference between the intentions are for a limiter, normalizer, volume maximizer, or anything else that would help me accomplish my goals.

Anyway, to accomplish my goals of futzing with the fidelity of the original signal as little as possible and bringing all the songs into a similar level, should I use the normalizer, mastering limiter or wave hammer or something else. Oh, I forgot another one Ive tried and gotten good results, though not as good as the mastering limiter: Replay Gain.

I have the feeling I may be asking a question where the answer is that x and y do the same thing, but different companies want to differentiate their products, so one calls it x and the other calls it y.

Thanks again for your help.
User avatar
marie breen
Posts: 500
Joined: Wed Aug 02, 2006 10:50 pm

Post » Thu Jan 13, 2011 5:11 am

I'm not trying to compress the sound, I'm trying to listen without having to run to change the volume control at every song."

If that i sthe case, then load the tracks up into CDA and drag down the levels of the louder ones until they all match. ANYTHING else does require compression.

Mastering limiter (I would imagine - I haven't been into it deeply) is presumnably a hardish-knee limiter to stop peaks from hitting 0dBFS. WH is both a compressor and volume maximiser with a lot more controlable parameters than ML..

I the freq dependant dynamics is OK, I would stick with WH or ML.


User avatar
Posts: 507
Joined: Wed Jul 26, 2006 3:43 pm

Post » Thu Jan 13, 2011 11:11 am

I'm still a litle confused about what you are asking, so let me tell you my usual proceedure. First I normalize - what I normalize to varies... from -0.05 to -2 or 3. Then I many times apply a little lite compression 2:1 at minus 2 - 4 db. To me it just tightens thing up.

Others may disagree, especially since compression varies from project to project and style to style.

BTW, I do not ever use the maximiser.
User avatar
Ice Fire
Posts: 515
Joined: Thu Nov 15, 2007 9:27 pm

Post » Thu Jan 13, 2011 9:36 am

This goes back to my not understanding the theory behind the plug-ins. I seem to remember Normalizing causing a lot of controversy. I guess I don't understand which of the plugs introduces the least change to an audio file, while still getting everything to similar levels.

Thanks everyone for bearing with me here.
User avatar
Annick Charron
Posts: 499
Joined: Fri Dec 29, 2006 9:03 am

Post » Thu Jan 13, 2011 8:45 pm

The only thing that causes the "least amount of change" to the signal while getting the levels equal is your brain. If that's truly your goal then what Geoff suggested with CDA is purely your only option.

- Normalization - can't take into account that some songs are much more dynamic than others. Say you have two songs, one of which varies from -12dB to -6dB and another that is mostly -30dB with a momentary peak at -1dB. Normalizing will raise the first one 6dB and the second one only 1dB. After normalizing the first will still sound much louder than the second. Suppose you have two songs, one of which varies from -30dB to -24dB and the second from -18dB to -3dB. Before normalizing the first will sound much quieter than the second. After normalizing it will sound much louder. You'll still have to manually adjust afterwards to match perceived volumes.

- Compression - deals with the problem above by "squishing" the sound so that there isn't as much difference between the loud peaks and the quiet passages. It's a lot easier to get a consistent volume. Wave hammer can be automatic enough that you don't have to adjust anything at all from one song to the next. However, the more automatic and easier you make it, the more it alters the dynamics to get there.

To really get all the songs to the same volume without compression you're going ot have to manually adjust the volume level on each one individually until they sound the same to you.
User avatar
Louise Lowe
Posts: 501
Joined: Fri Jul 28, 2006 3:08 am

Post » Thu Jan 13, 2011 6:37 am

Thanks for the input everyone.

I understand now that the reason I couldn't find a solution is that there is no solution.

Here's what I settled on. I'm keeping two sets of each file (hard drives are so cheap these days), one that is pure and unaltered, and another where I have used the Mastering Limiter set at a threshhold of 10dB, a limit of 1.5dB, the Character at 7 and the prevent interchannel clipping ticked. I use that one for casual pleasure listening.

Thanks again for helping.
User avatar
Posts: 526
Joined: Fri Jun 23, 2006 6:49 am

Post » Thu Jan 13, 2011 8:49 pm

Just another bit of input here... sometimes the solution is not louder but quieter. If you don't want to destroy the dynamic range of tracks that aren't compressed to hell then you can use the RMS normalizing function in soundforge to reduce volume as well as increase. Scan the overall RMS levels of some of your quieter tracks and set your RMS levels in the Normalize function to just below that. You'll find by doing this you'll pull DOWN the levels off loud tracks and maintain those of the softer ones. This is a good way to get a happy medium with many audio files without destroying their dynamic range. Provided you have a decent playback system it shouldn't be a problem as in the end, you need to push the amp a little louder overall to compensate for the volume loss on the tracks themselves but it'll save you from rushing to that control knob every 5 seconds.
User avatar
phil walsh
Posts: 517
Joined: Wed May 16, 2007 2:46 pm

Return to Sound Forge Pro