Voxengo Premium Membership - All Voxengo Plugins For a Fixed One-Time Fee
Winter Holidays 2017-2018 Discount
Forums     Plugins     Soniformer Soniformer 2 assistance.

This topic was created before release of the latest product version, and it may contain details irrelevant to this version.  Replying is disabled for this topic.

Last

Next

Previous

First



I've had Soniformer, Curve EQ, and Elephant for almost one year now, and no matter how many hours I spend reading the manual, I just can't understand certain aspects.  So..my questions are..What does a flat spectrum look like?  Do I make sure all bars are registering across the bottom of the screen, how high on the bass, how high on the 10k+ range?

Do I match the Threshold line with the spectrum output, meaning if I have a peak at the 106Hz level, and another at the 800Hz level, do I draw down harder, and raise the line during a valley portion?  Or..is it just a smooth decent low to high?

The blue bars pointing downward in spots when I'm looking at the Out Gain section, is this bad, what does it mean, how do I fix it?  Is it worse is the bars are moving down toward the first line below the 0 mark?

Finally, in all my reading about compressing it's stated that release times should be longer than attack times, (attack 5ms/release 112ms), yet some presets in Soniformer are the opposite.  Is this normal for this type of compressor?


Answers on Threshold setting depend on WHAT do you want would you likt to achieve.  If you want to suppress peaks at some frequency then set Threshold lower at these frequencies.  If you don't want to touch some frequencies, more Threshold up at those areas.  Do not forget to adjust Ratio at these frequencies as well.

As for the overall spectrum shape, I think I was not too precise by saying spectrum should be an 'almost straight line'.  It should look more like an 'inverted smiley', with roll-offs below 80 Hz and above 12kHz, and a slight wide bell shape boost centered around 700 Hz or so.

There should be no definitive dips and notches - that's for sure.

As for the attack/release differences, that's totally up to you.  In normal cases setting attack lower than release is a way to go.  If you have enough experience, nobody can stop you from experimenting with an opposite ratio.

daydad: The blue bars pointing downward in spots when I'm looking at the Out Gain section, is this bad, what does it mean, how do I fix it?  Is it worse is the bars are moving down toward the first line below the 0 mark?

There's nothing wrong about blue bars pointing downward - it simply means these frequencies get more of gain reduction than of gain amplification.


daydad: to understand what a 'neutral' well balanced spectrum is, I suggest you download Voxengo SPAN and set it's SLOPE parameter to 4.5dB (or 5dB) and load in your favourite pop-song to your host and watch what kind of spectrum it has.  The chances are that at the slope setting of 4.5dB it will look mostly like a flat line from about 60Hz to about 16 - 18kHz (there it should start dropping quickly).  Most well mastered and balanced mixes have a rather flat frequency response (because this way the mix translates best over most systems).

In soniformer you can achieve this same flat response.  I find it easiest to use the A+C mode and tweak from there so that it looks like a slightly sloping curve, like Aleksey said, an inverted smiley.

There are two very different ways of supressing the offending frequencies.  One is low-level compression (very small ratio, usually 1.1 to 1.2:1 ratio, threshold set to -30dB or even more) or peak limiting (ratios around 8:1 or more and threshold only at the tip of the peaks).  These two methods provide very different results.  Attack and release controls are also important to consider and fine tune.  Usually areas with quick transienst (percussive/picked instruments) can have a fairly quick attack and fastere release but if you have a lot of soft, slow moving material you might get better results with slower attacks and releases.  Note that the bass region (less than 120Hz) can begin to distort and sound weird if you set a too fast release for it.  It might also lose punch if your attack is too fast (sucks the 'movement' out of it).

If you have any tracks that you are trying to master you can send me a rough copy of it and I'll do some tweaking with soniformer and then send you the preset, with a description of what I've done.  This is usually the best way to learn.  Maybe Aleksey could provide some free to distribute track of some kind that needs a bit of 'mastering'?  This could be used as a good track to practise using soniformer and the other mastering tools you offer.  Maybe even consider a whole new section for registered users on the voxengo site called something like "mastering with voxengo tools" or whatever, with downloadable presets and audio clips. :)

Cheers!

bManic


To Aleksey and bManic,

Thanks for the quick responses. bManic, I have downloaded SPAN, but up until now I have not fully understood all the ins and outs.  I will do what you recommend and load some similar material in for comparison.  I don't like using recent stuff for my reference material just because of the way they "Master" these days.  I usually take stuff from the 80s and early 90s, when things still had a chance to breathe.

Anyway, I only use my Voxengo tools for mixdown/master work.  During tracking, and putting mixes together, I try to nip whatever problem area I can in the bud, (i.e.  I record everything line-in to Adobe Audition 1.5 at 32bit/96kHz.  When I analyzed my tracks I was getting nasty spikes at 30k and 48k.  I used a low-pass filter around 15k with a Q of 6 and like magic everything sloped smoothly down from low to high hitting the floor at 22-24k.

I will take you up on your offer regarding rough copy Mastering assistance, bManic, however since my material goes anywhere from 8-11min. and up, I'll send you a 11/2 - 2 min clip of a couple "problem children".  Not sure whether or not it matters, but I do a lot of pseudo-proggy stuff, several synths playing at the same time, with a Mini or ARP or distorted GTR on top of everything.  I try to make good use of panning though.

Thanks again!


A follow-up,

Now I get it folks!  In Curve EQ I was trying to literally flatten each spike to get a "straight line".  I had high control points going all the way up to +18 and lows and mids going down to -15.  Dumb, dumb, dumb!  I understand what "flat spectrum" means and what it looks like.  I also looked at several clips in Soniformer, and while Weather Report tends to favor the high side and has that "wide bell boost" around 700Hz, Rush seems to have two "bells", one at 300Hz and another at 700Hz.  I'm slowly catching on.

Thanks again


By the way, the reason for that 'inverted smiley' (as I have described it) is equal-loudness contour of our ears sensitivity.  At the loudness level we usually mix and master tracks nowadays, C weighting (a specific equal-loudness contour for such loudness level) is most usable.  And it does indeed have that 'inverted smiley' shape.  But usually any kind of weighting is only an unobtrusive helper - of course, you decide how your track should sound like.

And of course, you do not have to forcefully make track appear as a horizontal (or inverted smiley) shape.  It should be only a 'target' shape and not something you ultimately need to follow.  That is it - a helper.

One other problem you should understand when mastering tracks is that ears become accustomed to any spectrum SLOPE very quickly.  For example, if you try to listen to a very bright track for long, you'll quickly realise it does not sound as bright as it seemed to sound before you have started listening.  The same applies to overly dark sounding tracks.

To sum up, using spectrum analysis and selecting an appropriate Slope setting is very important when mastering.  Otherwise you may end up with a 'hyped' sound.


- double post removed -

bmanic: (...) Usually areas with quick transienst (percussive/picked instruments) can have a fairly quick attack and fastere release but if you have a lot of soft, slow moving material you might get better results with slower attacks and releases. (...)

Bmanic, now that puzzles me, maybe you can explain.

As I understand from the manual, the Sonifirmer is a Multiband Compressor with 32 bands with independent parameters for each band.

I learned to set the attack of a compressor a bit up (25 ms or so) if dealing with percussive material to keep the "bang" of drums.

Why do you recommend a faster attack on percussive parts ?

Doesn't that kill my drums' transients ?

Thanks,

Gerry


Hey Gerry,

bmanic has described some really useful starting points for using compression and for Soniformer.  However, there truly are not simple hard fast rules for every aspect of a mix.  At the heart of your question is really what you are after in using compression: what you are looking to achieve.

This gets more complex with something as flexible as Soniformer.  It is made to help simplify a lot of issues in mastering and balancing mixes, but until you get a better grasp of single, wide-band compression on individual instruments, it can be really tough to achieve the same thing with 32 bands of frequency dependent compression.

The fast attack/release time thing in dealing with percussion is helpful when trying to control volume dynamics without altering the overall sound of the percussive instruments.

This works well especially in the mid to high frequencies.  The transients happen so fast with some drums and percussion, that when you go to level out a few decibels of volume it can be helpful to keep the compressor's reaction to the peaks and natural attacks pretty quick.  This also helps avoid the compressor mis-firing and letting some peaks pass without being compressed.

But, you are also right in that slightly slower attack/release is usually intentionally used to get a strong "whack", attack out of the drums.  This is usually done in the mixing phase where the drums can be isolated from eachother so that compression is very specific to each drum.  There are a lot of combinations that work depending on the compressor being used and not only the attack/release times.

A Urei 1176 reacts very differently than an intentionally very clean digital plugin compressor, and both of them intentionally very different from a multiband limiter/compressor/expander.  Some compressors really shine when setting attack from around 14ms-45 ms and release anywhere from 50ms-175ms.

I think it might help you to look at the process like this:

Start with the individual tracks you've recorded.  Look at eq and compression settings for individual instruments as a way to make them balance and interact well with other instruments in the mix.  Be careful with time settings on the compressors used for each track so that they match what is happening musically with that instrument.  Bass guitar and low frequency instruments have to be handled carefully to avoid out of control boominess, muddiness, and compression settings that can lead to distortion.  Individual drums can get the attack/release that makes them hit hard before mixing them all together.

Once there is a balance to the mix, you can bring these elements together better in the mastering phase, with something like Soniformer.  Fixing specific bands of frequencies with dynamic eq and compression helps to control peaks of volume in these frequencies that apply to multiple instruments in the mix sometimes.  There might be a combination of bass guitar and bass drum causing different types of peaks from 40Hz-150Hz, and thus adjusting these bands together with the best sounding settings together, just in that band(s), can help solidify the whole mix where it doesn't affect the timing or dynamics at other frequencies.  Something that smooths out low end energy without messing with your snare attack-that sort of thing.

So, bmanic was offering solid advice, but there are a lot of rules that apply to what sounds good to you, and what helps give you some control in certain areas of the mix without losing something that you like.  Thats the beauty of multiband compression!

Perhaps you can take a specific mix that you have done and try working with it in Soniformer, and see if there are some specific spots in the mix that you want to improve that are giving you trouble.  Try writing down what you are struggling with specifically in certain frequencies or with certain instruments, and maybe we can offer some ideas on specific fixes.  That might help you learn the right approach in other areas as well!

bmanic: (...) Usually areas with quick transienst (percussive/picked instruments) can have a fairly quick attack and fastere release but if you have a lot of soft, slow moving material you might get better results with slower attacks and releases. (...)

straubgx: Bmanic, now that puzzles me, maybe you can explain.

straubgx: As I understand from the manual, the Sonifirmer is a Multiband Compressor with 32 bands with independent parameters for each band.

straubgx: I learned to set the attack of a compressor a bit up (25 ms or so) if dealing with percussive material to keep the ''bang'' of drums.

straubgx: Why do you recommend a faster attack on percussive parts ?

straubgx: Doesn't that kill my drums' transients ?

straubgx: Thanks,

straubgx: Gerry


Just to say, that this is a very interesting discussion, with good things to learn.

Thanks,

Narration

This topic was created before release of the latest product version, and it may contain details irrelevant to this version.  Replying is disabled for this topic.

Last

Next

Previous

First