| Author
|
Using voxengo span
|
psilonaut
IsraTrance Junior Member
Started Topics :
14
Posts :
35
Posted : Sep 11, 2007 05:36
|
I'm using voxengo span to look at the spectrum of my tracks as I write them and I just realised that if you adjust the slope value it changes the spectrum. For example a lower slope value seems to make the spectrum dip lower than it should in the mid-high freqs and having a higher slope gives a more realistic (to my ears anyway) curve.
So my question is what values for slope, block etc do people use when using voxengo span to make things as accurate as possible? |
|
|
Tomos
IsraTrance Full Member
Started Topics :
84
Posts :
981
Posted : Sep 11, 2007 10:58
|
I'm not sure about Span, but in terms of EQ I would love to know peoples thoughts on EQ slopes and curves. I have NO idea how they affect the sound, except what I can hear. When you've got so many things going on in a mix, I would think a slope is fairly inaudible.
I'm really into cutting unnecessary sound as much as possible using steepest cutoff slopes, and sometimes if an EQ has more than 1 filter, I use all of them to combine the slopes for ultra steep cutoff.
I do it all out of habit, looking at analyzers like Span and seeing how much sound is removed and a desire for the cleanest sound I can get. Whether its correct or not is just a guess. |
|
|
|
vipal
IsraTrance Full Member
Started Topics :
123
Posts :
1397
Posted : Sep 12, 2007 03:11
|
block 2048, speed 64, slope 0,0 max/real/instant/full
pure, 20 20k, scale -10 60
my standard settings for span
loaded a well mixed track, adjusted settings to get the best view and used this as my standard setting.
|
|
|
|
Auralviolence
Started Topics :
0
Posts :
58
Posted : Sep 15, 2007 01:43
|
|
When you change the slope, the DB resolution scale increases. That's the reason why it seems to you that your curve is more stretched verticaly, and, of course, your mid and high frequencies look "lower". If I understood your question rightly... |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 15, 2007 02:07
|
Block 4096, speed 8, slope 3.0, avg/realT/instant/Full
pure, 5Hz - 20 Khz -10 60.
UnderTow |
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Sep 15, 2007 03:37
|
It depends what part of the spectrum I'm trying to analyze. One setting is not good for all.
If I'm trying to analyze the bass I bump up the block size to one of the top 8192 or 16384 - a good idea, as longer blocks can analyze lower freqs better - however, this makes it very difficult to get a fix on top end stuff. I normally go down to 2048 or 4096 for general purpose.
Speed depends on what I'm doing as well. If I'm mastering or mixing I'll stick it on the master out, with the speed right right down, as I'm trying to get an idea of the average spectrum of the mix - I want to ignore the detailed waveshape and get a more general picture. If I'm working with a specific sound for sound shaping duties, or just understanding what is going on with it, I move the speed much higher.
Slope - I've never felt the need to go near this - its a bit of a red herring I feel. 3.0 db seems accurate to my ears, and I can't think why anyone wouldn't have it on 'Full'. Remember with the dB settings in slope, its the same information being presented to you, just in a different way. If I chuck proper white noise through it, 0.0 is clearly the most accurate - however, the human ear doesn't work like this at all! F-M curves for the win!
 .
http://www.soundcloud.com/speakafreaka |
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Sep 15, 2007 16:28
|
Tomos, in response to what you are saying - yeah, in one sense (the most important one at that) there is no need to have a really thorough understanding of *exactly* what your EQ is doing, as long as it sounds better - I mean, that's always the ultimate point of getting out the EQ in the first place.
However, some general hints to make it even better sounding 
Sometimes, the steep steep slopes can do more damge then good - especially when they are the cutting notch sort. Nearly all filters introduce a slight dip or boost around the edge of the peak going in the opposite direction to the one you were after - its quite deliberate, and without it EQ tends to sound quite strange - often we subconciously want this action. The thing is, when you start stacking up really steep notches, you start stacking up really steep boosts around the edge of the notch too, and it starts to ring. So instead of one problem frequency you have two - one on each side of the cut - which require more EQ... and the overall shape you are making to this ringing is two notches either side of lesser reduction, both of which introduce ringing, so more notches to eliminate that and so on and so on - this is just a really deep and quite shallowly sloped cut with slighty dodgily shaped lines - far better and CPU efficient just to use one notch where you can really see what you are up too 
Of course, there are many situations where you want that really steep cut, like if something is sticking up over the the rest of the mix and there is very little either side of it - possibly nothing actually - a lot of the time with the freq analyzer when there is a big ol' dip between two strong peaks there is very little actually in the dip. And then these steep notches make a lot of sense.
And really steep works great if you are doing just small cuts - a few dBs here and there on glassy freqs.
There are no hard and fast rules though - if it sounds good then keep doing it Just be aware that when you introduce a cut somewhere - at edge of the slope you are going to introduce a peak too (and vice versa), and the steeper the cut, the steeper the peak. EQ is great, but less is more I feel. I use the minimum amount needed to get a sound into the mix - not the maximum as I find my mixes sound better integrated for this.
The bottom line is, if I can hear the sound of the EQ, then I've applied too much.
.
http://www.soundcloud.com/speakafreaka |
|
|
|
Tomos
IsraTrance Full Member
Started Topics :
84
Posts :
981
Posted : Sep 15, 2007 19:40
|
Thanks, that's helped clarify some things. It's weird - I know these things, but only when they are restated do they click.
There was a distinctive mid-point in my production history where I became confident in using EQ and various other tools. My production actually took a dip in quality compared to when I was barely using anything. In EQ less is definitely more!
As you say - there are no hard rules, but as a matter of preference, when removing unnecessary low frequencies do you cut, or use a low shelf pulled right down? I would assume the boost at the edge of slopes would be less dramatic than using a high pass cut. I guess its a matter of taste too.
|
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Sep 15, 2007 21:40
|
I am more and more using shelves everywhere in my mixing - I still like to lowpass the very top end of things just a smidge.
With the bottom end, alot of people - even professionals - do not follow best practice and will quite happily tell you to use a high pass on your bass, etc, etc, etc... this makes precisely no sense. Before people jump down my throat, let me explain why this makes no sense.
Its impossible to have frequencies below the funadmental of the oscillator - they just are not there. Span will tell you they are at default block size - but they are not - another good reason to bump up the block size for low frequency work! So, what exactly is the point of sticking a high pass filter at 30 hz when there is nothing there? Well, what you are actually doing is attentuating the fundamental (which can be very desirable) with the edge of the slope - now if you wish to attentuate a specfic frequency, doesn't it make sense to know precisely how much you are attentuating it by, does it make any sense to attentuate through shifting frequency and moving the slope? Of course not, the correct tool for the job is notch EQ. On a kick drum, with a shifting pitch, I can see it makes sense to use a HP, on a bass, with set fundamental frequency it makes absolutely no sense at all, as a notch filter is the tool we use for isolating and attenuating individual frequencies.
.
http://www.soundcloud.com/speakafreaka |
|
|
|
|
