hey all, quick question, I know that boosting with Neb's great EQs makes sense as it sounds lush, but does it also make sense to cut with them also? or is that a waste of resources, as one might as well just do that with the DAW's in built EQ?
If you cut, as an example, with the W295b, it sounds nothing like an algorithmic DAW EQ. I don't know what it is but you can dial in pretty severe negative levels, and it still sounds natural in a way.
yeah. If you notice when you do a cut with a standard digital EQ....if its the lows it tends to rob all the power and leave the signal like a skinny twig. If you cut the highs it makes a muffled dull sound.
My theory is because digital filters are too literal and static to be musical. They literally cut freq taking away the life and musicality of the source. The reason Nebula sounds better both cutting and boosting is because it retains/adds to the dynamics and harmonics of the sound when changing the EQ. This then means there is still some movement in the sound - its not been 'zapped' with digital precision. Essentially - to change the EQ of a source sound in a musical fashion is about more than simply EQ. Most digital filters only focus on the EQ.
This phenomenon became most clear to me when playing with my Moog Source synthesiser. With software synths and filters - if you use a low pass filter on your sound - it is so literal that it completely cuts off the high freq. Whilst this is precise and accurate for the task - it actually makes for a dull (as in - lifeless unexciting) sound.
I was trying to understand why when I had the low pass filter closed in my real analogue synth - the sound was still exciting even though the filter was closed - there still was the 'feel' of high freq existing in the sound. It had 'air' and life to the high freq even though the sound had been filtered. This must be because of subtle resonances happening as a result of dynamic and harmonic movement in analogue circuits.
Nebula samples this movement - so when you make a cut - the sound still breathes
I hate cutting with algorithmic eq's as well as boosting. In fact, I just hate algorithmic eq's, LOL! When boosting and cutting with Nebula, I always get what I expect. What's even better is that boosting/cutting with different eq's like CDSM's 1084 or Alex B's 1081, really shows which eq has which character/vibe. It really feels and sounds like a color palette, instead of shades of grey with algorithmic eq's. A little experience with hardware doesn't hurt either.
SWAN wrote:..there still was the 'feel' of high freq existing in the sound. It had 'air' and life to the high freq even though the sound had been filtered. This must be because of subtle resonances happening as a result of dynamic and harmonic movement in analogue circuits.
Nebula samples this movement - so when you make a cut - the sound still breathes
i don't want to sound like i'm making nebula out to be anything less than what it is here, but i feel some need to say that, for the vast overwhelming majority of 'eq' programs made for Nebula, this statement is only part right. the majority of nebula EQs- by far- do NOT sample dynamics. if you actually do the math involved in sampling you can easily see why- it would require thousands upon thousands of samples- for a single program. for each control you sample, you have however many positions on that control that you are going to sample that control at. some eqs have a few fixed selections for the frequency of a peak/cut filter, so in that case you would just count those selectable frequencies. maybe there that control only has 5 positions. other EQ's offer continuously variable frequency selection, so if you want to sample that behavior, the position count for that one control just shot up to possibly hundreds (if you want smooth transition across the frequency range).
then you have your other controls. let's look at gain. say your EQ allows +/- 12db boost/cut for the band. as a developer you have to decide how much resolution is 'ok' for the end result. you have to weigh all of the other controls, how many positions sampled they need, and then you decide where you can get away with more or less resolution (positions sampled). gain can be interpolated much better in nebula than frequency differences, so you can get away with less positions sampled on a gain control. lets say you choose to sample 5 positions there- one at 0db for flat (although NAT allows you to sort of skip this one, and only use one flat for all frequencies), one each for +/-6db, and one each for +/-12db. sounds like big jumps right? ok, so you could go with 3db steps and use 9 positions.
lastly you have width. you would need to actually look at the response of that control on an analyzer while you sampled to see how many positions you need here- to see how it changes shape as you turn that knob. you want to make sure the interpolation that Nebula will do later will be able to smoothly transition between the positions you pick. so lets say you use 5 for width.
the math then, that you need to do is like this: you multiple all of the positions for each control by each other. so you have 5 selectable frequencies X 9 gain selections X 5 width. that's already 225 samples. not so much. plenty of EQ programs go higher. only thing is that only accounts for 1 kernel. so say you want 8 kernels. 8 X 225 = 1800. not sure if i've seen too many programs go that high.
ok now- that's withOUT dynamic steps. let's add 5 steps at 5db increments (which only gives you a 20db range, and not a whole heck of a lot of resolution in that range). 5X1800=9000. consider how many samples you would need if instead of a fixed frequency selector with 5 positions you did a smoothly variable eq where you sampled 100+ positions. or if you wanted a full 10k. you'd need maybe 100,000+ samples.
Nebula eqs are almost NEVER dynamic. the only exceptions i know are henry olanga's titan eq, which can have dynamics because it does away with the controls and offers the different positions to you through various presets, which is why i like that set so much- it DOES capture dynamics (even though there is a big tradeoff for not having much control), and then there is my own stuff from my freq tweeker sets. so obviously now it may look like that's why i went through all this trouble of explaining this, to promote my own stuff. well, no. the 'eq' i did for frequency tweekers 2 is simple, with only a bass and treble knob. only uhh, i don't remember, maybe 5-8 positions were sampled on each knob. then i forget how many dynamic steps but not as many as a typical dynamic program has, so there was wider gaps between the steps. this STILL amounted to a few thousand samples. it's as simple as an eq can get and to have dynamics it still takes thousands of samples. also, it's more of a specialized thing, being from a weird old tube amp, so i'm not saying it compares to the high-end EQs other devs have done, nor could it substitute for them.
you are right about the harmonics. i think that's the main reason you hear more of an analog quality with nebula EQs compared to algorithmic ones. that and the fact that there are frequency specific phase shifts and things that Nebula also captures, and that stuff contributes to the 'sound' of things. but by and large, nebula eqs lack dynamics. anyway, i'm not knocking nebula EQs at all. it's just that i do have some kind of interest in the process, being a dev, and so i would like to see a bit more understanding about what goes into doing this stuff, since it takes so much time and effort on the part of all devs. i felt like henry should have actually mentioned how his titan eq featured dynamics, when he plugged it here in the forums, but i don't think he did in that thread. the eq lacks controls because of how he sampled it, which he mentioned, but the tradeoff gave it something else that no other high-end nebula eq has, which he didn't mention. any dev has to make those kinds of choices when they set out to do something. i think it's cool that he did that set differently, which nobody else had even considered to do an eq that way, even if it sounds like he only did it that way kind of accidentally (for lack of time in the studio to do a 'proper' eq). i think he maybe inadvertently opened a door to an alternate way of doing things...
Hi Tim - I must have misunderstood some of nebula eq...I remember Alex B stating that his nebula eq had 'dynamics'...maybe he meant in a different sense...but in my head it started to make sense why they sound different...thats interesting...so its just the harmonics...