Dubstepforum.com

nowaysj wrote:Once again, have to turn sigs off to read this damn forum, the formatting is so bad, and so can't check toons.

Soundcloud

defo needs more harmonic in the middle now though.
I remember the subs I had back in the day being more dynamic still thoughh.... *scratches head*
I was probably just a lot higher on weirder drugs.

Memory is shit.

Am on a laptop that actually has a very good sound system for a lappie, and there is no sub there. I wonder where this thing cuts off, maybe 90hz or so

decklyn wrote:Just open up a frequency analyser. A pure sine should be just a slit. If it takes up anymore frequency than that then there is harmonics.

FWIW, a perfect sine won't show up as a slit in a digital frequency analyzer. In a theoretical continuous system it would, but it's a feature of discrete fourier transforms (which I really should learn more about...) that that doesn't actually happen in the digital world.

I'm not 100% sure where I stand on this - I know that wavetable based algorithms aren't intrinsically identical to mathematical wave generation, but I'm not sure whether it's theoretically different in a way that is indistinguishable at real world digital resolutions or whether it actually makes an audible difference. Anyway, I was a bit bored this evening, so I chucked a bunch of waves from Synth1 (decent freebie synth but I'd be suprised if it had a particularly special sine wave oscillator) into audacity and stuck them next to some mathematically generated sines at the same pitches. Anyone want to pick out which is which?

* all in one wav file, 44k / 16 bit PCM
* The order is randomized for each of the 4 pairs, the pitches are from about G#1 down to B0.
* the lengths and fades in and out were done by hand in audacity - so they aren't identical but there won't be any clues to which is which in the attack / release phases
* the synth tones were generated with as close to an init setting as I could get, and then each was peak-normalized to -3dB. The mathematical sines were generated at 0dB and then turned down to -3dB.

https://www.yousendit.com/download/ZW9E ... MlZFQlE9PQ

Personally I'm not sure I can hear a difference. But I'm not listening on great speakers. Now I'm going to stick some halfstep beats over it and send it off to a label or two...

Shit, I'm pretty sure I can hear a difference... maybe a kinda placebo though, I'll use this in my next tune.

the goldbaby packs are a worthy investment as well

i dont get the deal here. but i would or so recomend the goldbaby tape 808 pack. and saturation is your friend.

trying to gte best pure sinewave via digital rendering which orig cooledit was best to do in, but tape 808 you can get a single sine sampled high qual from analog that is a bit better. all about the pure sub line.

slothrop wrote:

decklyn wrote:Just open up a frequency analyser. A pure sine should be just a slit. If it takes up anymore frequency than that then there is harmonics.

FWIW, a perfect sine won't show up as a slit in a digital frequency analyzer. In a theoretical continuous system it would, but it's a feature of discrete fourier transforms (which I really should learn more about...) that that doesn't actually happen in the digital world.

I'm not 100% sure where I stand on this - I know that wavetable based algorithms aren't intrinsically identical to mathematical wave generation, but I'm not sure whether it's theoretically different in a way that is indistinguishable at real world digital resolutions or whether it actually makes an audible difference. Anyway, I was a bit bored this evening, so I chucked a bunch of waves from Synth1 (decent freebie synth but I'd be suprised if it had a particularly special sine wave oscillator) into audacity and stuck them next to some mathematically generated sines at the same pitches. Anyone want to pick out which is which?

* all in one wav file, 44k / 16 bit PCM
* The order is randomized for each of the 4 pairs, the pitches are from about G#1 down to B0.
* the lengths and fades in and out were done by hand in audacity - so they aren't identical but there won't be any clues to which is which in the attack / release phases
* the synth tones were generated with as close to an init setting as I could get, and then each was peak-normalized to -3dB. The mathematical sines were generated at 0dB and then turned down to -3dB.

https://www.yousendit.com/download/ZW9E ... MlZFQlE9PQ

Noone going to tell me which is which, then?

deadly habit wrote:trying to gte best pure sinewave via digital rendering which orig cooledit was best to do in, but tape 808 you can get a single sine sampled high qual from analog that is a bit better. all about the pure sub line.

j

When using an 808 slice you're still going to experience aliasing issues on account that you've only got the sample at 1 pitch so as soon as you try and tune it to a different pitch your samplers aliasing is going to kick in.

slothrop:

Listening on Mackie HR824s:

I feel a bit schooled here.
I would feel very confident recommending Synth1 to people then and will probably go download it right now for easy subs.
This changes everything I thought - through good equipment there is 0 discernible difference to my ears at least. Both shake the room which is all I'm out to do.
The difference is slight at best.

Thanks!!!

the dub lemon wrote:

deadly habit wrote:trying to gte best pure sinewave via digital rendering which orig cooledit was best to do in, but tape 808 you can get a single sine sampled high qual from analog that is a bit better. all about the pure sub line.

j

When using an 808 slice you're still going to experience aliasing issues on account that you've only got the sample at 1 pitch so as soon as you try and tune it to a different pitch your samplers aliasing is going to kick in.

joys of a quality hardware sampler heh

akai samplers don't alias and introduce artifacts as u pitch shift?

The good softsamplers these days have pretty much zero aliasing.

hurlingdervish wrote:its all about the pitch envelopes...

operator in ableton makes NASTY fuggin 808 sounds and sub bass

THiiiiiiiisss!! I only wish it was easier makin it not click, sometimes it can get pretty random but that's what i love about it. And you can always resample the shit out of it for exxxtreme results

future one wrote:The good softsamplers these days have pretty much zero aliasing.

yeah - I totally agree - ended up sampling my Korg MS10 doing sine notes [through Brick tube pre and TL tube comp] chopped up the samples - taken at dim intervals for loading up - works great in Kontakt - link will work for 20 days
Not the purest sine wave, but full of thick analogue goodness...

http://files.me.com/nzhouse/xsra6a

chur!

cixxxj wrote:

hurlingdervish wrote:its all about the pitch envelopes...

operator in ableton makes NASTY fuggin 808 sounds and sub bass

THiiiiiiiisss!! I only wish it was easier makin it not click, sometimes it can get pretty random but that's what i love about it. And you can always resample the shit out of it for exxxtreme results

lol I asked a question about this the other day. the clicks. they're not a big problem though.

The ms10 only has triangle waves as the purest waveform, but with that delicious lowpass filter you can get it sounding really deep. As far as the difference between synthesized and mathematically generated and then sampled sines, there really is no discernible difference. Most synths, even "analog modeling" synths use a lookup table for their basic waveshapes. These use several thousands of values to represent one cycle of the shape and have very efficient interpolation algorithms for smoothing any stepping in the waveform. Also, most modern synths use band limited impulse trains, so there is zero (close enough) aliasing. The power of your sub has more to do with how it is mixed into the track than the minutiae of how theoretically correct the sine is

knobgoblin wrote:The ms10 only has triangle waves as the purest waveform, but with that delicious lowpass filter you can get it sounding really deep. As far as the difference between synthesized and mathematically generated and then sampled sines, there really is no discernible difference. Most synths, even "analog modeling" synths use a lookup table for their basic waveshapes. These use several thousands of values to represent one cycle of the shape and have very efficient interpolation algorithms for smoothing any stepping in the waveform. Also, most modern synths use band limited impulse trains, so there is zero (close enough) aliasing. The power of your sub has more to do with how it is mixed into the track than the minutiae of how theoretically correct the sine is

oops - totally forgot it was a triangle wave - should've looked more closely at the waveform - normally when I've used this on the ms10 the triangle very little middle and high freq content - might have to build me some sine waves..
Still, the pure triangle wave mixed with a heavily effected square wave in Kontakt seems to be doing the trick - not using the lfo/filter on the triangle, but just the lfo/amplitude instead on the tri, then with the square wave I've sampled off the ms10 [opened the filter while sampling] using prophet 53 filter, 4 pole hpf and finishing it off with Twang inside Kontakt - the layer works well with the bottom end and mid-hi sitting nicely in the mix

moznz wrote:

future one wrote:The good softsamplers these days have pretty much zero aliasing.

yeah - I totally agree - ended up sampling my Korg MS10 doing sine notes [through Brick tube pre and TL tube comp] chopped up the samples - taken at dim intervals for loading up - works great in Kontakt - link will work for 20 days
Not the purest sine wave, but full of thick analogue goodness...

http://files.me.com/nzhouse/xsra6a

chur!

MAN THANKS!

I've got the cool edit pack of sine waves on my old HD. Il upload when I get in from work.