|
Roland JP 8080 DC Offset with the Supersaw OSC (pg. 2)
|
View this Thread in Original format
| johno27 |
| Cool I'm going to do the same. and transpose the pattern up like 8 octaves ;) |
|
|
| Eric J |
OK, so when I transpose the pattern up several octaves, it effect is still present but much less pronounced. The higher I go in the register, the less pronounced the effect is. It is to the point that when I am up in the C8 register its practically unnoticable. If I move down to the C4/C3 register, it is very obvious in the waveform.
I only have the ability to add a high pass filter at about 20Hz since none of my plugins EQ any lower than that, but adding an UAD Cambridge Eliptical 6 High Pass at 20Hz makes no difference to the DC Offset. The Eliptical 6 setting looks to be about 48db/octave. This is the only EQ I have that has anything even close to a 96db/octave filter, so if anyone else has an EQ that will go lower, I'd love for them to try as well.
It appears that the posters above may be corret in that this is just an artifact of so many detuned saw waves.
johno27, let me know if you find anythin else out. |
|
|
| johno27 |
I suspected moving the pattern up quite a few octaves would reduce the problem.
I don't think using the hipass will make any difference in this case. Even though it's always a good idea to filter that band out of most things.
I agree that it appears to be an artifact of saw waves in general, especially on the jp. I guess the move saw's you have the worse it gets :) |
|
|
| johno27 |
I guess the question now is.. what do we do about it, ignore it? I can't think of anything to do to correct this apart from maybe trying to run the signal through some analog gear/filter.
The mastering guy I spoke to said not to worry about it. I just don't see how it's possible to ignore, it's eating up headroom which will cause a problem during mastering as far as I can see. |
|
|
| johno27 |
Eric, can you post us a screen-shot of what your waveform looks like when u use the supersaw patch around C4?
I'm trying to compile some examples to take through to another mastering engineer.
Does it look the same if you switch off the supersaw oscillator and use a normal saw one in it's place? |
|
|
| Eric J |
| quote: | Originally posted by johno27
Eric, can you post us a screen-shot of what your waveform looks like when u use the supersaw patch around C4?
I'm trying to compile some examples to take through to another mastering engineer.
Does it look the same if you switch off the supersaw oscillator and use a normal saw one in it's place? |
I'll post some pics later on tonight when I get home from work.
I havent tried it with a regular saw osc instead of a supersaw osc, but based on my previous experiments, I would theorize that the effect will be less pronounced but still present as long as I have at least two saw oscs detuned from each other. I will try this tonight to confirm.
It is my suspicion that this is an artifact of detuning all those saw waves. My supersaw patches are usually one performance with the upper patch set to a Supersaw on OSC 1 and a regular saw on OSC2, then the lower patch is a Supersaw on OSC 1 and usually a square or triangle on OSC 2. SO that equates out to 15 saw waves all detuned from each other. Thats a LOT of saws!
Hopefully someone who knows more about it than me can confirm this. |
|
|
| Eric J |
Here is a picture of the waveform:
At first the filter is only about 15% open, to it looks pretty even, but when it really looks bad is when the filter opens up past 50% and the amp release is up about about 50%. |
|
|
| Eldritch |
| Well, that's because you are filtering out most of the harmonics. The JP-8000 supersaw is actually a ramp wave afaik (Basically an inverted saw). That's why the peaks are at the bottom of the wave. This is perfectly normal. Supersaws eat up alot of headroom because it's like seven saws playing at the same time. The reason the wave is spiking like that is because the supersaw has more energy in the high frequencies. |
|
|
| camsr |
| multiband compressor. |
|
|
| johno27 |
That looks exactly like the waveforms I get, and I can assure you it's not just the supersaw. The normal saw wave oscillators do it too.
Eldritch, so I agree that this is "normal" in the sense there's nothing we can do about it as it is a by-product of the synthesis, however, surely that wave when fed into mastering is going to create HUGE headaches with any attempt to limit/normalize the audio. It'll be very hard to master the track as "hot" as other releases when all these spikes are playing havoc with the headroom?
camsr, I tried running it through a standard stereo compressor and that wasn't able to remove the "spikes", I'm not sure a multiband would have any better luck and would land up acting more like an EQ with highs being removed overall.
If I can't get this figured soon I'm going to sell all my synths and draw my waveforms by hand from now on!! argh!!! *kiding* :) |
|
|
| aquila |
| Perfectly normal imho too. The transients of a sawtooth (be it a supersaw or otherwise) will always dominate one side of the dc scale. |
|
|
| johno27 |
Aquila, I agree but surely this has to affect the mastering side of the track?
If those transients fell away to a greater extent in the mixdown I could understand, but with a saw sound being at the right level in the mix the transients affect the entire track in the same way.
I dare anyone to take away like the one shown above and try run it through a limiter and see what happens. It's no good, it goes into gain reduction and starts to distort before the peaks have even gone close to 0db. |
|
|
|
|
