Submit bug reports and feature requests for the JJOS-XL and 2XL
By boy_blunder Sat Mar 10, 2012 8:09 pm
It would be good if you could choose if you wanted a straight or curved adsr envelope so you could get a more natural decay and more ways of shaping sounds.

You could have the option to have (attack/decay curves) straight/straight, straight/curved, curved/straight etc.
User avatar
By bliprock Sun Mar 11, 2012 7:50 am
it all ready has a curve you just dont see it. I am sure these are 2 pole filters and so this has its own inherent curve already.
its not something you can understand unless you are familiar with the mathematics of filters. ie
http://www.timstinchcombe.co.uk/index.php?pge=poles

there that saves me explainin it.

So its just a graphical representation of a filter, and therefore I do believe it gives you a good idea, but not absolute true real life curve of ADSR.

See its complicated, and so we dont see that, instead we see an artists impression to what is really going on.

http://www-k.ext.ti.com/SRVS/Data/ti/Kn ... aqs/lp.htm

thats a simple idea, and its mathematics are used maybe in JJOS and Aksysy, but not exactly sure how the implementation is, but its a 4 poles total over 2 filters. I think JJOSXL he split it so its, 3 poles for first and one pole for second, not sure. Maybe just 2 poles each as original aksys.
But the have an inherent curve to the filter. Thats how filters work in reality. So what your asking for is already in there.
By boy_blunder Tue Mar 13, 2012 11:02 pm
Cheers for your reply, those are good reads. You're right about the filters, they're both two-pole. Tested them with a spectrum analyser and white noise.

What do you mean about artists impression? The frequency responses are plotted from mathematical formulas, so they're theoretically how a filter looks. It would sound differently to us a we perceive lower frequencies as less loud than the highs though.

The first link about analog filters: aren't the inherent curves you're talking about the irregularities due to the characteristic of the system component, diode/transistor? Which wouldn't be present in digital filters.

But I wasn't asking about curving the filter, it was curving the attack or decay of the sample using the amplitude ADSR envelope. It would just mean that when multiplying the amplitude of the sample you do it with a square.

Then again these limitations are why I like using my mpc :)
User avatar
By bliprock Wed Mar 14, 2012 3:36 am
What I am saying is that the filter is curved, the MPC screen does not show true filter shape. My bad you are talking about ADSR and I think it would be the same. Its a representation, not the true curve that happens inside. Its like an artists impression. See links for 2 pole filter response in links above to see how a 2 pole filter attenuates, but of course its attenuation is not like this in MPC window. I beleive the AMp env wold be similar.
User avatar
By bliprock Fri Mar 16, 2012 4:17 am
yep that is good. Its a complicated for majority of users i think, and I have yet to see anything about it here before, but I have not read everything or searched for this here, so am not 100%. Conformation of these things would be good, but we do know that it would have to be software in the case of the MPC, and then its all conjecture. Filters not linear but ADSR might be, all depends on what amplitude transfer function and how it is implemented in the OS.