Is the Hive normal filter the same as the reporo5 rounded filter

[poopinpants]

I was playing to make the same sound from repro5 into Hive 2 and found that the normal filter in Hive with resonance sound the same as repro. Is it true? Is other u-he synths also a source for any Hive filters?

[EvilDragon]

It's not the same filter.

[poopinpants]

Oh ok do you know if it is same as a filter in Zebra or DIVA?

[EvilDragon]

Nope it's entirely Hive's own.

[Urs]

Hive's "normal" filters are much simpler in topology than any of Repro's. The general sound may be similar though, they're both based on cascaded one pole filters, but Repro's filter distortion promotes even harmonics while Hive's generally promotes odd ones.

[lunardigs]

@Urs, just curious; how do you know what harmonics are produced by a given design? Do you measure the output for specific things? Is it a simple evaluation, like listening or looking at a spectrum analyzer? Rather, does the filter's design indicate what you should get?

[Urs]

@Urs, just curious; how do you know what harmonics are produced by a given design? Do you measure the output for specific things? Is it a simple evaluation, like listening or looking at a spectrum analyzer? Rather, does the filter's design indicate what you should get?

It's not voodoo, just very plain maths. Other than that we use whatever is convenient at any moment in the design process, from estimating the transfer function in a circuit simulator own reading papers to comparing by ears to looking at spectrum analysers.

Hive's filters have perfectly symmetric distortion and IIRC they're DC coupled. Hence, they should promote odd overtones. Even overtones are promoted with asymmetric distortion and/or DC offsets.

Repro's filters are based on the Curtis 3320 chip which works like no other filter I know of and has an inherently asymmetric distortion.

[lunardigs]

... Repro's filters are based on the Curtis 3320 chip which works like no other filter I know of and has an inherently asymmetric distortion.

Do you think a filter like this will make it into Runciter or Zebrify3?

[recursive one]

What is the slope of Hive's bandpass and highpass filters?

Or does it depend on the engine?

[EvilDragon]

I think they're 2-pole.

[recursive one]

I think they're 2-pole.

I wonder what was the reason not to include the 4-pole ones then. Is it supposed that two HP or BP should be used in series insntead?

[Urs]

I think it depends on topology to a bit.

Only "Clean" is a true multimode filter (State Variable) which generates LP, HP, BP and the other outputs. HP would be 12dB, BP 6dB. "Normal" is a cascade/ladder design which requires mixing of outputs to approximate different filter responses. I guess I did 12dB for HP and BP. "Dirty" is a Steiner-Parker aka Sallen-Key design which requires "injecting" multiple inputs into the signal flow of the filter. I presume HP is 6dB and so is BP, but I can't be sure after all these years, and I'm not sure I have the time to dissect the code right now...

[recursive one]

I think it depends on topology to a bit.

Only "Clean" is a true multimode filter (State Variable) which generates LP, HP, BP and the other outputs. HP would be 12dB, BP 6dB. "Normal" is a cascade/ladder design which requires mixing of outputs to approximate different filter responses. I guess I did 12dB for HP and BP. "Dirty" is a Steiner-Parker aka Sallen-Key design which requires "injecting" multiple inputs into the signal flow of the filter. I presume HP is 6dB and so is BP, but I can't be sure after all these years, and I'm not sure I have the time to dissect the code right now...

Thanks for the explanation!

BP filters in Hive do sound great, but for certain cases I wouldn't mind more agressive BP filtering, I use 24dB bandpasses in other synths quite a bit.

Maybe that would be something for a very remote update?

[Urs]

Well, you know, when people say "24dB BP", it's commonly just 12dB. Because it's a 12dB slope to both sides, which is what you get from a 4-pole filter.

[recursive one]

Ah, just different terminology. Thanks, I see.

So basically if I want to recreate a patch from, say, Sylenth I use Clean or Normal mode where the original patch has 24 dB BP and Dirty for 12 dB in the original patch (I understand I probably won't be able to match the exact sound but that's not really the point, I sometimes use Sylenth patches as starting points for making my own sounds in Hive)

I like BP filters in Hive as they are, work great for my psytrance leads and other such things.