MINIMAX controller resolution

batfink · Post by **batfink** » Mon Jul 07, 2003 3:51 am

Hi
I've just purchased a large Pulsar setup and will shortly be ordering Minimax; however I have a question.
When, for example, altering the rotary controls in Miniscope II, they appear to have only 5-bit resolution (is this correct?) giving rise to audible stepping.
Has Minimax improved this spec and are all rotary parameters now 7-bit?
Owing a Macbeth M3X I am familiar with the great depth of tonal variation offered by closely tuning multiple oscillators, so clearly a greater bit-depth would significantly enhance SFP instruments such as Miniscope II (and presumably increase DSP load!).

Thanx

kimgr · Post by **kimgr** » Mon Jul 07, 2003 11:56 am

Most controls in SFP are 32bit afaik !!!
Except if you use remote midi, then you're limited to 7bit naturally...

Kim.

spacef · Post by **spacef** » Mon Jul 07, 2003 12:39 pm

the resolution of most parameters (pots, faders) under sfp is of approx 2,1 million values (if i remember correctly, eventhough it can be lower, and go to 4,2 millions sometimes).

The resolution you get with midi or in a sequencer is all about the resolution of Midi which only accept 128 values for midi Controls.

In SFP, with the mouse, you get access to higher resolution by using the big circle technique explained in the manual.
By using a hadware midi remote controller, you already have access to finer values.

It's a limitation of midi protocol i think, not SFP. NOAH has hardware direct access to those values (i mean the live controllers communicate directly with the synth parameters, not through midi but "through hardware").

I've heard of hardware remote controllers that can send higher values than 1/127 thru midi, but i wonder if it works or if any sequencer accepts this ? (i have a sequencer approach because i record a awful lot of CC tracks).

<font size=-1>[ This Message was edited by: spacef on 2003-07-07 14:04 ]</font>

visilia · Post by **visilia** » Mon Jul 07, 2003 3:02 pm

But to answer your question more directly: no, the MiniMax has not any 'zipping' issues, even when using only 128 MIDI steps. It probably interpolates between discrete steps.

Cheers,
Vincent

batfink · Post by **batfink** » Tue Jul 08, 2003 2:59 am

OK thanks guys. My question arose from counting the number of (clearly) audible steps when turning the pitch and filter pots throughout their range - both using a mouse and the 7-bit slider on my Roland PC160A controller.
I counted 32 steps for each parameter which corresponds to 5, rather than the usual 7, bits. This resolution does seem unexpectedly low and I'm wondering if it may be a bug or something? I'm running SFP v3.0 but will upgrade to 3.1c when my new board arrives.

Thanks

Mr Arkadin · Post by **Mr Arkadin** » Tue Jul 08, 2003 3:52 am

There was no SFP 3.0, just Pulsar 3.0, SFP is 3.1a, b or c. You really need 3.1c to run MiniMax, it's not recommended to run in 3.0.

aMo · Post by **aMo** » Tue Jul 08, 2003 5:54 am

We have a workaround for this problem in Modular.
We simply smooth it out.
We do this with a constant value slider (or similar) and a filter, wich we connect to the cutoff-mod, the filter smooths out the zipping.
So it is very lightly that Creamware finally added this to one of their synths.

So even if it is 7 bit or 5 bit or whatever, the problem has been solved, and shouldn't have to persist in future synths from Creamware (although they might let some knobs be unsmooth because of DSP usage of having a filter on every knob).