Parva Development Update 2016/07/20


Last week I released Parva Mainboard Firmware v0.50, which included a wide range of bug fixes, improvements, and added features. Most significantly, I was able to resolve the relative tuning issue through the mainboard firmware; however, some of the other fixes and features require new voicecard firmware.

Unlike the mainboard, the new voicecard firmware didn’t require a full rewrite, but there are some significant changes in the DCO code to eliminate popping/crackling noises and the aliasing that can be heard when playing high notes. This new code is fully tested and almost ready for release. The one remaining issue is the ramp control calibration function.

Some of you have noticed the inconsistent oscillator levels, PWM widths, and glitchy triangle waves – all of these are caused by inconsistent ramp core trimming. Fortunately, each voice card contains circuitry to route signals back to the master microcontroller, which can then automatically re-calibrate the ramp cores. I’m fine-tuning the calibration code right now, and hope to have it working tonight, but it is a fairly complex and tedious process.

NEW - the weekly report about the development of the Parva

Any news about the update ?


The new voicecard firmware is complete – including the self-calibration function – and is going through regression testing now.


Anything new to report?


I expect this relates to the info that Brad already mentioned, but while I was prepping for a video comparing basic, unfiltered saw and square waves on various analog synths, I thought I would throw the Parva into the test, and this is what I got:

  1. at low osc level settings (around 30), the saw wave mostly looks like a saw wave, with it slowly growing spikier at lower frequencies. In fact, at the lowest usable octave, a large number of the harmonics are gone, and the overall volume is reduced (it’s VERY spiky)
  2. at the highest osc level settings (127), all sorts of harmonic artifacts are present, making the wave look more like a standard analog square wave than a saw wave. I expect this likely accounts for the “metallic” sound that a lot of folks have mentioned. I thought someone mentioned this was overdriving causing that, but whatever the reason, it starts affecting the basic saw wave at fairly low levels (someone else mentioned that as well), and the misshaping of the wave is more prominent at lower frequencies.

It is was it is, and I’m hoping the new firmware for the voice cards will take care of a lot of that, but my real question is this (and apologize if it was answered someplace and I missed it):

How is it the Parva demoed in the “Futuresonus Parva Preview” video (Published on Feb 16, 2015) seems to NOT suffer these same issues? Based on that alone, I would have assumed large parts of the anomalies we’re seeing would have already been worked out.

Again, fingers are crossed for that voice card firmware update.



That appears to be phase inverted. If you flip the picture over, you’ll see that it’s a pretty standard ramp wave. The v50 firmware update will even out the amplitude over the whole frequency range.

With the Clean OSC option on, a single saw oscillator will not overdrive the mixer or filter at any level setting.

The oscillator VCA gain range was increased before production. Other than that, the voice hardware is exactly the same.


Just back from vacation an seeing this. Great info! Any hint, by when the regression testing will be finished?


After a few more minor changes, it looks like it’ll finally be done this evening.


That would be cool, some new stuff to play with :slight_smile:


are we any closer to this evening?


… Christmas evening?


As I said last week, there is just one more known bug to fix, and I’m working on it.


I solved it! I’ve been running up against a limitation of the voice card microcontroller, but I just figured out a way to make it do what I want. A quick test looks good, and it’s going through “official” testing now.


nice to hear! so I will warm up my Parva, again :slight_smile:


The testing has gone well, and I will be uploading the new firmware today.


And there was great rejoicing throughout the land :slight_smile:


I installed the 0.51 software version yesterday and I’m very pleased with the results - the Parva sounds way better.

Are there plans for a good set of presets? There seem to be very few around, including in these forums.