P600 Pulse Width Survey #98
-
|
In order to correct for typical PW deviations it is important to ge a better understanding of the variations of instruments. Everyone ready to participate and capable of using/trying OS versions alpha 13.* (* marking potential hotfix versions) and handling SysEx patches can contribute by sending in 3 patch syx files prepared as described below. |
Beta Was this translation helpful? Give feedback.
Replies: 9 comments 24 replies
-
|
pwm.zip |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, I have your values here (again different):
But: I found something odd: your 3rd SysEx (patch 22) for the "End" was corrupted. It contained a different number where the "Storage Magic" would be expected. I have no idea how this can happen. Of course, this type of situation would explain failed loads of patches. Does it load properly when selected? Could you please store it under a different number and export it again for me? |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
|
Of note I set my midpoint, start and end, using an oscilloscope. If I press 22 it loads fine. I did not try and send it back to the P600. |
Beta Was this translation helpful? Give feedback.
-
|
That came out with the correct Storage Magic. Mysterious... |
Beta Was this translation helpful? Give feedback.
-
|
Couldn´t get the python script to work, so here are my alpha13 PW patches: |
Beta Was this translation helpful? Give feedback.
-
BTW: when I made the python script work for Windows I found out the command is "py syx_converter.py ". So just "py", not "python", maybe that's what didn't work in your case. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, the "py" now gets the syx_converter to act, but it spits out another error message: |
Beta Was this translation helpful? Give feedback.
-
|
BTW, the script works fine on my Mac. the command line for me is python3 syx_converter.py patch22.syx Note the patch name has to be ".syx," not ".py" |
Beta Was this translation helpful? Give feedback.
-
|
Yes, the ending should be syx but the "*.py" ending should also work (the file can have any ending). el-folie's error message is daunting as it points to elementary encoding setting, possibly related to the operating system. I'll look into it when I have time - it would be nice if the script worked for everyone. |
Beta Was this translation helpful? Give feedback.
-
|
My P600 (remeasured):
Unfortunately, no one else contributed so far. For a survey that is a little thin. |
Beta Was this translation helpful? Give feedback.
-
|
Maybe it´s just that a perfect knob mid point/50% duty cycle is not that important for most users. But of course, for pure patch compatibility matters it´d still be nice if more users would send in their data. |
Beta Was this translation helpful? Give feedback.
-
|
Did you notice your start, mid, and end drift at all as the P600 warmed up? Edit: VCOA Mid 30912 VCOA End 62400 |
Beta Was this translation helpful? Give feedback.
-
|
My preliminary conclusion would be
The PW range is limited (first impression: <=95%). It always starts pretty much at zero. The We could scale the PW with carefully chosen values with the aim to cover the PW range better with the pots. it would b better to have more data, though. What it does not explain is why @el-folie reports that the LFO range is not sufficient to go from zero to zero. The limited voltage range of PW would make that even easier? BTW: in 13.2 the highest LFO amount has slightly increased due to the compatibility fix. |
Beta Was this translation helpful? Give feedback.
-
|
Well, 0.1 % deviation should still be ok. ;-))) |
Beta Was this translation helpful? Give feedback.
-
What are the three columns in the table? Variations in spec or different chip variants? Ah, I see - looked it up myself. Sorry to bother you. The ranges seem too large for an instrument. I guess there must have been cherry picking to use the ones with lower 100% voltage. Our 3 instrument are below 5V, obviously. |
Beta Was this translation helpful? Give feedback.
-
|
min typical max. Also, I was looking at the "buffer" after the DAC and before the PW input and it is interesting. Maybe I was wrong about the available voltage range for PW.
|
Beta Was this translation helpful? Give feedback.
-
|
From the Service Manual: |
Beta Was this translation helpful? Give feedback.
-
|
Nice! So the circuit makes sure full PW cutoff even at lower DAC gain settings than 5v. As those are 1% resistors, the PW difference between P600s could more likely be CEM3340 batch variation, like Matrix12x already assumed. |
Beta Was this translation helpful? Give feedback.
-
|
In principle you could replace 410 and 413 with pots to calibrate the PW voltage characteristic. But a things stand I feel that we don't have enough data to make a one-size-fits-all adjustment on the software side, especially when you look at the range in spec. I think we need to decide if the case is strong enough for a calibration procedure (like the one for bender). I find the explanation for the noise filter interesting. The LFO in the Z80 firmware must be really steppy!? |
Beta Was this translation helpful? Give feedback.
-
|
You see how low the cutoff frequency is for C443 (0.002uF) and R415 (4.7Mohm)? lol |
Beta Was this translation helpful? Give feedback.
-
|
9.4ms/16.9Hz. Funny, it never appeared to sound bad on Z80. But then again I never listened to the P600 as concentrated as I do since imogen´s great updates. ;-) |
Beta Was this translation helpful? Give feedback.
-
The original idea was just to have the knob´s mid point match the 50% PW - but in the end it´s not that important as the sound engine works as it is. Both zero point cutoffs are reachable. If someone´s bothered too much by the notch to the left on the knob he/I can fiddle with the electronics a bit. If a PW cal procedure means again much work for you in coding I don´t know if it´s worth it. It´d be an example of striving for perfection, yes, but not really necessary in this case. The slow detune on fine-tune on the other hand was extremely worth it, can´t get over how good it sounds now, can´t thank you enough. :) |
Beta Was this translation helpful? Give feedback.
-
|
I don't think we need a PW cal procedure. |
Beta Was this translation helpful? Give feedback.
-
|
The recent level of bug reporting (and fixing) shows that it is important at some point to converge to a "feature complete" version in which you only fix things but not introduce new issues. Therefore, yes, I feel maybe it would be better to leave it as it is just to make sure we come to an end. On the other hand, I think it more about patch portability than UI perfection because a PW value obviously leads to different wave shapes on different machines, especially if you create patches which oscillate close to or traverse the far zero point. |
Beta Was this translation helpful? Give feedback.
-
|
That´s true. In that regard it´s a pity that other users didn´t send in their PW syx data, that would have led at least to a better 50% point/shift for the knob. |
Beta Was this translation helpful? Give feedback.
-
|
OK, I created an enhancement issue for the calibration procedure but marked it "backlog". So it is documented but can wait for another time. Great excursion here - I am learning so much. I'll close this discussion by marking it "answered". BTW: I would have marked your posts as "answer" but they were replies, so that was not possible. |
Beta Was this translation helpful? Give feedback.
OK, I created an enhancement issue for the calibration procedure but marked it "backlog". So it is documented but can wait for another time.
Great excursion here - I am learning so much. I'll close this discussion by marking it "answered".
BTW: I would have marked your posts as "answer" but they were replies, so that was not possible.