P10/OB-X8 Desktop Tuning Stability

[g3o2]

I still wonder how it comes that the OB-X8 seems considerably more stable in terms of tuning than the P10 - comparing desktop versions. Has anyone here observed the same thing?

[seqpro3]

I still wonder how it comes that the OB-X8 seems considerably more stable in terms of tuning than the P10 - comparing desktop versions. Has anyone here observed the same thing?

I haven’t heard anything or read anything in regards to that . I don’t have a desktop of either. I have the OBX8 synth keyboard. Its tuning is stable . I haven’t heard anything 
about the Prophet 10s synth keyboard tuning being unstable.  Wonder if you need to let the Prophet 10 warm up. You’re supposed to let it warm up and then tune it.  Does your P10 drift after it’s been on for awhile and you’ve been playing it?

[seqpro3]

I still wonder how it comes that the OB-X8 seems considerably more stable in terms of tuning than the P10 - comparing desktop versions. Has anyone here observed the same thing?

Did you read the post on how many times do you tune your Prophet 10?

[Analog Prophet]

I still wonder how it comes that the OB-X8 seems considerably more stable in terms of tuning than the P10 - comparing desktop versions. Has anyone here observed the same thing?

I haven’t heard anything or read anything in regards to that . I don’t have a desktop of either. I have the OBX8 synth keyboard. Its tuning is stable . I haven’t heard anything 
about the Prophet 10s synth keyboard tuning being unstable.  Wonder if you need to let the Prophet 10 warm up. You’re supposed to let it warm up and then tune it.  Does your P10 drift after it’s been on for awhile and you’ve been playing it?

As it is analog technology it has to warm up. I have not paid attention to how frequently I had tuned (the auto tune process) the P5 rev 4 and the OB-X8 more than it required more frequently tuning when they were new. That is natural as they build up a table of tuning it recognize related to temperature in the specific room, if I understand it right. When temperature is changing analog synths require tuning.

[g3o2]

Thank you for your feedback. Please note that I always let the P10 or the OB-X8 warm up for 20 minutes.

Yet, I have come to notice that I don’t need to tune the newer OB-X8 as much as the one and half years older P10, both in the same room. That has surprised me. After a post warm-up tune _each_ session, the P10 does remain reliable most of the time during the session.

Of course, on the OB, tuning is a separate and shorter process from full calibration, while, on the P10, AFAIU, tuning&calibration is a combined longer process.

It’s nothing that I find too annoying but it does make the OB-X8 appear more reliable.

[Pl@ton]

My P10 use to be unstable, I had to tune it 1 or 2 times a day, but since 6 months it's ALWAYS in tune (after around 20 min of warm-up of course).
Maybe you could try to delete the temperature/tuning tables and start over?

[g3o2]

Yup, I’ve done this last week and I’ll see how it will go. Can anyone please confirm that there is no visual feedback upon pressing Record+Tune?

I’ve always suspected my P10 to be a returned product because some knobs were already broken in while others were not - unless those were part of Sequential’s factory test programme. So maybe the tuning table was also durably influenced by that short time of ownership in someone else’s environment, even though after a year and a half this does seem unlikely.

[LPF83]

I still wonder how it comes that the OB-X8 seems considerably more stable in terms of tuning than the P10 - comparing desktop versions. Has anyone here observed the same thing?

It could be related to a smaller desktop case with 10 voices for the Prophet versus slightly larger case and only 8 voices for the OBX-8.  Heat accumulation inside the case could affect tuning stability.

I have the keyboard version of the P10 versus the desktop version of the OBX8, so that's not a fair comparison.  But I will say my Prophet 12 desktop (an example of high voice count in a small form factor) gets very warm.  Tuning issues aren't present because the oscillators are digital, but it demonstrates there is often a relationship between voice count, case size, and heat dissipation.

[g3o2]

It is true that the P10 desktop dissipates more warmth to the outside than the OB-X8 desktop. I’d suspect the 2 two extra voices to be the cause of that, rather than the size difference between both desktop versions, which is not that impressive.

Once warmed-up, the P10 peaks at an apparently steady 59-62 degrees temperature. From what I have read, this seems normal for the desktop version.

I guess I’ll have to see how it all goes after the tuning table reset on the P10.

[LPF83]

It is true that the P10 desktop dissipates more warmth to the outside than the OB-X8 desktop. I’d suspect the 2 two extra voices to be the cause of that, rather than the size difference between both desktop versions, which is not that impressive.

Even if the size difference seems insignificant by looking at them (and I'm going by Sweetwater's listed specs here since I only have physical access to the OB-X8),  the OB-X8 is listed at 4 x 22 x 8.5 inches while the P10 is listed at 3 x 20.75 x 7.4, which comes out to 748 cubic inches for the OB-X8 and 460.65 cubic inches for the P10.  That's about a 60% case volume difference, with 20% fewer VCOs for the OB-X8.

I have no way of knowing that's the source of the tuning requirement difference, but it seems plausible.  Even if the physical volume itself didn't affect the measurable heat transfer much, a smaller case can force designers to place components closer together, which can create smaller individual hotspots, that could potentially affect analog components depending on where in the system they appear.  On top of all that, the presence of hotspots could make it harder for a temperature sensor to accurately keep up.

It's one theory anyway... I won't pretend to sell it as the correct one

[g3o2]

It is indeed a 5 liter difference. Of course, this theory relies on the assumption that other users observe the same thing, which has so far not yet been the case in this thread The only thing I hope of course is that there is nothing wrong with my P10

[LPF83]

It is indeed a 5 liter difference. Of course, this theory relies on the assumption that other users observe the same thing, which has so far not yet been the case in this thread The only thing I hope of course is that there is nothing wrong with my P10

How often do you find you need to tune it?  Even with my keyboard version of the P10, I find I need to tune it slightly more often than most other synths.  In that regard, it could just be a side effect of the board design itself.  The response from Pl@ton seems consistent with the hotspot theory, as it would mean that the calibration mechanism needs more learning time than it might on another synth.

I haven't seen the inside of an OB-X8 to compare, but I know with the P10, the second set of 5 voices rests on top of the main board (which means it would be subject to heat coming off the components below it).  Maybe that is the differentiating factor, if the OBX-8 places all voices on a single board without a heat source underneath it?

A mystery I find harder to explain is why the tuning duration is so different between the two synths.  My OBX-8 takes about 12 seconds to tune all voices, while the Prophet 10 takes about 90 seconds.

[g3o2]

I tune my P10 once after warm-up usually. The OB-X8 sometimes does not even need tuning.

To me it seems like P10’s tune button performs both tuning&calibration, while the OB’s equivalent would be the full calibration in the globals and not just what the tune button does. The duration of those two processes is much closer.

[LPF83]

I tune my P10 once after warm-up usually. The OB-X8 sometimes does not even need tuning.

To me it seems like P10’s tune button performs both tuning&calibration, while the OB’s equivalent would be the full calibration in the globals and not just what the tune button does. The duration of those two processes is much closer.

That would make sense, as I haven't compared the timing of full calibration.