Oscillator Issues

[Gerry Havinga]

thanks guys and thanks about my mum.

Yes mine is def faulty, but the mainboard that tech got into send to me didn't sound 100% on triangle anyway.. tbh without having it in front of me and testing it myself it seems hard to get exact definition of what 'a bit off' or 'as expected' means, my problem is if I have the new board I have to send the old one back (at my own expense) from UK to the US, but if I try to swap with my retailer I have to wait around a possible 2 more days (collection and then new one) when am quite busy with music and don't like sitting in silence for couriers. And even then there's no guarantee that swap wouldn't have the issue or other issues so I guess all I can do is go for the board and eat the costs of returning the other one (even though none of this is my fault). As said, I felt stuck between a rock and a hard place with no 100% guaranteed fixed board to be sent to me, tech did describe it in detail to his credit but, I dunno, DSI themselves should know if they are sending out a perfect or 'as expected' board to me rather than me have to do mind gymnastics to decide on the best course of action or if I'll end up with a still futzed board just not quite as bad...

was very busy last week so didn't have time to even think about it til today really. And yes, one of my triangle voices (same exact voice as the pulse has issues with) tends to sound more like a saw when set up as per the test.

I was writing on here (more than support @ DSI) still trying to work out what the deal is with everyone's rev2 and what is acceptable as 'normal' but does seem everyone has the pulse issue to some degree. I hope any replacement board doesn't increase the damage from the one voice to two voices + cost me money too. If the synth wasn't so damn good I'd have just returned it by now, but I can't, it's a cracker!


*and most definitely sounds and reacts like a full bodied, rich analog synth, with the best/liveliest DCOs I've ever personally heard (using the new slop), can get very close to a VCO richness, though not rawness, which is fine as I don't always want that or need that in most cases.

@Propheteer I am also sorry to hear about your mother, my condolences.

I own an 8 voice Rev 2 and going through the test(s) described earlier in this threat have not been able to detect, by ear, any of the issues in relation to the Saw and Saw-Tri. I did notice at the very far end of the ranges for the Pulse setting my Rev 2 starts to show similar behavior as described earlier. But I also believe this (at the far end both side of the scale) is normal behavior.

[Propheteer]

thanks guys and thanks about my mum.

Yes mine is def faulty, but the mainboard that tech got into send to me didn't sound 100% on triangle anyway.. tbh without having it in front of me and testing it myself it seems hard to get exact definition of what 'a bit off' or 'as expected' means, my problem is if I have the new board I have to send the old one back (at my own expense) from UK to the US, but if I try to swap with my retailer I have to wait around a possible 2 more days (collection and then new one) when am quite busy with music and don't like sitting in silence for couriers. And even then there's no guarantee that swap wouldn't have the issue or other issues so I guess all I can do is go for the board and eat the costs of returning the other one (even though none of this is my fault). As said, I felt stuck between a rock and a hard place with no 100% guaranteed fixed board to be sent to me, tech did describe it in detail to his credit but, I dunno, DSI themselves should know if they are sending out a perfect or 'as expected' board to me rather than me have to do mind gymnastics to decide on the best course of action or if I'll end up with a still futzed board just not quite as bad...

was very busy last week so didn't have time to even think about it til today really. And yes, one of my triangle voices (same exact voice as the pulse has issues with) tends to sound more like a saw when set up as per the test.

I was writing on here (more than support @ DSI) still trying to work out what the deal is with everyone's rev2 and what is acceptable as 'normal' but does seem everyone has the pulse issue to some degree. I hope any replacement board doesn't increase the damage from the one voice to two voices + cost me money too. If the synth wasn't so damn good I'd have just returned it by now, but I can't, it's a cracker!


*and most definitely sounds and reacts like a full bodied, rich analog synth, with the best/liveliest DCOs I've ever personally heard (using the new slop), can get very close to a VCO richness, though not rawness, which is fine as I don't always want that or need that in most cases.

@Propheteer I am also sorry to hear about your mother, my condolences.

I own an 8 voice Rev 2 and going through the test(s) described earlier in this threat have not been able to detect, by ear, any of the issues in relation to the Saw and Saw-Tri. I did notice at the very far end of the ranges for the Pulse setting my Rev 2 starts to show similar behavior as described earlier. But I also believe this (at the far end both side of the scale) is normal behavior.

Thanks.

[cloudswimmer]

Ok two days in now with mine and so far so good. The synth sounds a lot like my Oberheim DCO synths on steroids. Will report back if anything pops up. And I too offer my condolences Propheteer.

[Propheteer]

Cheers Cloudwimmer.

Peace.

[Propheteer]

Was told, hopefully good news, that this is indeed fixable in software now. In fact I didn't even get a new board in the end because of this (thankfully). Not sure on an ETA for the new firmware or if they can really fix all issues but I had a feeling it was just some offsets wrong for shape mod on certain voices causing the anomolies, hope it was something that simple.

Anyone else heard from DSI support about new firmware correcting this and if so have you been sent it yet?

[sangel]

Was told, hopefully good news, that this is indeed fixable in software now. In fact I didn't even get a new board in the end because of this (thankfully). Not sure on an ETA for the new firmware or if they can really fix all issues but I had a feeling it was just some offsets wrong for shape mod on certain voices causing the anomolies, hope it was something that simple.

Anyone else heard from DSI support about new firmware correcting this and if so have you been sent it yet?

I sent my prophet 08 back to Thomann and they checked it.. tbey just did a reset and a filter calibration and the problem went away.. I read however from another person in this thread that it’s possible that the problem comes back...

[Propheteer]

Yeah that's just the in-menu calibration though? that we have all tried? A new firmware should (hopefully) fix it at a fundamental level, at least that's what support told me recently after originally going to send me a new board.

I still have hope... for now!

[kris]

Hi all,

with drops of sweat on my forehead I worked through this thread tonight after I discovered my Rev 2 16-voice had the described triangle discontinuity (triangle with 0% shapemod still sound a bit like a saw on some of the voices) problem on 4 of the 16 voices, on both Osc1 and Osc2.

I contacted support, and they confirmed (within 14 minutes of me sending the email - praise where praise must be) that there was an issue in the calibration routine within the firmwares prior to 1.1.0.

I installed 1.1.0, and the problem was gone immediately even without running the Osc calibration as they had suggested.

Happy, I love my Rev2!

And DSI!

[dsetto]

Good news, kris. Thank you for reporting!

[Propheteer]

<snip>Same on triangle, the 'bit of saw' seems related to shape mod. the same voice/note that acts up in pulse is also the one acting up on triangle (when shape mod is at minimum). So in triangle/basic program init (shape mode zero) any out of calibration voices will have a bit of 'saw', and with pulse that same voice when shape mod is at max will have a collapsed wave (no sound). Not sure if this will help anyone or DSI to pinpoint if something can be done in software but it seems it's a calibration issue outside of the power of the built in calibration menu (so a FW fix may update the calibration routine to correct for this shape mod issue on random voices?)

Some did say I was probably wrong when I hypothesised that it should be fixable in software, back when I posted this, am glad I wasn't! 

That said, I've not even updated to 1.1 yet (didn't even realise it was released as they said they would send me it so assumed it wasn't as they hadn't) so thanks to the poster above for tipping us off. Hopefully it fixes mine too...

[Propheteer]

well... thought it was too good to be true

I've updated to 1.1, I tried first without re-calibrating (still had the issue) then re-calibrated (still had the issue).

The issue has changed it seems, though it's been a while since I last tested.

For me the same ONE VOICE remains different. Now the triangle going more 'saw like' doesn't happen on repeated key press (it seems to never get to that voice as if its not cycling through the whole set of voices anymore but just using the first one), but play a chord with that 'basic program' and lo and behold there's that one brighter voice AGAIN.

With shape mod adjust to full, that one voice now is far DULLER than the rest (as if it has gone around the range and looped back on itself to duller rather than brighter).

That same voice when I hold it and switch to pulse is indeed still a problematic voice, it simply doesn't sound at all (when shape mod is at full)

So, for me it seems the new 1.1 FW has done absolutely nothing at all towards fixing this. Am gutted as had a lot of hope this would finally put that issue to bed so I could stop worrying about it in the long term.

[DavidDever]

well... thought it was too good to be true

I've updated to 1.1, I tried first without re-calibrating (still had the issue) then re-calibrated (still had the issue).

The issue has changed it seems, though it's been a while since I last tested.

For me the same ONE VOICE remains different. Now the triangle going more 'saw like' doesn't happen on repeated key press (it seems to never get to that voice as if its not cycling through the whole set of voices anymore but just using the first one), but play a chord with that 'basic program' and lo and behold there's that one brighter voice AGAIN.

With shape mod adjust to full, that one voice now is far DULLER than the rest (as if it has gone around the range and looped back on itself to duller rather than brighter).

That same voice when I hold it and switch to pulse is indeed still a problematic voice, it simply doesn't sound at all (when shape mod is at full)

So, for me it seems the new 1.1 FW has done absolutely nothing at all towards fixing this. Am gutted as had a lot of hope this would finally put that issue to bed so I could stop worrying about it in the long term.

Install the new board, then–if it's busted, then this should fix it.

If the new board is faulty, then send the unit back / sell it / walk away.

Chances are that you're hearing this because there is something amiss in the first place (could have been a faulty / wrong-valued SMT component, for all you know).

The root cause, on the other hand, is not your problem, but the manufacturer's.

[Propheteer]

install WHAT board?

 I don't have a board! I would rather have avoided having to fit a board, as well as having to return my board (from the UK to the US at my own expense!).

There was no board sent, I was waiting to decide on if it was worth it when I was told by DSI that they had fixed it in software so no need to send me a board now.

needless to say, had I got another 'new' mainboard sitting here, or had been sitting here all these weeks I'd have already fitted it and tested it, I'm not lazy (just busy).

Of course it's their problem, I still hope it IS fixable in software mainly because I can't be arsed to waste time dismantling a brand new synth, and posting another board back to the US (esp as I was told the 'new board' they had there also showed some issues on the pulse but not the triangle). These kinds of things are usually not expected when you buy 'brand new' and when you pay over a grand for something.

[Propheteer]

ok....

at last. Good news. Seems that although I updated, synth rebooted itself, and I ran the calibration (once). It didn't appear to fix it at all. I was going to do a video today to send to DSI and thought maybe I'll run calibration one more time and bingo!

It's now working ok It looks like it must have needed switching off after calibration and on again or needed it run twice for some reason (probably the former but I just tested after calibration yesterday without rebooting - duh).

so the triangle and the pulse issue seems to have gone, and now the pulse (all voices) on full shape mode go virtually uniformly silent which is a good thing. I like PWM that doesn't hold back and can collapse down like that.

I'll see if it holds up over the next few days but can't see why not now. Has anyone else in this thread updated to 1.1? Hopefully it fixes everyone's problem and we can all get back to raving about the virtues of this great synth instead of this gremlin.

Thanks DSI.

[Sleep of Reason]

I don't get why people keep saying 1.1 did anything for them...

No additional changes have been made other than incrementing the version number.

[Propheteer]

I don't get why people keep saying 1.1 did anything for them...

No additional changes have been made other than incrementing the version number.

From my email with DSI last night

OS 1.1 was made to address the issue as you've noted others that have had similar issues have had it fixed with this. Shoot me a link to the video once you have it up.

Best regards,

Mark
Dave Smith Instruments

Maybe because it has? I wouldn't say it had if it hadn't. And do note I even asked him in email if 1.1 was actually made to 'do anything' and he confirmed it was. This morning I did as above and it worked. I can't argue with fact regardless of why or what they told you on here.

MAYBE they just didn't want to shout about the issue as it was only affecting a small number of users (or at least only a small number of users were aware), but please don't insinuate I'm a moron who's affected by placebo, the damn thing worked after updating and a power cycle.

[Robot Heart]

Glad the OS solved your issue and you're back to working condition!

To clear things up, the fix in question was already implemented in 1.0.9.4, as mentioned in bug #9 of the release notes in the same thread where you quoted me. Here it is for reference.

What Mark meant by saying 1.1 addressed the issue is that "we've addressed the issue in our official OS release". Not "we've done something extra between OS 1.0.9.4 and OS 1.1 but we're not owning up to it". I can assure you, nothing has changed between 1.0.9.4 and 1.1 except for the version number. Period.

As this issue has been deemed resolved, we are locking this thread. Please contact our dedicated support channel if you need further clarification on anything discussed herein: support (at) davesmithinstruments.com