The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Synths for sale     Latest  Trending
Dave Smith and Tom Oberheim OB6
Old 5th March 2019
  #5641
Lives for gear
 
flat's Avatar
Quote:
Originally Posted by magikroom View Post
So, the OB6 is my favourite Synth I've bought, however as I've made Patches for 10 different PolySynths, I get asked "Which Synth, this vs that vs another". I get this a lot and decided to do little Reviews and Score each Synth in 5 Areas, each out of 20 for a final score out of 100. Here's the OB6

Great little review. Will try and catch your Sunday show.
Old 6th March 2019
  #5642
Lives for gear
 

Quote:
Originally Posted by Channelizer View Post
I emailed Sequential and got the following response. I agree, it's not a bug, like I said earlier, it's just a limitation of the speed of midi ... NRPNs can cause unwanted effects because they split a number over two packets and both packets can't arrive at the same time. If anyone else wants to take it up with them, feel free. For my expander, I'm inclined to use CCs rather than NRPNs which will give more smoothness but slightly less resolution on a few parameters.

Hi Steve-
Thanks for contacting Sequential. That's an accurate observation you've made regarding what happens when you modify the filter frequency (or other controls with a range greater than 128 values) via NRPN and get to the point where you have to change from a value of 127 to 128. Since the LSB has to change from 127 to 0 and the MSB has to change from 0 to 1, and these two commands can't happen completely simultaneously, what you're reporting occurs. What the software team tells me is that it's a MIDI bandwidth limitation, and there's not really a way to handle it differently given how things are structured. You would encounter this on our other instruments, as well.
I hope this answers your questions.
Best,
Andy
Sequential LLC
while i can see how this occurs and accept it is what my ob6 does, this actually makes no sense to me... sound engineering design would have part of a message in a temporary buffer awaiting the other half and only execute when the entire message is in.
Old 6th March 2019
  #5643
Lives for gear
 

Quote:
Originally Posted by Mystic38 View Post
while i can see how this occurs and accept it is what my ob6 does, this actually makes no sense to me... sound engineering design would have part of a message in a temporary buffer awaiting the other half and only execute when the entire message is in.
Consider what the MIDI spec says (I quoted it a page or so back):
Quote:
If 128 steps of resolution is sufficient the second byte (LSB) of the data value can be omitted. If both the MSB and LSB are sent initially, a subsequent fine adjustment only requires the sending of the LSB. The MSB does not have to be retransmitted. If a subsequent major adjustment is necessary the MSB must be transmitted again. When an MSB is received, the receiver should set its concept of the LSB to zero.
To me, that says that the full range of the parameter should be covered by going 0 to 127 on the MSB and that if you need finer control then you use the LSB, so this situation wouldn't arise because changing the MSB should zero the LSB.

So suppose your parameter had an actual resolution of 255 steps. If you only used the MSB each increment would be 2 steps of the actual value; to get the individual steps between those you'd send a single LSB with a value of 64. So the series of individual messages to increment the actual parameter value from 126 to 130 would go:
MSB 63 = 126 (LSB resets to 0)
LSB 64 = 127
MSB 64 = 128 (LSB resets to 0)
LSB 64 = 129
MSB 65 = 130 (LSB resets to 0)
and there wouldn't be any jumps because each time you transmitted the MSB the LSB would be zeroed.

Maybe this is arrogant for me to say, given Dave Smith was one of the founders of MIDI, but I do think the way they've implemented the NRPN's here is out of line with the spec, and that if they'd stuck to the spec then there wouldn't be this problem.
Old 7th March 2019
  #5644
Lives for gear
 

oops i had missed that so thanks for the wording on the spec, so I agree, from that, it seems pretty clear that it is not implemented correctly...the spec adequately covers part transmission in both msb and lsb cases.

Quote:
Originally Posted by AlanC3 View Post
Consider what the MIDI spec says (I quoted it a page or so back):

To me, that says that the full range of the parameter should be covered by going 0 to 127 on the MSB and that if you need finer control then you use the LSB, so this situation wouldn't arise because changing the MSB should zero the LSB.

So suppose your parameter had an actual resolution of 255 steps. If you only used the MSB each increment would be 2 steps of the actual value; to get the individual steps between those you'd send a single LSB with a value of 64. So the series of individual messages to increment the actual parameter value from 126 to 130 would go:
MSB 63 = 126 (LSB resets to 0)
LSB 64 = 127
MSB 64 = 128 (LSB resets to 0)
LSB 64 = 129
MSB 65 = 130 (LSB resets to 0)
and there wouldn't be any jumps because each time you transmitted the MSB the LSB would be zeroed.

Maybe this is arrogant for me to say, given Dave Smith was one of the founders of MIDI, but I do think the way they've implemented the NRPN's here is out of line with the spec, and that if they'd stuck to the spec then there wouldn't be this problem.
Old 8th March 2019
  #5645
Pym
Lives for gear
 

... No?

This isn't how it is supposed to work. Although, to give you credit, technically it would fix the problem in this specific case, it is absolutely not the way it is written in the MIDI spec

The most simple reason I can give you is for this to work, you have to know what those bits in the MSB represent... and you can't know that until you know how many fine points are in between each MSB bit. You can't know that without knowing the RANGE of the parameter. Currently there is no way to indicate what the range is of an NRPN, this is why there is a +/- command for the NRPNs. Without knowing the range of the NRPN how do you know how many bits in the LSB are valid? Doesn't work.

Knowing the range would be immensely helpful to MIDI controllers that wanted to do NRPNs (or even CCs for that matter) but it isn't in the spec as it currently is.

I have a way to fix this but it involves a global flag that doesn't respond to incoming NRPN data until both bytes have been received when it gets an MSB value. The reason I did it the way it is some devices actually only send an MSB sometimes, as a bandwidth saver, but it isn't normal. Either way you cause a problem. The OB6/P6 don't have a global location left in the 0-9 buttons so I need to think about how to implement this if we choose to later


Quote:
Originally Posted by Mystic38 View Post
oops i had missed that so thanks for the wording on the spec, so I agree, from that, it seems pretty clear that it is not implemented correctly...the spec adequately covers part transmission in both msb and lsb cases.
Old 8th March 2019
  #5646
Lives for gear
Quote:
Originally Posted by Pym View Post
I have a way to fix this but it involves a global flag that doesn't respond to incoming NRPN data until both bytes have been received when it gets an MSB value. The reason I did it the way it is some devices actually only send an MSB sometimes, as a bandwidth saver, but it isn't normal.
Thank you for sharing your insight, Chris, it is very much appreciated.

May I suggest that the controller most likely to be used to control an OB-6 would be an OB-6. A user just wants to record some filter sweeps from the front panel via MIDI. IMHO It should respond to itself correctly, before all other considerations.

I can’t speak for everyone in the world, but I believe the global flag you describe should be the default behaviour: wait for both bytes before updating the sound.

If it is really something that the user must have control over, perhaps it could be a new option in the global “param rcv” (presently OFF, CC, nr) - something like “nr2” (“NRPN 2 byte”), to save a dedicated button?

Thanks again for visiting and participating!
Old 8th March 2019
  #5647
Pym
Lives for gear
 

Yeah I could probably just make it a receive mode, that would probably be easiest.

Honestly I may just set it to default, I can't imagine too many people having a difficulty with it the opposite way around, although it does happen or I wouldn't have made it this way in the first place. I'll think about it

Quote:
Originally Posted by infindebula View Post
Thank you for sharing your insight, Chris, it is very much appreciated.

May I suggest that the controller most likely to be used to control an OB-6 would be an OB-6. A user just wants to record some filter sweeps from the front panel via MIDI. IMHO It should respond to itself correctly, before all other considerations.

I can’t speak for everyone in the world, but I believe the global flag you describe should be the default behaviour: wait for both bytes before updating the sound.

If it is really something that the user must have control over, perhaps it could be a new option in the global “param rcv” (presently OFF, CC, nr) - something like “nr2” (“NRPN 2 byte”), to save a dedicated button?

Thanks again for visiting and participating!
Old 8th March 2019
  #5648
Lives for gear
 

Quote:
Originally Posted by Pym View Post
You can't know that without knowing the RANGE of the parameter. Currently there is no way to indicate what the range is of an NRPN, this is why there is a +/- command for the NRPNs. Without knowing the range of the NRPN how do you know how many bits in the LSB are valid?
Ah. I see what you're saying.

If you wanted to use the NRPNs to set values for a parameter with a range that didn't match some power of 2 then the LSB values would be spread out in an irregular fashion, and that would make it hard to get smooth control changes without sending a lot of redundant data unless you knew the exact range and hence exactly when - and with what value - to send the LSB.
Old 8th March 2019
  #5649
Pym
Lives for gear
 

Yup, exactly

Quote:
Originally Posted by AlanC3 View Post
Ah. I see what you're saying.

If you wanted to use the NRPNs to set values for a parameter with a range that didn't match some power of 2 then the LSB values would be spread out in an irregular fashion, and that would make it hard to get smooth control changes without sending a lot of redundant data unless you knew the exact range and hence exactly when - and with what value - to send the LSB.
Old 8th March 2019
  #5650
Lives for gear
 
John01W's Avatar
This is awesome.....Thanks a bunch Pym!
Old 8th March 2019
  #5651
Lives for gear
 
bobule's Avatar
 

Thanks Pym for the knowledge drop!
Old 8th March 2019
  #5652
Lives for gear
 
Synthpark's Avatar
Quote:
Originally Posted by xanderbeanz View Post
The VCO2 to VCO1 Xmod isn't consistent on mine, I'll get a nice tuning on one note and then other notes will sound out of tune or inharmonic, is this normal? A limitation of VCOs? I have better results modulating shape at audio rates than the VCO 1 pitch.
If you think that is normal, it is not.
I have the same problem and I realized what it is.

When I run LFO2s in low frequency mode,
2 out of 6 vco2 are tuned a quarter and a half tone higher.
No way to fix it, calibration does not help.

If it was only VCO1 as a destination, but also Filter FM will sound slightly different for one voice.

I am really andgry and fed up with this company. It is the second time that I ordered an OB6 and there is a quality issue.
First it was the sub oscillator, now it is xmod. That is a 2465 Euro synthesizer, not a 500 Euro synth. I am not going to contact support ... "would you be willing to open the unit?". NO I AM NOT. They will send some other crappy voice cards, which will probably have some other issue. YOU DSI SHOULD PAINSTAKINGLY TEST YOUR SYNTHS NOT ME.

If DSI could just deliver a fine working unit, but that seems almost impossible. That is modern "made in USA". Really, then I prefer made in China, even if the raw tone is not as great as here.

This extension you are talking about sounds like a joke, given that not even the standard functions are working 100%.
Old 8th March 2019
  #5653
Lives for gear
 

thanks for that ..

your proposed solution of ensuring the entire message is received was actually in my first post on this topic.. given it is a common scheme in many comm protocols, both open and closed in nature.

then the clarifications of the spec started...lol

Quote:
Originally Posted by Pym View Post
... No?

This isn't how it is supposed to work. Although, to give you credit, technically it would fix the problem in this specific case, it is absolutely not the way it is written in the MIDI spec

The most simple reason I can give you is for this to work, you have to know what those bits in the MSB represent... and you can't know that until you know how many fine points are in between each MSB bit. You can't know that without knowing the RANGE of the parameter. Currently there is no way to indicate what the range is of an NRPN, this is why there is a +/- command for the NRPNs. Without knowing the range of the NRPN how do you know how many bits in the LSB are valid? Doesn't work.

Knowing the range would be immensely helpful to MIDI controllers that wanted to do NRPNs (or even CCs for that matter) but it isn't in the spec as it currently is.

I have a way to fix this but it involves a global flag that doesn't respond to incoming NRPN data until both bytes have been received when it gets an MSB value. The reason I did it the way it is some devices actually only send an MSB sometimes, as a bandwidth saver, but it isn't normal. Either way you cause a problem. The OB6/P6 don't have a global location left in the 0-9 buttons so I need to think about how to implement this if we choose to later
Old 8th March 2019
  #5654
Gear Addict
 

Quote:
Originally Posted by Synthpark View Post
If you think that is normal, it is not.
Actually, it is normal. Th resulting signal from high rate frequency modulation of an oscillator can vary dramatically depending on the phase and frequency of the carrier and modulator respectively.

The P6 and OB-6 have free running VCO oscillators. Slight tuning differences between voices and oscillators can yield drastic differences in output spectra.

As such, XMOD will not yield perfectly consistent operation voice to voice as the phase and frequency of the oscillators is not perfect like digitally generated FM.
Old 8th March 2019
  #5655
Lives for gear
 
string6theory's Avatar
Quote:
Originally Posted by cbmd View Post
...

The P6 and OB-6 have free running VCO oscillators.

...
And, what lovely sounding free running VCO’s they are!

I happen to really like some of the unexpected sonic artifacts that come out of Poly and X-Mod. It gives these masterful synths even more personality and character.
Old 9th March 2019
  #5656
Lives for gear
 
Synthpark's Avatar
Quote:
Originally Posted by cbmd View Post
Actually, it is normal. Th resulting signal from high rate frequency modulation of an oscillator can vary dramatically depending on the phase and frequency of the carrier and modulator respectively.

The P6 and OB-6 have free running VCO oscillators. Slight tuning differences between voices and oscillators can yield drastic differences in output spectra.

As such, XMOD will not yield perfectly consistent operation voice to voice as the phase and frequency of the oscillators is not perfect like digitally generated FM.
You can read, right? Apart from the standard audio range setting
I put my second VCO also into low frequency mode.

The voices don't have the same pitch for the second VCO, despite the calibration. Nothing to do with the usual tuning. So even using the second VCO as filter FM will not sound consistent. Its like to have a bad second LFO.

I was also "normal" to assume by a lot of people (believers) there are insonsistencies for the sub oscillator from one voice to the next, until the new OB6 proves otherwise. A lot of stuff seems "normal" for DSI, until you realize it is just careless manufacturing and variations from one unit to the next. (I see similar things in the medizine company where I work, a lot of customer complaints about poorly assembled devices, a taste of modern "made in Germany" ).

Man I just want a new perfectly calibrated OB6. The sound is great. Just not the quality control.
Listen yourself. It is not about 5 percent tuning. cheers.

I added MKS80 at most extreme xmod setting and there are NO variations, a synth not calibrated over the last 15 years. (regardless of its musical sense).
Attached Files

VCO2 low frequency mode.mp3 (295.9 KB, 2226 views)

mks80.mp3 (866.3 KB, 2189 views)


Last edited by Synthpark; 9th March 2019 at 01:13 AM..
Old 9th March 2019
  #5657
Gear Addict
 

I was referring to your assertion that it is not normal for voice to voice inconsistencies when performing oscillator FM using Xmod.

Concerning the issue you've noted, I've just checked an OB-6 in our office and am hearing similar pitch variance between the voices with keyboard tracking off, low frequency on, and the oscillator set to its highest frequency value.

In this case, oscillator 2 is not meant to track pitch perfectly and we do not calibrate the oscillator in low frequency mode. Low frequency mode is primarily meant for modulation purposes and slight tuning differences in oscillator 2 between voices will not yield unsatisfactory results in use.

The OB-6 is an analog VCO based synthesizer, there is no perfectly calibrated unit and variation from unit to unit and voice to voice is part of it's charm and sonic signature. If you want perfect, get a digital synth.
Old 9th March 2019
  #5658
Lives for gear
 
Synthpark's Avatar
Quote:
Originally Posted by cbmd View Post
I was referring to your assertion that it is not normal for voice to voice inconsistencies when performing oscillator FM using Xmod.

Concerning the issue you've noted, I've just checked an OB-6 in our office and am hearing similar pitch variance between the voices with keyboard tracking off, low frequency on, and the oscillator set to its highest frequency value.

In this case, oscillator 2 is not meant to track pitch perfectly and we do not calibrate the oscillator in low frequency mode. Low frequency mode is primarily meant for modulation purposes and slight tuning differences in oscillator 2 between voices will not yield unsatisfactory results in use.

The OB-6 is an analog VCO based synthesizer, there is no perfectly calibrated unit and variation from unit to unit and voice to voice is part of it's charm and sonic signature. If you want perfect, get a digital synth.
Get me a digital synth? Why did I just prove that an analog synth can do xmod without too much variation? I know more when I go in the store tomorrow and compare with what they have. Their demo unit should say more.
Old 9th March 2019
  #5659
Lives for gear
 

Quote:
Originally Posted by Synthpark View Post
Get me a digital synth? Why did I just prove that an analog synth can do xmod without too much variation? I know more when I go in the store tomorrow and compare with what they have. Their demo unit should say more.
If you're just doing a single note you will probably have better luck using the one voice unison mode.
Old 9th March 2019
  #5660
Lives for gear
 
Synthpark's Avatar
Quote:
Originally Posted by cbmd View Post
I was referring to your assertion that it is not normal for voice to voice inconsistencies when performing oscillator FM using Xmod.

Concerning the issue you've noted, I've just checked an OB-6 in our office and am hearing similar pitch variance between the voices with keyboard tracking off, low frequency on, and the oscillator set to its highest frequency value.

In this case, oscillator 2 is not meant to track pitch perfectly and we do not calibrate the oscillator in low frequency mode. Low frequency mode is primarily meant for modulation purposes and slight tuning differences in oscillator 2 between voices will not yield unsatisfactory results in use.

The OB-6 is an analog VCO based synthesizer, there is no perfectly calibrated unit and variation from unit to unit and voice to voice is part of it's charm and sonic signature. If you want perfect, get a digital synth.
I just returned from the store. Yeah, seems to be like you described for other units as well (one Prophet 6, one OB6). Ok, it is what it is. Waste of energy to get unhappy about it. Maybe I will get me a Prophet 6 as well, since I feel even more "emotionally connected with it" as somebody described it here (who actually felt disconnected with the P6, hard to believe).
Old 9th March 2019
  #5661
Lives for gear
 

On "free running voltage controlled oscillators oscillators".
Are there any VCOs that are not free running ?
I thought only some old digital (software) oscillators stopped when not in use.
Old 11th March 2019
  #5662
Gear Nut
 
shmuelyosef's Avatar
Quote:
Originally Posted by Channelizer View Post
I don't think anyone can claim an add-on midi controller as their idea. It has been done before for DX7s, Matrix 1000s, you name it, just about everything apart from the OB6! Good luck with yours anyway.

If anyone wants to get involved with helping me spec this then I can take their ideas into account, thank them for their input, and if they buy one will have it better matching their requirements. If not, I'll take this off-line and just get on with it. You might find it for sale somewhere if you bother to look.
It appeared to be <infindebula>'s idea...
Old 11th March 2019
  #5663
Gear Addict
 

Just got my baby outta pawn. A couple months ago I saw a damn near mint condition mono/poly, but was short on funds so I loaned the ob6 to the pawnshop.. wouldn't say regrettably either because now I have a monopoly and ob6. But have definitely been missing it the whole time and man now that we're reunited again I am so happy . Such an awesome synth. And one of the few synths where I can just get lost in just the presets for hours at a time without even too much knob twiddling. Gotta hand it to the programmers, they made some really beautiful patches
Old 11th March 2019
  #5664
Lives for gear
 
John01W's Avatar
Quote:
Originally Posted by Syntheseyez View Post
Just got my baby outta pawn. A couple months ago I saw a damn near mint condition mono/poly, but was short on funds so I loaned the ob6 to the pawnshop.. wouldn't say regrettably either because now I have a monopoly and ob6. But have definitely been missing it the whole time and man now that we're reunited again I am so happy . Such an awesome synth. And one of the few synths where I can just get lost in just the presets for hours at a time without even too much knob twiddling. Gotta hand it to the programmers, they made some really beautiful patches
Congrats on the welcome home for your OB-6!
Old 11th March 2019
  #5665
Gear Addict
 

Quote:
Originally Posted by John01W View Post
Congrats on the welcome home for your OB-6!
Thanks mang! So stoked to have it back home
Old 8th April 2019
  #5666
Lives for gear
 
bobule's Avatar
 

Hey yall, I found an old 100% OB-6 arrangement of a Squarepusher midi file I made a few years ago for his 'Midi Sans Frontiers' project.
Never submitted, never finished or mixed but I enjoyed listening to it while archiving old tunes last night so thought I would share.



I still love the OB-6 to bits!
Old 8th April 2019
  #5667
Lives for gear
 
TheBrightSide's Avatar
Quote:
Originally Posted by bobule View Post
Hey yall, I found an old 100% OB-6 arrangement of a Squarepusher midi file I made a few years ago for his 'Midi Sans Frontiers' project.
Never submitted, never finished or mixed but I enjoyed listening to it while archiving old tunes last night so thought I would share.



I still love the OB-6 to bits!

Excellent, some great OB-6 action here.
Old 24th April 2019
  #5668
Here for the gear
 
RobertSlump's Avatar
 

I am about to pull the trigger on this awesome synth.
However, i do want to be certain before buying the OB6.. how does this synth perform as a masterkeyboard?

I will have a simple setup:

OB6 as masterkeyboard.
Novation Peak.
Scarlett 18i20 (2nd Gen).
Cubase 9 with software synths and plugins.

i want use the OB6 as a masterkeyboard to trigger the Novation Peak, and to trigger software synths, plugins etc.. within Cubase 9.
I have been reading about issues with this when done through USB.. does anyone reconize this or are these not common issues?
Old 24th April 2019
  #5669
Gear Nut
 

Quote:
Originally Posted by RobertSlump View Post
I am about to pull the trigger on this awesome synth.
However, i do want to be certain before buying the OB6.. how does this synth perform as a masterkeyboard?

I will have a simple setup:

OB6 as masterkeyboard.
Novation Peak.
Scarlett 18i20 (2nd Gen).
Cubase 9 with software synths and plugins.

i want use the OB6 as a masterkeyboard to trigger the Novation Peak, and to trigger software synths, plugins etc.. within Cubase 9.
I have been reading about issues with this when done through USB.. does anyone reconize this or are these not common issues?
yes and its apparent with ALL DSI's

I have the rev2 really bad using it via usb but slap a reg midi cable in you're good to go
Old 24th April 2019
  #5670
Deleted cec0d5b
Guest
Quote:
Originally Posted by Pajor Hill View Post
I have the rev2 really bad using it via usb but slap a reg midi cable in you're good to go
Really bad how?

I’ve got a Rev2 and use it exclusively via USB. What problems have you experienced?
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump