Author |
Message |
torsig1967
Joined: Mar 29, 2006 Posts: 162 Location: Sweden
Audio files: 2
G2 patch files: 1
|
|
Back to top
|
|
|
kkissinger
Joined: Mar 28, 2006 Posts: 1353 Location: Kansas City, Mo USA
Audio files: 41
|
Posted: Wed Mar 04, 2009 10:47 am Post subject:
This is normal |
|
|
The antidote is to put a little touch of lag (i.e., glide or portamento) into the CV that controls the oscillator's frequency. Even if you don't want an audible slide, you can add just enough that you don't get the "pop".
Does the DSI-Evolver allow you to do this?
Another work around would be to roll off some of the bass -- however who wants to do that? _________________ -- Kevin
http://kevinkissinger.com |
|
Back to top
|
|
|
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Wed Mar 04, 2009 11:32 am Post subject:
|
|
|
Heh - I can hear this on my Evolver too! Glide doesn't help - the lowest setting is too audible (at least for my taste). I can't think of anything that'd help here. It's curious that it's mostly noticeable when you switch from a low note to a high note. You'd think that if the oscillator pitch was lagging behind the envelope you would hear a more audible high pitched pop on the low note. The charm of analog I guess...
/Stefan _________________ Antimon's Window
@soundcloud @Flattr home - you can't explain music |
|
Back to top
|
|
|
torsig1967
Joined: Mar 29, 2006 Posts: 162 Location: Sweden
Audio files: 2
G2 patch files: 1
|
Posted: Wed Mar 04, 2009 12:45 pm Post subject:
|
|
|
Antimon wrote: | Heh - I can hear this on my Evolver too! Glide doesn't help - the lowest setting is too audible (at least for my taste). I can't think of anything that'd help here. It's curious that it's mostly noticeable when you switch from a low note to a high note. You'd think that if the oscillator pitch was lagging behind the envelope you would hear a more audible high pitched pop on the low note. The charm of analog I guess...
/Stefan |
Seems my unit isn't broken then, if you have the same phenomenon.
Why it's more noticeable in that direction I have no idea. Maybe it's faster "downwards". I'm pretty sure it's the oscillator frequency lagging behind the envelope. Adding a minimum of glide only makes things worse in my opinion.
What I'd want to do is get inside the unit and trim the glide rate so it's faster. But I have no idea if that's possible. Are there any service manuals out?
Unfortunately I don't find this charming at all. I think it's evidence of a poor circuit design and it renders the unit useless for the kind of jumping basselines I wanted to use it for. I have used many analog synths over the years and not even a cheap design like the TB303 has this. I'm quite disappoined since the Evolver is supposed to be some kind of modern Prophet. |
|
Back to top
|
|
|
torsig1967
Joined: Mar 29, 2006 Posts: 162 Location: Sweden
Audio files: 2
G2 patch files: 1
|
|
Back to top
|
|
|
alanzo
Joined: Jan 14, 2009 Posts: 13 Location: Burlington, VT
|
Posted: Fri Apr 24, 2009 9:44 am Post subject:
|
|
|
Has anyone found a way around this? I think it has to do with a glide of 0 not being "off", rather being really fast. There is an "off" for the glide, but it then makes the OSC mono-tone. |
|
Back to top
|
|
|
EdisonRex
Site Admin
Joined: Mar 07, 2007 Posts: 4579 Location: London, UK
Audio files: 172
|
Posted: Sat Apr 25, 2009 5:52 am Post subject:
|
|
|
Just as a curiosity I tried this myself. If I make sure all 4 voices are low, and then hit a high note multiple times, I get the pop for each voice, the first time in. I also get the pop coming back to the low notes.
Experimenting with it, it happens both in normal and finger mode. Adding glide does not really work as the glide is too noticeable. VCA attack at 15 will mask it, at the expense of the attack being a tiny bit mushy.
Also I notice that it does it actually all the time, for relatively short intervals too. The first frequency change of the analog oscillators in each voice gives a bit of a pop, even for fifths, octaves, 2-3 octaves, etc.
It is more noticeable with less harmonic content, for instance a triangle wave as opposed to a saw. But then again, it's still there. Has anybody written to Dave Smith about this? This acts more like a microcode problem than anything. _________________ Garret: It's so retro.
EGM: What does retro mean to you?
Parker: Like, old and outdated.
Home,My Studio,and another view |
|
Back to top
|
|
|
alanzo
Joined: Jan 14, 2009 Posts: 13 Location: Burlington, VT
|
Posted: Sat Apr 25, 2009 5:56 am Post subject:
|
|
|
EdisonRex wrote: | Has anybody written to Dave Smith about this? This acts more like a microcode problem than anything. |
That's what I'm thinking, too. They may be able to fix it with a software update. I wrote to them yesterday and linked them to this thread ... I'll post their response when I get it. |
|
Back to top
|
|
|
EdisonRex
Site Admin
Joined: Mar 07, 2007 Posts: 4579 Location: London, UK
Audio files: 172
|
Posted: Sat Apr 25, 2009 6:00 am Post subject:
|
|
|
Sounds like a reasonable course of action. Guess we can see if they respond to you then. _________________ Garret: It's so retro.
EGM: What does retro mean to you?
Parker: Like, old and outdated.
Home,My Studio,and another view |
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005 Posts: 4044 Location: England
Audio files: 32
G2 patch files: 15
|
Posted: Sat Apr 25, 2009 10:48 pm Post subject:
|
|
|
DSI always have answered my emails very promptly, sometimes Dave answers them but the other guys seem very knowledgable as well.
I wouldn't hold out too much hope for an update though I have had a couple of requests for changes since the Poly Evolver came out and havnt seen an update yet. |
|
Back to top
|
|
|
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Sun Apr 26, 2009 2:26 am Post subject:
|
|
|
I guess one could try downgrading to an earlier firmware version (have to ask DSI for this, no old versions on their web site) for one or all of the processors, if the pop issue is critical enough. Just one of my random crazy thoughts.
Edit: I'm not saying that this would fix the issue, just thinking that maybe the pop wasn't there before and has been introduced somewhere along the way. Anyone with old firmware in their Evolver that can check this out?
/Stefan _________________ Antimon's Window
@soundcloud @Flattr home - you can't explain music |
|
Back to top
|
|
|
alanzo
Joined: Jan 14, 2009 Posts: 13 Location: Burlington, VT
|
Posted: Sun Apr 26, 2009 5:01 am Post subject:
|
|
|
Antimon wrote: | Anyone with old firmware in their Evolver that can check this out?
/Stefan |
Version 1.0 of my Poly Evolver Rack had the issue. Here's the response from DSI :
Code: |
The short answer is not really. The Curtis analog ICs we use have some artifacts that are unavoidable in our designs. Since the ICs allow us to build real analog synths at a reasonable price, it’s just a trade off that comes with the instruments. These artifacts are masked in normal use, so generally they just end up adding to the random variations in the sound that make analog synths sound more “natural” than their digital counterparts, which of course are always perfect in repeating a sound. I hope this answers your question, and please don't hesitate to contact us with any other questions regarding your Poly Evolver Keyboard.
Many thanks,
Tracy
Dave Smith Instruments
|
I think that means there's no way around it :\ -- They must be using different chips in the Prophet 08 because it does not have the issue (I used to own one). |
|
Back to top
|
|
|
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Sun Apr 26, 2009 5:13 am Post subject:
|
|
|
A quick and clear answer, at least. Since torsig brought this up I've actually started to notice it more when playing. I'm into messy, trashy sound though (what I seem to do most with the Evolvers), so I'm not that bothered.
/Stefan _________________ Antimon's Window
@soundcloud @Flattr home - you can't explain music |
|
Back to top
|
|
|
torsig1967
Joined: Mar 29, 2006 Posts: 162 Location: Sweden
Audio files: 2
G2 patch files: 1
|
Posted: Thu Apr 30, 2009 8:26 am Post subject:
|
|
|
I don't agree this artefact makes the synth sound "warm" it rather reminds me of how some early virtual analog synths sounded; cheap and sloppy.
I remember the JD800 having simliar problems with artefacts on percussive sounds.
Personally I dislike this so much I'm going to get rid of my DSI. |
|
Back to top
|
|
|
cerebrosis
Joined: Apr 12, 2009 Posts: 5 Location: arizona
|
Posted: Sun May 31, 2009 11:53 pm Post subject:
|
|
|
the evoler can get crackly and noisey from simply having a single parameter up to high. if the clip light comes on when you hear the cracks and pops for sure something is turned up to high. the digital osc clip at about 50% volume. |
|
Back to top
|
|
|
torsig1967
Joined: Mar 29, 2006 Posts: 162 Location: Sweden
Audio files: 2
G2 patch files: 1
|
Posted: Mon Jun 01, 2009 4:11 am Post subject:
|
|
|
cerebrosis wrote: | the evoler can get crackly and noisey from simply having a single parameter up to high. if the clip light comes on when you hear the cracks and pops for sure something is turned up to high. the digital osc clip at about 50% volume. |
Yes, that's probably true. But the problem I've described has nothing to do with this. It's not related to the signal level. It comes from the (analog) oscillators. |
|
Back to top
|
|
|
dorremifasol
Joined: Sep 28, 2006 Posts: 814 Location: Barcelona, Spain
Audio files: 7
G2 patch files: 49
|
Posted: Wed Jun 03, 2009 12:56 am Post subject:
|
|
|
When I studied the Cheetah MS6 I found the same issue.
Have you checked if the portamento value / setting is zero? _________________ Cheers,
Albert |
|
Back to top
|
|
|
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Wed Jun 03, 2009 2:42 am Post subject:
|
|
|
dorremifasol wrote: | When I studied the Cheetah MS6 I found the same issue.
Have you checked if the portamento value / setting is zero? |
Yep. The issue seems to be that the portamento cannot be set to really be zero. Could it be something that could be fixed in an update?
/Stefan _________________ Antimon's Window
@soundcloud @Flattr home - you can't explain music |
|
Back to top
|
|
|
cerebrosis
Joined: Apr 12, 2009 Posts: 5 Location: arizona
|
Posted: Wed Jun 03, 2009 12:06 pm Post subject:
|
|
|
torsig1967 wrote: | cerebrosis wrote: | the evoler can get crackly and noisey from simply having a single parameter up to high. if the clip light comes on when you hear the cracks and pops for sure something is turned up to high. the digital osc clip at about 50% volume. |
Yes, that's probably true. But the problem I've described has nothing to do with this. It's not related to the signal level. It comes from the (analog) oscillators. |
i dont have the glide problem with my pek, sorry. |
|
Back to top
|
|
|
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Wed Jun 03, 2009 1:17 pm Post subject:
|
|
|
cerebrosis wrote: |
i dont have the glide problem with my pek, sorry. |
Good for you!
Though bad for us others. If everyone doesn't have this problem then it may be down to hardware problems in some units. I've started to get annoyed by it too, now that torsig1967 brought it up. Oh well, the PEK is still a great machine anyway too me.
/Stefan _________________ Antimon's Window
@soundcloud @Flattr home - you can't explain music |
|
Back to top
|
|
|
EdisonRex
Site Admin
Joined: Mar 07, 2007 Posts: 4579 Location: London, UK
Audio files: 172
|
Posted: Thu Jun 04, 2009 12:52 am Post subject:
|
|
|
You know, other than the academic interest in the problem, it completely doesn't matter to me. I am quite happy with my PEK and use it on an increasing basis. If I let every zit with every piece of hardware I own get to me, I wouldn't own any hardware. (and if I let every zit with every piece of software I own, I wouldn't have that either). _________________ Garret: It's so retro.
EGM: What does retro mean to you?
Parker: Like, old and outdated.
Home,My Studio,and another view |
|
Back to top
|
|
|
Antimon
Joined: Jan 18, 2005 Posts: 4145 Location: Sweden
Audio files: 371
G2 patch files: 100
|
Posted: Thu Jun 04, 2009 3:18 am Post subject:
|
|
|
What do you mean? I threw out my NMG2 the second I found out that the delays are only 16-bit!
I can accept though that this pop can be too disturbing if you're trying to make soft sounds with fast attack - if you want those sounds you get to choose some other piece of gear.
/Stefan _________________ Antimon's Window
@soundcloud @Flattr home - you can't explain music |
|
Back to top
|
|
|
torsig1967
Joined: Mar 29, 2006 Posts: 162 Location: Sweden
Audio files: 2
G2 patch files: 1
|
Posted: Sat Jun 06, 2009 2:27 am Post subject:
|
|
|
cerebrosis wrote: |
i dont have the glide problem with my pek, sorry. |
That's interesting.
Could you record and post a short sample playing and sounding like the first clip I posted here? It would be interesting to hear the difference.
If you do that remember to use only one analog oscillator turn all kinds of modulation off and use high bitrate on the mp3 encoding. So that no other artefacts appears.
By the way, I wrote DSI and asked about this and got this answer from "Andrew":
Quote: | "There is an artifact of the analog circuitry that you can sometimes hear when going from low to high. (Actually, I believe it's there going the other way, too, but it's less noticeable.)" |
Seems they are well aware of the problem.
So I got myself a Novation Super Bass Station instead (analog too) and it's quite funny (or annoying) but this machine has a similar problem but less noticeable.
None of my truly vintage analog synths (MicroMoog, Korg MP, Pro One and others) has this artefact. So it has nothing to do with being "vintage" or being "analog". It's just poor and cheap circuit design. |
|
Back to top
|
|
|
Munchausen
Joined: Jun 18, 2009 Posts: 3 Location: PA
|
Posted: Fri Jun 19, 2009 4:17 am Post subject:
|
|
|
Seems I have the same problem with my MEK. I have a feeling it's on all of them. For some reason only a few people have ever noticed....strange. It's a pity that there seems to be no effort to fix it. |
|
Back to top
|
|
|
recnsci
Joined: Jun 24, 2009 Posts: 4 Location: Belgrade
|
Posted: Wed Jun 24, 2009 7:46 am Post subject:
|
|
|
torsig1967 wrote: |
Seems they are well aware of the problem.
So I got myself a Novation Super Bass Station instead (analog too) and it's quite funny (or annoying) but this machine has a similar problem but less noticeable.
None of my truly vintage analog synths (MicroMoog, Korg MP, Pro One and others) has this artefact. So it has nothing to do with being "vintage" or being "analog". It's just poor and cheap circuit design. |
Hi there,
IMHO reason is not poor or cheap circuit design, it's one of side effects of using DCO (whitch Bassstation uses as well).
Classical VCO has circuit that converts voltage into current, and that current charges capacitor, which results in linear rise of capacitor voltage. When voltage reaches some threshold, electronic switch of some sort is triggered that discharges capacitor. Thus, you have sawtooth waveform on capacitor with constant amplitude, but with ramp rise rate (and thus VCO frequency) proportional to capacitor current. Problem is (among other things) that you need high quality voltage to current converter at VCO input, which means its expensive and a bit trickie to do.
DCO uses same circuit as VCO, but discgharge of capacitor is triggered from some sort of digital timer (usually on some sort of processor). It's like having VCO with sync input, and you sync it to pulses from digital timer. Good part is that now you can use much simpler VCO, because precision comes form digital timer. OTOH there are two problems: ramp rate is now independently controlled from oscilator frequency and (much bigger issue) digital timers are not smoothest things to control.
Timer is usually independent block in MCU. It has register which tells how many master clock cycles it counts, befor triggering and resetting it self to start new count (a bit like divide down scheme in organs/string machines). Now imagine you have new note on with different pitch somewhere in middle of counting period. To have smooth transition (like with VCO) MCU has to retrive current timer value, somehow compute relative position in cycle and generate new count register value that would result in VCO like change in pitch. I don't want to go in details, but this is not simple AND sometime not doable at all.
So, easier scheme would be to simply update timer with new value when it finishes it current cycle OR to reset timer (and thus DCO).
OK back on topic, when you examine waveforms posted by torsig, you see that after low note to high note transition you have glitch at start of first high note that is much longer in duration that high note period. I suspect that timer that controlls DCO counts whole cycle and than interrupt routine for count register setting is triggered. And, on first low note after high note you have glitch as well, but much shorter than low note period (and comparable to high note period).
Bottom of the line, DSI can probably do nothing about this.
There is however workaround (sort of). Dave could probably squeeze DCO reset on note on. That means that DCO would restart on same position on each note. You would loose oscilator beating, but on other side it could be quite usefull for bass sequences. A little side note: Juno 106 has DCOs that restart on note on (due old simple timer); some percieve this as sterile, but I find that consistency of note attack is usefull in lot of occasions. Further more, this DCO reset option could be nicely put in, for example, Osc Slop page in menu as additional item. Somebody could ask DSI if this is doable.
OTOH, I could be completely wrong on all accounts. But to further investigate this issue, I would recommend this test: first use Saw wave, because all other waves are derived from saw, and on saw its easiest to see whats going on. Second, don't trigger envelops. Use long note with full sustain level, but send one of sequencers (or LFOs) to Osc pitch. When recorded, this would provide much better insight on what actually happens.
Sorry for long post
cheerz
Urosh |
|
Back to top
|
|
|
|