Causes of (and Solutions to) SY-1000 Poor Pitch to Midi Performance

Started by Bluesbird, October 22, 2021, 08:13:10 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

JustinH

Thank you Bluesbird for the glitch chart.  My use of the SY-1000 is mostly as a guitar effects unit, midi sound module driver (strings, pads), and combining the two.  I have a Roland JV-1010 for the PCM sounds.  Changing to Type 2 Chromatic and Low Velocity cut of 5 got rid of my glitches! I can now play the strings, organ, and pads from the JV-1010 very cleanly in live performance without worrying about that occasional odd sound!

My setup is simply a U.S. Roland Ready Strat (GK-3 I think), the SY-1000, and midi out to the Roland JV-1010, using the effects loop to bring back the line out from the JV-1010.  Wish I could do it all in one box, but this works for me.  I manually set the PCM sound from the JV-1010.  I need to learn a way for my SY-1000 to send a patch select to the JV-1010.

This simple information has really made this perfect for my purposes!  Thanks again!

Justin

Unlimbo

Quote from: admin on December 17, 2021, 10:41:41 AM
Live gig i use the faster GR-33 if I need pianos and saxophones
Is the GR-33 faster than the GR-55?
Six string synths.

admin

Quote from: Unlimbo on December 18, 2021, 09:11:59 AM
Is the GR-33 faster than the GR-55?

For my use. Triggering ,,internal synths. Yes GR-33 is night and day for me

It's also faster for Guitar to MIDI. Vs GR-55 , SY-1000

And GR-33 is perfect MIDI Mode 4 tone module for Tripleplay ( for independent MIDI CH per string

Roland GR-33 info
https://www.vguitarforums.com/smf/index.php?board=140.0

jozart

Why is an external MIDI filter better than the built-in LOW VELO CUT?

Pg. 30 & 79

admin

Quote from: jozart on December 18, 2021, 11:30:11 AM
Why is an external MIDI filter better than the built-in LOW VELO CUT?

Pg. 30 & 79

Agreed

And All tips & tricks for GP-10 Guitar to MIDI apply to SY-1000 guitar to MIDI Strategies
https://www.vguitarforums.com/smf/index.php?topic=11913.msg115410#msg115410

And GR-55 Guitar to MIDI tips -stil apply to SY-1000
https://www.vguitarforums.com/smf/index.php?topic=3787.msg25564#msg25564

Shawn's tips
https://www.vguitarforums.com/smf/index.php?topic=6878

jozart

Quote from:  Bluesbird on December 18, 2021, 12:48:40 PM
Because the low velocity cut (even when turned up to maximum) on the SY-1000 does not stop the output of low velocity glitch notes which the low velocity filter on the DAW will catch and filter out. Why don't you give the low velocity cuts on Ableton or Bitwig a try and then come back and comment. I think I've described well enough the method of how to do that.  Have you even tried what I suggested, admin?

Here is a link for a video screen capture that I just did that shows the 10 velocity glitch notes. Look at the bottom of the screen where you see the midi effect rack.

https://1drv.ms/v/s!Aje0A-m1QrOz8DaUJsNDYsVczAhE

This is on Feel 2, chromatic 2 hitting the low E string at the 8th fret.  Feel 1, 3, and 4 put out low velocity glitch notes at the level of 3. I don't know why this is the case--it is just what I have directly observed.  As you can see, the velocity cut filter is set so that notes below the red line won't come through and create garbage in your vst synth. 

Otherwise, just ignore this thread and do what you want.

Hey, thanks for the thorough explanation now and before. It wasn't clear to me what the difference was. I play 5-string violin with a Cantini pickup and I use an iPad, so your guitar experience with Ableton doesn't always translate and I wanted to be clear before I set out.

chrish

Quote from:  Bluesbird on December 18, 2021, 12:48:40 PM
Because the low velocity cut (even when turned up to maximum) on the SY-1000 does not stop the output of low velocity glitch notes which the low velocity filter on the DAW will catch and filter out. Why don't you give the low velocity cuts on Ableton or Bitwig a try and then come back and comment. I think I've described well enough the method of how to do that.  Have you even tried what I suggested, admin?

Here is a link for a video screen capture that I just did that shows the 10 velocity glitch notes. Look at the bottom of the screen where you see the midi effect rack.

https://1drv.ms/v/s!Aje0A-m1QrOz8DaUJsNDYsVczAhE

This is on Feel 2, chromatic 2 hitting the low E string at the 8th fret.  Feel 1, 3, and 4 put out low velocity glitch notes at the level of 3. I don't know why this is the case--it is just what I have directly observed.  As you can see, the velocity cut filter is set so that notes below the red line won't come through and create garbage in your vst synth. 

Otherwise, just ignore this thread and do what you want.
you quoted Jozart. Is it me or are you still pissed? Admin thanked you for the work you did on this and for sharing that work here.

Personally I read the thread and noted how you are dealing with the P2M limits of the SY1000. Could you post some real world unedited sound examples of your methods in action?


admin

Quote from:  Bluesbird on December 20, 2021, 07:25:15 AM
In Reaper DAW, you can filter the low velocity glitch notes using the Midi Velocity Scaler/Compressor. The volume offset is the only setting you need change. In the attached example the volume offset is set to -15, which will turn notes from 15 and below off.  Play Feel 1, 3, and 4 create glitch notes at a velocity of 3, so you can set it to a volume offset of -4 for these settings.  On Play Feel 2, velocity glitch notes of 10 are produced so you need to set it a bit higher. A setting of -11 on volume offset when using Play Feel 2 will do the trick.





But this strategy is Useless for live performance at gigs.

admin


Chumly

Perhaps the best sig created by human hands "Otherwise, just ignore this thread and do what you want."
I think it's much more interesting to live not knowing than to have answers which might be wrong. - Richard P. Feynman

jassy

Bluebird thanks for this thread and your findings.
In this P2M area Roland seems to do things very absurdly.

chrish

Quote from: jassy on December 20, 2021, 01:24:39 PM
Bluebird thanks for this thread and your findings.
In this P2M area Roland seems to do things very absurdly.
except that my GR50 P2M works great, and that unit is from the 90's.

Wasn't that synth one of the early guitar synths that Al DiMeola used to get that , ,cleanly picking each note and shredding faster than a speeding bullet, pan pipe tone?.

Of course that was an internal sound, but still, pared with aGodin Multiac SA or Wilcox Atlantis hex fx and the right external sound module or synth, the GR50  P2M works great.

Roland was a Pioneer into this type of hex processor technology, not "absurdly" at all.

kimyo

Quote from: chrish on December 20, 2021, 01:53:39 PMRoland was a Pioneer into this type of hex processor technology, not "absurdly" at all.

like sony was a 'Pioneer' with betamax? 

this thread is about the sy-1000's performance, not the gr-50's.  they are light years apart.  what is absurd is the degradation over time in roland's ptm tech.

chrish

Quote from: kimyo on December 20, 2021, 02:38:38 PM
like sony was a 'Pioneer' with betamax? 


no like Roland was a pioneer in hex processing, guitar synths, VG series cosm, etc. And now the SY1000 is a breakthrough product in sound quality for an analog modeling subtractive guitar synth. It's P2M is not that relevant when there are other tools available that do a better job and don't require elaborate workarounds.

Perhaps Roland should have left it off, but then we know they would be taking an online thrashing for that decision.

When I was a carpenter I never blamed my tools for something that I couldn't create.





kimyo

Quote from:  Bluesbird on December 20, 2021, 04:10:00 PMWith the low velocity midi filter, the SY-1000 pitch to midi performance moves into second place below Fishman and leaves Midi Guitar 2 in the dust.

are you saying that the filtering has decreased the latency?  your previous trial put the sy-1000 behind the gr-55 and nowhere near the tripleplay/midiguitar.  what is your new number for the sy-1000?

Quote from:  Bluesbird on October 24, 2021, 08:47:55 AMI just did a full test along the lines of the Wayne Joness speed tests: https://www.joness.com/gr300/MIDI_SPEED.htm

1   G50, 11.08
2   GI-20, 15.96   
3   GI-10, 18.58      
4   GR-55, 19.01   
5   Z3, 20.11   
6   GR-55, 20.31   
7   GR-700 Turbo, 28.54
8   GR-700, 29.87
9   MC1, 33.62
10   GM-70, 40.39   

The SY-1000 in my tests sits closely to the GR55, between it and the GR700.  I believe the SY-1000 is performing up to Roland specs and I suspect that the dissatisfaction stems from other pitch to midi solutions that now exist on the market.  Monophonic mode of Jam Origin Midi Guitar 2 is close to the Tripleplay, and the Tripleplay scores are similar to the G50 at 11.08 ms. 

kimyo

it does seems like chromatic modes 2 and 3 will have lower latency than off or 1.

certainly the elimination of glitches alone will improve the playing experience, thank you for this exploration, i will definitely give this another shot now. 

i have the gr-55 hooked up atm and will have a look to see if it generates glitch notes like the sy.

kimyo

on the gr-55 with velocity cut set to 0 i don't get any glitch notes.

there is a difference, though between chromatic on and off, if off, the gr-55 sends two pitch wheel commands per note prior to the note on.

if the sy is doing something similar, i could see that some synths would get confused by strummed chords.

fwiw, a quick check on the gr-55 shows that chromatic on reduces latency by 3-4 msec.

jassy

Quote from: chrish on December 20, 2021, 01:53:39 PM
except that my GR50 P2M works great, and that unit is from the 90's.

Wasn't that synth one of the early guitar synths that Al DiMeola used to get that , ,cleanly picking each note and shredding faster than a speeding bullet, pan pipe tone?.

Of course that was an internal sound, but still, pared with aGodin Multiac SA or Wilcox Atlantis hex fx and the right external sound module or synth, the GR50  P2M works great.

Roland was a Pioneer into this type of hex processor technology, not "absurdly" at all.

With absurdly I mean, how would you rate a company that makes a brilliant product in the 90s (p2m) and in 2020 implements "the same" technology but with worse results? Isn't this absurd?

AlakaLazlo

Quote from:  Bluesbird on December 12, 2021, 09:34:34 AM
More good news on the SY-1000 pitch to midi front.  Think of this as a digital scrunchie.

First, my compliments to Bluesbird for attempting to figure out a way to improve the usefulness of this product and then sharing his (obviously time intensive) results here. I'm constantly disapointed when people legitimatly try share hard won knowlege and get attacked for it, which is why I've backed off doing so despite having been working with guitar synths since the 70s. (Remember the 360 Systems Slavedriver?) Post "I like X" or "hey, try Y" on an internet board and the onslaught of "screw you, Z is better and everyone knows it!" is enough to make one consider drinking heavily. Moreover, criticize a product because it doesn't appear to perform up to what one expects and others (I hate the term fanboys) come out and attack you for expressing a valid opinion. (I love my Helix, but the Helix rhetoric on-line is pretty harsh...)

I tried the SY1000 at NAMM and was not impressed. I also sat in on the GF meeting with the very pleasant folks from Roland. I decided not to buy one. Having followed this devices various threads, I have been somewhat disapointed (and share others' frustrations) that this device doesn't seem well implimented and that obvious firmware updates aren't forthcoming in a timely manner.

I've only owned one Roland P2M device (the GI-20) which I gave up on relatively quickly. I've tried them all (GR300 to the GR55) and they just never really worked for me (of course YMMV). I understand the physics of converting audio to midi and suspect that the updates made by Terratec (later Axon) which apparently carried over to the Fishman products may be the theoretical limits.

That being said, this multi-channel midi-filter processing scheme looks very promising. Looks to be replacable on pretty much any serious DAW and with the proper tools, might be replicable in a live situation. I'm going to do some experimenting in Cubase (I have, but almost never use Ableton) and report back whether this helps the P2M functions on an Axon AX50. I suspect it may. If so, it would seem relatively simple for Roland to integrate a programmable midi filter in the SY1000 which can be set on a per-patch basis (unless the firmware uses so much of the memory that there isn't room) or in an external device that can store various midi filter configs and recall them via midi CC. 

Thanks again Bluesbird. Keep up the good work.
Hexstainocaster, Fender Strat and Electric XII, Godin ACS, Axon AX50 - Moog One, IIP and Mini, SEM, Dot.Com/Moon/STG/FSFX 110, Cubase Pro, 2xMR816, HR824, NS10M, Komplete, Omnisphere/RMX/Trilian, Z3ta+, Analog Lab, Slate MTi2, ML1 and Everything Bundle, Social Entropy Engine, ESQ1, DX7, Lavalamp.

shawnb

I agree - thanks for sharing, Bluesbird!

There always seems to be some issue with P2M on these devices.  Every generation...

Pre- GR-55, it was always slightly glitchy and the dynamics were flat.  But the sounds were actually pretty warm and usable; I love my old GR-30.

Starting with the GR-55, I believe they attempted to enhance the dynamics and add new P2M features such as nuance.  To their credit, dynamics were improved, but the glitches became far more pronounced.  Negating any benefit.

It feels like MIDI out on all subsequent products is similar.  You've gotta find some set of tricks to tame the glitches...  So thanks for sharing!

(I have always liked enhancing my guitar sounds with some synth sounds.  That's what I aim for with these products.  For me, the SY series is the best by far, and I don't have to use MIDI at all...)
Address the process rather than the outcome.  Then, the outcome becomes more likely.   - Fripp

admin

QuoteI have always liked enhancing my guitar sounds with some synth sounds.  That's what I aim for with these products.  For me, the SY series is the best by far, and I don't have to use MIDI at all...)

I feel the same

SY-1000 is perfect form factor

Brak(E)man

Quote from: shawnb on December 21, 2021, 11:49:30 AM

It feels like MIDI out on all subsequent products is similar.  You've gotta find some set of tricks to tame the glitches...  So thanks for sharing!

(I have always liked enhancing my guitar sounds with some synth sounds.  That's what I aim for with these products.  For me, the SY series is the best by far, and I don't have to use MIDI at all...)

Quote from: admin on December 21, 2021, 11:53:12 AM
I feel the same

SY-1000 is perfect form factor

I would agree wholeheartedly.

Having started out with gtm in the 80s and using the same technique to get rid of glitches
(but not in real-time back then of course , instead with auto edit ) , but also having the guitarsound in the foreground.
With GP-10 , SY-300 I went back to using the guitarsound in front with synthsounds alongside because of necessity. ( GP due to less possibilities compared to vg99 , SY 300 due to lack of polyphony )

I'm doing the same with SY-1000 , that even with some of the major complaints I have , is the best so far imho.

( it's still baffling that the alt tune doesn't work work for me since GP-10 and gtm that worked imho best with GP-10 isn't up to par in the SY but...)
swimming with a hole in my body

I play Country music too, I'm just not sure which country it's from...

"The only thing worse than a guitar is a guitarist!"
- Lydia Lunch

chrish

Quote from: jassy on December 21, 2021, 02:20:58 AM
With absurdly I mean, how would you rate a company that makes a brilliant product in the 90s (p2m) and in 2020 implements "the same" technology but with worse results? Isn't this absurd?
not to me. As Shawnb stated Roland was trying to improve the P2M dynamics and it appears by doing so they tried to cram too much into the DSP. Perhaps they did this in response to customer feedback.  It appears Bluebird got it working and when Roland implements his work in new firmware then bobs your uncle.

Back when the SY300 came out, many here were hoping for a hex version and some said adding hex processing to it would be too expensive.

Well Roland came through making an inexpensive hex version of the dynamic synth and added an osc synth, a GR300 model and as requested the famous vioguitar.. The normal 1/4" input dynamic synth seems to also be an improvement over the one in the SY300.

The SY1000 is one heck of a monster guitar synth.

Over the years, Roland has given us so much to add to our guitar creativity that I would never use the word absurd to describe them.

BROCKSTAR

"The normal 1/4" input dynamic synth seems to also be an improvement over the one in the SY300"

It's not though. It's actually the same and I proved that by doing a back to back video starting at 4:06 with both units. Exactly same sound and everything. The same technology via normal 1/4 is the same in the old sy-300 as it is in the new sy-1000, sy-1 and sy-200.


aliensporebomb

All I can see is with v1.06 and v1.07 I have massive increase in guitar to midi performance. 
My music projects online at http://www.aliensporebomb.com/

GK Devices:  Roland VG-99, Boss GP-10, Boss SY-1000.