hi everyone! I want to ask Roland fr accordion users, if you noticed the very high latency of the sound coming out of L/mono. I did a test, I think it's quite relevant. I recorded with the microphone in cubase on a track, the hit of the keyboard with a spoon of metal and on another track/other input in the audio interface, the sound from the L/mono output. I used the same audio cables (klozz mc5000 balanced) both at the microphone and at the output accordion, neutrika plugs. the sound from the accordion L/mono output reaches 5ms after the sound of the key. the stroke of the keyboard, until it makes contact, is approximately 2 mm. it is modified, it cannot be said that it is a delay due to the stroke of the keyboard. on the speaker of the accordion it does not exist or cannot it perceives the latency of the sound. maybe there is a defect in my accordion? I use tones from the balkan set. I have not tried with other sets
Roland fr4x latency
- Thread starter geky
- Start date
Before blaming the accordion, I would ask:
- what kind of computer are you using?
- what kind of audio interface are you using?
- what are your settings? (44.1, 88, 96khz... what is your buffer size)?
- how are you capturing this difference?
- are you seeing the same thing on the RIGHT mono channel?
- are you seeing the issue when in stereo mode?
- did you purchase your 4X new or used?
- what kind of computer are you using?
- what kind of audio interface are you using?
- what are your settings? (44.1, 88, 96khz... what is your buffer size)?
- how are you capturing this difference?
- are you seeing the same thing on the RIGHT mono channel?
- are you seeing the issue when in stereo mode?
- did you purchase your 4X new or used?
Ventura
Been here for ages!
until you plug the Roland audio out directly into a simple audio amplifier
and listen, you cannot assume the latency is the fault of the accordion
and listen, you cannot assume the latency is the fault of the accordion
You need to include a lot more info... latency is 90% of the time a function of the path the sound has to take from:
- key press to audio exiting the audio jack
- audio exiting the audio jack reaching the audio interface preamp
- audio interface preamp sending it to A/D converter
- A/D converter sending it to the computer
- computer sending it to the DAW
- DAW registering/recording the data
- then the whole process moves to the path out the speakers which also can have latency, but is not part of this process I *think* you are showing above.
You would be the first person to have noted this... so I would think it unique to your setup... too many potential variables outside of the 4X that could cause this.
- key press to audio exiting the audio jack
- audio exiting the audio jack reaching the audio interface preamp
- audio interface preamp sending it to A/D converter
- A/D converter sending it to the computer
- computer sending it to the DAW
- DAW registering/recording the data
- then the whole process moves to the path out the speakers which also can have latency, but is not part of this process I *think* you are showing above.
You would be the first person to have noted this... so I would think it unique to your setup... too many potential variables outside of the 4X that could cause this.
While I agree, it would be VERY hard to hear an 8ms latency between L/R channels... I mean it could be anything, up to and including a defective USB cable from the audio interface to PC or bad DAW settings or inaccurate app... so many possibilities as well I would look at before placing the blame on the 4X.
I don't see the relevance of your questions. What does the interface and sample rate matter? the questions have no logic. the comparison is made according to the sound of the keyboard recorded at the same time. if one lags, the other lags. to clarify things, I used rme ucx2, neumann u87ai microphone, sampling rate 48khz/24 bit with buffer 48. round trip latency 1.9 ms. the accordion is bought new. pc configuration: cpu 12900k gen12. 3200-5400 ghz. 8 cores 12 threads. 64gb ram ddr5 4800mhz, ssd m2 nvme gen 4.0×4. Windows 10 ×64 configured Hiperformance. I don't blame anything, I just asked. I don't sing with the sounds from fr4×, I only use it as a midi controller.
this is exactly the problem, that I noticed the latency of the accordion when playing live, both on a digital and analog mixer. I didn't enter the accordion in the audio interface on the microphone input, I entered it on the line, so it's not a preamp on the line. I'm not the only one who noticed latency at Roland fr. latency exceeding 4-5 ms is perceived. not by all instrumentalists.
Ventura
Been here for ages!
if there was ANY latency in the audio output of any Roland
keyboard device, FR or otherwise, then every musician
performing live with sound re-inforcement onstage would
hear an echo
we don't
there is no latency in the audio outputs
if there was any delay in the MIDI keyswitch data then the sound coming
from (any) FR would have an echo between the internal speakers and
the sound coming out of an expander through an onstage amplifier
we don't
because there is no latency in the MIDI note-on note off timing
whether triggering internal sounds, or sounds from an external
sound engine
good luck figuring out what is causing your actual latency problem..
it's not the accordion
keyboard device, FR or otherwise, then every musician
performing live with sound re-inforcement onstage would
hear an echo
we don't
there is no latency in the audio outputs
if there was any delay in the MIDI keyswitch data then the sound coming
from (any) FR would have an echo between the internal speakers and
the sound coming out of an expander through an onstage amplifier
we don't
because there is no latency in the MIDI note-on note off timing
whether triggering internal sounds, or sounds from an external
sound engine
good luck figuring out what is causing your actual latency problem..
it's not the accordion
Because I was trying to understand and help, and there are many sources of latency. If you don’t see the relevancy, then maybe you are not understanding the process. Please go back to post #5 to see why the potential for latency exists.
If you are using it as a midi controller, why aren’t you using the MIDI out to MIDI in by-passing the whole analog side??
If you are seeing something that other FR-4x users are not and have never heard of before, I suggest you reach out to whomever you purchased the V-Accordion from and try explaining it to them.
BTW, to experience what 4ms latency sounds like in a live environment, have the accordion speakers turned on, output the wire analog output to an amplifier, stand 4-feet away from the amplifier’s speaker… there is your 4ms latency. Want 10ms? Stand 10 feet away.
Last edited:
Caps
Active member
Maybe I don't understand what is going on but just out of curiosity (because it is raining and I'm bored) I fed the output of left mono to one input and the right output to the other input on my 50 Mhz 4 channel oscilloscope and saw no time difference between the left and right output. Tried this on both my 4X and 3S. Also tried it on the stereo headphone output of each and still could see no difference.
Ventura
Been here for ages!
God i remember hiding under the table in my makeshift space ship
with my dad's green-screen oscilliscope giving me controls to
manipulate and guide me through the stars
with my dad's green-screen oscilliscope giving me controls to
manipulate and guide me through the stars
Caps
Active member
Ha! When I was 12 or 13 I made a really bad oscilloscope out of a couple old TVs. Bought the schematics from an ad in a popular science magazine.. Sorta worked - no calibration, no time reference other than a crystal oscillator. But I could look at the pretty sine waves and see weird patterns. Looking back, it is a wonder I didn't electrocute myself.
Airy
Active member
Hi Geky,
I don't see your problem resolved, yet. Are you still interested? I also have FR4x, RME interface and Cubase and in the upcoming days I will have some spare time and we could investigate this more closely. Maybe in a private chat to keep it focused? And then we'll tell the group afterwards about our findings?
I don't know what is a reasonable latency, but you cannot expect that in a digital instrument that has to do some processing there would be no latency at all. And also i an acoustic accordion there would be latency because of the time it takes for the reeds to react to the air pressure and produce sound. Now, I do know you're mainly concerned with the midi controller, but that too has to do some processing and the overhead in a more complex instrument like the FR4x may cause additional latency.
But as Jerry pointed out sound travels at a limited speed as well (about one foot per ms), so your 5ms delay is like placing a speaker in an analog mic-amp-speaker chain 5 feet away from you. This is really nothing. I play in a quintet sitting in a half-circle so player 1 and 5 are sitting about 12 feet apart, which is a 12ms delay and we really have the impression everything sounds exactly at the same time. In a larger orchestra (about 25 players the distance between the players furthest apart is maybe 30 feet and even that goes unnoticed, especially by the conductor who is standing maybe 20 feet away from the players furthest away from him. In a large symphonic orchestra the distances are even greater. So really the latency of 5ms between the keypress on the digital accordion and the sound going through the amp or recording device seems insignificant to me, and considering the processing that needs to be done I'm still surprised at how fast a digital instrument does all that processing in so little time.
But as Jerry pointed out sound travels at a limited speed as well (about one foot per ms), so your 5ms delay is like placing a speaker in an analog mic-amp-speaker chain 5 feet away from you. This is really nothing. I play in a quintet sitting in a half-circle so player 1 and 5 are sitting about 12 feet apart, which is a 12ms delay and we really have the impression everything sounds exactly at the same time. In a larger orchestra (about 25 players the distance between the players furthest apart is maybe 30 feet and even that goes unnoticed, especially by the conductor who is standing maybe 20 feet away from the players furthest away from him. In a large symphonic orchestra the distances are even greater. So really the latency of 5ms between the keypress on the digital accordion and the sound going through the amp or recording device seems insignificant to me, and considering the processing that needs to be done I'm still surprised at how fast a digital instrument does all that processing in so little time.
Airy
Active member
That is my assumption as well: Geky is not hearing the 4ms delay, there must be something different in his signal chain which assures him and his friends (!) that there is a noticible (!) delay.
But science is not about trying to prove your gut-feelings but to define the experimental setup (which Geky partially did already) and then do the measurements and/or compare with results from other experiments.
BTW and totally off-topic: Some nagging about “science” by the musician (and sort of jack of all trades) Benn Jordan:
Last edited:
The latency I see when recording is from the computer software! When I plug my Rolands to other audio devices (like a Boss RC-505 looping station) there is no latency.
I’ve created multitracks using audio interface and computer many times, when playing back a song and me adding a 2nd, 3rd, etc… instrument, I hear no latency and I generally listen through the external stereo’s speakers that is playing from my computer, to my audio interface to stereo to my earbuds. I know where the latency is, but it’s so low I do not hear it. Also because of my setup, should I wish, I can listen with 0ms latency when monitoring myself, my audio interface is also an analog mixer, so I can basically cheat, which is not the case for geky.
I can EASILY insert 5ms latency just by adding even a small and inefficient VST plug-in in to a stream. This is why round trip latency can rise from 1.5ms to 5-6.5ms. Is that what is happening? Don’t. Know, my questions are irrelevant according to him.
I know he is complaining about 3 things:
- latency when playing live
- latency with MIDI
- latency differences between left and right channels when recording through his chain (what that chain is we are clueless, he’s not shared it)
All 3 are completely different with different potential solutions.
So, assuming his setup is working properly, the first one is normal, he is just showing a lack of experience playing live and hearing the normal lag we all hear when playing live and are far from the speakers, that is why live events have wedges in front of each musician so that singers/players can hear themselves. Is this the case? We don’t know, he’s not shared this info.
The second one he did not really prove it was the 4x, he was showing apparent latency while sending his sound out from the left mono channel, captured, converted to MIDI and assigned to a sample… an incredibly intensive task, delays should be expected!
Third… while I will say that it is in THEORY possible if there is something wrong with his v-accordion, he needs to find out if it is his setup or his v-accordion. He is playing in an area or system where 5-20ms common, no matter what his interface, so what is he doing? That is what I was trying to get from him, but apparently, none of my questions to him had any relevance, any he actually sounded as if he lost his temper with me… lol
I’ve since relegated him to the “if you are smarter than me, you may have the privilege of figuring it out on your own”.
I can EASILY insert 5ms latency just by adding even a small and inefficient VST plug-in in to a stream. This is why round trip latency can rise from 1.5ms to 5-6.5ms. Is that what is happening? Don’t. Know, my questions are irrelevant according to him.
I know he is complaining about 3 things:
- latency when playing live
- latency with MIDI
- latency differences between left and right channels when recording through his chain (what that chain is we are clueless, he’s not shared it)
All 3 are completely different with different potential solutions.
So, assuming his setup is working properly, the first one is normal, he is just showing a lack of experience playing live and hearing the normal lag we all hear when playing live and are far from the speakers, that is why live events have wedges in front of each musician so that singers/players can hear themselves. Is this the case? We don’t know, he’s not shared this info.
The second one he did not really prove it was the 4x, he was showing apparent latency while sending his sound out from the left mono channel, captured, converted to MIDI and assigned to a sample… an incredibly intensive task, delays should be expected!
Third… while I will say that it is in THEORY possible if there is something wrong with his v-accordion, he needs to find out if it is his setup or his v-accordion. He is playing in an area or system where 5-20ms common, no matter what his interface, so what is he doing? That is what I was trying to get from him, but apparently, none of my questions to him had any relevance, any he actually sounded as if he lost his temper with me… lol
I’ve since relegated him to the “if you are smarter than me, you may have the privilege of figuring it out on your own”.
Squeezeplay
Active member
Maybe a little “ different view “ but , I can not understand why a person will step up, purchase a product ( instrument) and then begin to pick it apart, spending more time running it down than playing and enjoying it. No it ain’t perfect….nothing is.
