Topic: ?Zero latency?

Hi there,

I like the Pianoteq technology, but the problem I have is the latency. I've read a lot of about it here, but anyway, is there any chance to get a zero latency? Not sure about it, maybe it is because of the technology. I'm trained at Conservatory and Academy of Performing arts and playing the piano from my four. I'm able to get 1.3 ms (0.7ms with rattle)  in the Pianoteq but still not satisfied with that. I can hear that very clearly the tone is after I touch the key.
Please, let me know if there is anybody who's got 0 latency. Maybe the answer is, the latency cannot be zero !?!?

Thanks a lot.

Re: ?Zero latency?

Hi Pianoshrek2

You need to lower your sample rate, that's the only way to get extremely low latencies. Lower it as much as you need to in order to get the feel you want when you play. i presume you have optimized your PC for audio? If not, check out this link to the focusrite knowledge base

http://www.focusrite.com/answerbase/

It will give you various tips to get your audio running smoothly

Re: ?Zero latency?

Sigasa: I think you meant that the sample rate should be increased in order to reduce latency - yes?  A given buffer-size worth of samples will be played faster, hence reducing/improving latency, as the sample rate is increased.

Greg.

Re: ?Zero latency?

It is interesting to see in this lecture
http://www.speech.kth.se/music/5_lectur...asure.html
that an acoustic piano has a latency in the range of 20 to 30 ms (staccato forte case), measuring time from the moment when the finger touches the key to when the strings start moving. So this is more or less what one would expect from a digital installation when summing all latencies: keyboard + software + soundcard.

Re: ?Zero latency?

There may be a way to get zero latency, but I don't know of it, and don't know of anyone that has achieved it (and I don't think it's necessary).  The best I am aware of is 5 ms (0.005 seconds).

I'm curious as to why you need zero latency - a grand piano certainly does not achieve this.  In fact the latency varies from bass (most) to treble (less), and is noticeable.

The question about latency was raised on another music forum, and the answer was as follows:

If you were playing with a group of musicians on stage and one of them was four metres (13 feet) from you, his sound would take 4.00 divided by 340 m/s = 0.012 seconds approximately to reach you.  The speed of sound in air is about 340 m/s = 1115 fps.

So his sound would take 12 milliseconds to get to you and vice versa.

Of course it's much worse with a symphony orchestra.  For an pipe organist, 12 milliseconds would be a welcome gift.  For the most part, musicians have been able to overcome this latency problem. 

Glenn

Last edited by Glenn NK (30-12-2009 22:48)
__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

Glenn, it's better you erase this segment from your comment.

I'm already seeing someone requesting to Modartt add a variable latency adjust along keyboard range.

Glenn NK wrote:

In fact the latency varies from bass (most) to treble (less), and is noticeable.

Last edited by Beto-Music (30-12-2009 23:15)

Re: ?Zero latency?

Beto:

I was referring to the latency on a grand piano - which varies across the keyboard and can't be changed, and won't likely ever be changed as long as the bass hammers have more mass than the treble hammers.  And this cannot be changed - heavier hammers are required to excite the bass strings.

Glenn

__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

No when we're back in latency issues, let me ask one thing:

How much does the MIDI itself cause latency? I remember that I have felt some latency when putting my hardware piano's local control off and then playing it back to same machine trough computer / DAW. So if this is true, even if software like PTQ could be zero latency still there is midi based latency. If it exists how much is that? And if we speak about software-based latencies (metering it with ms) are we forgetting this MIDI-latency...?

P.S: Right now I'm using PTQ standalone with 96KHz sample rate and 192 buffer size. This gives me 2.0 ms (or what it really is?) latency which feels quite ok.

Last edited by Ecaroh (31-12-2009 00:11)

Re: ?Zero latency?

I must remamber we all have also a bio latency in our organic bodies. It' know that our nervous system have speed near a Indy car, 300km/h, to receive and send a nervous impulse.

If you was a brontosaurus (27 meters long), maybe this bio-latency would anoy you about move the point of the tail.

;-)

Re: ?Zero latency?

Hi All,

thank you for your comments. I'm not a technical based guy, but everything was done as Sigasa wrote. The best latency I can get in Pianoteq is 0.7 ms with some crackles (64 samples) and 1.3 ms with nice sound (128 samples) and still can hear/feel the difference compared the grand piano which is next to me - (for example 5 ms is really huge difference for me). But is this real latency from key to the speakers or headphones?
BTW I've got absolute pitch (which is not about the latency :-)
What I only feel the sound is later comparing my acoustic piano. I can immediately identify when I'm playing the Pianoteq and not the internal sound from the original instrument or the grand one.
When I play the instrument and PianoTeq together, it is very very difficult to hear there is some difference in the sound (maybe that's because of the sounds are merged) but in discant part of the piano I can hear the difference very well!!! Try that if you can hear that also :-)
Anyway if I play the Pianoteq individually the difference is there.
My piano fellow (he is also trained at Academy of Performing Arts) can identify the difference immediately also.

to guillaume: I'm not familiar with the physics also :-) but anyway, is that really only about when the finger touches the key to when the strings starts moving ? What we exactly feel is the hit to the string and not only starting touch the key. So what we are wating for, is the hit and this is not only about to start to touch/hit the key (of course with the playing). Maybe I should write that I'm able to feel when the hammer hit the string (and it doesn't matter if this is because of real physical contact or because of starting vibration).

to Glenm NK: it is nice to know these calculations but when you play with a group on stage you never wait for the sound from other play-mates. Never :-) It is about same feeling of the music and rhythm also (sometimes it is about visual sense). If you've got good and same feeling the sound is all together.....Of course the symhopny orchestra is also playig with the feeling and that's why a conductor is there. If you are in the middle of the orchestra, sometimes the sounds is quite strange, because of latency.....I know that, I feel that :-)

Anyway my colleague and me can feel the latency and I really don't know how to fix it, cause I'm big fan of Pianoteq. At this time I cannot imagine I'm able to create a professional recording with this issue ;-(

But anyway thank you!!!

Last edited by pianoshrek2 (31-12-2009 00:56)

Re: ?Zero latency?

pianoshrek2 wrote:

...The best latency I can get in Pianoteq is 0.7 ms with some crackles (64 samples) and 1.3 ms with nice sound (128 samples) and still can hear/feel the difference compared the grand piano which is next to me - (for example 5 ms is really huge difference for me). But is this real latency from key to the speakers or headphones?...

"Real" latency could only be measured using a separate recording device that could record the output of Pianoteq or your acoustic piano and compare that to an acoustic sound made by the key at the bottom of its throw by examining the waveform in an audio editor.  There is no such thing as zero latency* only perceived zero latency.  The latency numbers that appear in either Pianoteq or your audio card's interface do not represent the actual time from finger to ear -they are a calculated guestimation.  As far as detecting anything less than 5ms of "actual" latency -I sincerely doubt that humans have that capability.  There are a few scattered experiments on the web, but I admit I haven't found a definitive study.  But I have a feeling that it is your keyboard controller where you are "feeling" the difference.  Maybe with a different velocity response, either from your keyboard settings or from within Pianoteq, you may not have the same perceived difficulty.  There are plenty of accomplished pianists who use Pianoteq and other MIDI controlled instruments without difficulty.  That said, there is no substitute for the acoustic instrument and you may never be satisfied with even highly reasonable facsimiles.
*quantum mechanics theory might hold an exception.

Aside:  There is a controller -possibly discontinued- for marimba/percussionists called the Marimba Lumina which uses proximity sensors in the heads of mallets to trigger the MIDI response prior to actual contactor with the striking "pad" or surface.  This recreates the "immediate" response that percussionists are accustomed to.  Possibly a piano controller (a piano is a percussion instrument, after all) could incorporate a similar scheme though I'm not sure how it would be different from simply altering velocity settings.

Last edited by Cellomangler (31-12-2009 02:31)
"Downing a fifth results in diminished capacity."

Re: ?Zero latency?

Pianoshrek2: If you can, try Pianoteq on a totally different system, and preferably one on which it is already functioning to the owner's satisfaction. 

Another thing to try might be to actually record your problematic system with a microphone, so we can measure the total latency. Play a single note using something hard (such as the nail of your finger), so that it makes a distinct "click" that will come through on the recording. We will then be able to easily measure the total latency - we can measure the time from the "click" to the beginning of the actual piano sound from Pianoteq.  I've never done this but it should work.

It is possible that your total latency is still too high, despite the fact that your audio driver latency is very low.
(yes, one reason for this could be MIDI latency, in theory)

Greg.

Re: ?Zero latency?

skip wrote:

Pianoshrek2: If you can, try Pianoteq on a totally different system, and preferably one on which it is already functioning to the owner's satisfaction.  Greg.

I agree. It could be a very slight problem with your midi controller's response or something else.

Three other things to check:

1. On my keyboard, which is a rompler, no matter what I do, I'm playing through a preset on the keyboard. For the vast majority of parameters, the preset's parameters are ignored--the amp envelope settings are overridden, etc. No problem there.

But, if I accidentally make a change to some of the parameters in the midi keyboard's instrument preset, or change the preset, that may get reflected in the response of any vsti, including PianoTeq. (Please don't ask which parameters or what keyboard preset. I haven't had time to track down all of the variables. I just have to be sure to stay with one preset and not move any knobs by accident, since I have cc's routed to Pteq.)

For a test, you might try several presets on your midi keyboard and see if there is a slight shift in the latency even when over-riding the preset to play PianoTeq. (It's possible, too, that your keyboard has some kind of undocumented "piano mode" or "synth mode" that it goes into on various presets, without telling you that it will affect the velocity response of the keyboard when used as a midi controller.)

2. Check your keyboard's global and instrument settings. Could be something there with the velocity response. A slow ramp? Or just try increasing the default attack time or lowest velocity setting a little to hear what happens.)

3. And recheck your midi cc's. Is there a chance that you've assigned one to a cc that's shared by PTeq, and have it set slightly off?

Sorry if I'm telling you what you already know.

Cheers.

Last edited by Jake Johnson (31-12-2009 02:54)

Re: ?Zero latency?

I've just done the mic test, and unfortunately it fails dismally on my system - 70ms, which appears to be roughly twice that of a real piano.  This is using the lowest buffer size possible with my hardware (128 samples), and a sample rate of 44.1kHz. I am using the M-Audio Fast Track Ultra for both audio and MIDI. (this is a USB interface)  The audio recording was done using a completely seperate audio interface.

That said, I think the response is fantastic, and I can't tell the difference between it and my hardware digital piano. I am however a very basic player. (I never get to play real pianos these days either)

EDIT: I tried it again, playing the note more forcefully, to reduce the key delay, but this didn't help much - 65ms.

Greg.

Last edited by skip (31-12-2009 02:54)

Re: ?Zero latency?

I repeated the test with my Kawai MP9000 digital piano, and it is significantly better - 30ms. This matches a real piano nicely.

I then tried a sample library in Kontakt, and the result was roughly the same as Pianoteq. (maybe a few ms faster but I'd have to do a lot of repetitions to see whether it was consistent. Not enough to worry about in any case)

I am now throwing all my computer hardware in the bin.

Greg.
p.s Not really.

Re: ?Zero latency?

If only Digitech, Lexicon, -one of the usual suspects- would just make a Digital Forwarder... (opposite of a delay).

"Downing a fifth results in diminished capacity."

Re: ?Zero latency?

If Greg's measurement is correct, we are dealing with real latency problem here. So experienced latency is sum of at least these:
Keyboard mechanics and processing, MIDI-conversion and transfer, software(s), audio interface, speaker system and its distance to ear.

To live-music player like myself this latency is maybe the most critical thing which is against using software-pianos. In fact I've experienced that "same" latency feels more problematic on stage. I don't know if it's the overall sound systems which is somehow increasing delayed piano sound (I'm using near personal monitors) or am I just more critical on that moment. Maybe at home I've "adjusted myself" to play with some latency (somebody on this forum say that they handle 15ms software-latency. This need practice or should we say brainwash??)

Anyway if latency is something like twice than real piano then it's not surprise that many acoustic players don't accept it. No matter how good the sound, it just doesn't feel right. And I'm not sure that it's a good thing to develop this new way of playing and hearing (but I'm not sure that it is a bad thing either). Shouldn't we instead try to develop a digital model which is close to real thing also in latency....? Easier to say than to do of course.

But why there's no more complaint about latency? I think the reason is simply that people using softwares are used to deal with it and many doesn't even need this immediacy. Those producer-type-guys need good sounds, they don't need good feeling of playing. For a live-player playing feeling is the most important and then comes sound (sound is most of the time terrible anyway because of the sound systems and bad acoustics). To those guys string resonances and pedal noises etc. are simply irrelevant.

P.S: Let me give you one another example of latency. In my opinion Roland Finland f***ed up their V-piano demo tour with latency. I was testing the V-piano first time and I felt that there was latency. For many classical players it just felt "strange". Afterwards I heard that Roland was using some Genelec DSP-system which was causing latency (at least some ms). Their monitors were also not very close to player. So these milliseconds matter and they have also business value if you wanna new customers! From those who were there at the Roland show none have bought V-piano.

Last edited by Ecaroh (31-12-2009 11:46)

Re: ?Zero latency?

It'd be good if someone with a more traditional setup (not using USB) could do the latency test.  Btw, a mobile phone might be good enough to do the recording - sit the phone near the key that you play, to pick up the click.
(assuming the phone has voice recording - many do now)
Then transfer the recording to a computer for analysis.

I might repeat the test on my old desktop machine, in any case.

Greg.

Re: ?Zero latency?

Hi Pianoshrek2

you could try enabling the delay compensation and/or level compensation and see whether this helps you to achieve the desired response while you play? You will find these switches in the 'output' section to the left of the mic panel.

Last edited by sigasa (31-12-2009 11:31)

Re: ?Zero latency?

It must certainly be a difficult life for a musician who can be disturbed by a latency of 1 ms. As pointed out by Glenn, sound travels 30 cm in that time. That means you will really have to compensate for the difference in distance between your ears and the different registers of your piano. Oh, yes, and do not move your head too much; it would spoil everything.

If I calculate correctly and Beto-Music's Indy car speed for the nerves is correct, it would take in access of 10 ms for vibrations or to reach your brain from your finger tips, excluding 'processing' time. Very complex to use that information again to evaluate such small latency differences.
Human perception is wonderfully complex, but really 1 ms is very very short in acoustics. The lowest A on your piano needs 40 times that period to complete a single wave.

Re: ?Zero latency?

pz wrote:

It must certainly be a difficult life for a musician who can be disturbed by a latency of 1 ms.

We still don't know pianoshrek's total latency - all he has reported is the sound driver latency. As I have shown in my test, my total latency is very large - 70ms, despite a pretty small ASIO buffer size of 128 samples. (just 3ms @44.1kHz, which is a tiny fraction of the total latency)

Btw my speakers were about 1m away, adding about 3ms to the latency.

Just for kicks, I measured an ancient Yamaha PSR-230 portable keyboard: 10ms. 

I'm curious to know whether the Kawai digital piano was intentionally designed to have a latency the same as a grand piano!

Greg.

Re: ?Zero latency?

skip wrote:

We still don't know pianoshrek's total latency - all he has reported is the sound driver latency. As I have shown in my test, my total latency is very large - 70ms, despite a pretty small ASIO buffer size of 128 samples. (just 3ms @44.1kHz, which is a tiny fraction of the total latency)

Greg.

Greg, your measurement shows very convincingly that the 'Pianoteq latency' is insignificant compared to the controller latency (including MIDI effects). And so it should be: a midi controller used for playing piano should never react as soon as the first contact with the key, this would result in very poor control of the tone.

Obvious to e.g. an oboe or timpani player, but maybe not to all pianists, what you 'time' as a player is not the moment you start your movement (start to touch the key), but the timing of the actual sound itself. This should not be as immediate as possible, but just has to be more or less realistic. Would be very interesting if you yourself could do the same measurement on a real piano (grand preferably, the mechanics of an upright are different), as I suspect the result is really tied to individual playing technique.

This latency number for computer/audio latency was very useful in the days when it was 200 ms, but now appears to, in my opinion, makes people worrying about the wrong things. The main point I had is that some people do not seem to realize how short 1 ms really is; perhaps we should go back to writing it as 0.001 second.

Re: ?Zero latency?

pz wrote:

Obvious to e.g. an oboe or timpani player, but maybe not to all pianists, what you 'time' as a player is not the moment you start your movement (start to touch the key), but the timing of the actual sound itself. This should not be as immediate as possible, but just has to be more or less realistic.

May I refer you to Guillaume's post a bit earlier in this thread, where he referred to a study which tested the latency of a grand piano: the result was that the time between first contact with the key, to when the string starts to vibrate, is about 30ms.  I haven't read the article yet, but it would be reasonable to assume that for testing purposes, they struck the key rapidly, because otherwise the key-press duration would dominate the result, and it would not be very useful. The test I am doing should hopefully be a reasonably accurate reproduction of the same test, because I am measuring the time between the click of my fingernail hitting the key, to the first sound.  I am doing a *specific* test, to get a feel for the overall latency.

I have repeated the test with my desktop PC, using a PCI audio interface (M-Audo Delta 66, set to the same buffer size of 128), and an integrated MIDI port, and the result is better - about 40ms.  This is what I feared - I think the USB interface that I am using with my laptop is introducing an extra delay - more than 20ms extra. This is quite a large penalty! I will see if there is a way that I can reduce the latency using the USB interface. Unfortunately the "high performance" mode of the Fast Track Ultra doesn't work on my system - the audio is choppy.

Would be very interesting if you yourself could do the same measurement on a real piano (grand preferably, the mechanics of an upright are different)

I will probably do this, to see whether the result matches the aforementioned study. It will be very interesting.

I am interested in latency tests from other users. (particularly the base poster - Pianoshrek)

Greg.

Last edited by skip (31-12-2009 13:56)

Re: ?Zero latency?

With the Fast Track Ultra connected to the desktop, I have measured about the same latency. If I then keep the MIDI on the FTU, but use the audio of the PCI card, the latency reverts back to about 40ms.
So, it appears to be the FTU's audio that is introducing the extra latency.  The high performance mode of the FTU actually works on the desktop machine, but it doesn't make any measurable difference. It does allow the buffer size to be reduced to 64, but that only improves the latency by 1.5ms. 

Greg.

Re: ?Zero latency?

pz wrote:

This latency number for computer/audio latency was very useful in the days when it was 200 ms, but now appears to, in my opinion, makes people worrying about the wrong things. The main point I had is that some people do not seem to realize how short 1 ms really is; perhaps we should go back to writing it as 0.001 second.

Hello All,

Regarding latency issues, I completely concur with pz's observation that people might be worrying about the wrong things.  Are you aware that 20ms is one-fiftieth of a second and that a "very large" 70ms is approximately one-fourteenth of a second?  From some of the rather dire concerns I am reading in this thread, it sounds as though you believe the latency is on the order of half a second or more.

Here is my take on latency of a real grand piano:

Although the literature states the response time of a real piano is on the order of, say 30 to 40ms from the onset of depressing a key, I believe the latency mentioned in this thread is defined as the timing delay one experiences after the midi keyboard has bottomed out and the sound is heard. 

When one plays a real grand with the intention of creating a pianissimo sound, the player is pressing the key down so "slowly" that the hammer strikes the string BEFORE the key bottoms out in the keybed.  In other words, the hammer shank is in free flight after the escapement and has already contacted the strings while the front of the key is still moving downward by the player's finger. Typically the hammer contacts the string approximately 10ms (1/100th of a second) before the key front bottoms out in the keybed.

When one attempts to play fortissimo, the hammer of a real grand piano typically contacts the string AFTER the key front has bottomed out in the keybed, because the high force of the action upon the hammer shank has caused the shank to flex (e.g., the hammer itself remains stationary by its own inertia for a minute fraction of a second AFTER the hammer shank has started to move).  In addition, bass note hammers have more inertial mass than upper octave hammers.  Typically, this difference in timing -- delay, if you will -- is on the order of about 10ms difference (one-one hundredth of a second).

Somewhere between pp (where a real piano string is contacted before the key bottoms out) and ff (where a real piano string is contacted after the key bottoms out), there must be a loudness level (say, mezzo forte or forte) where the string is contacted at precisely the time the key bottoms out in the keybed.

* * * * * * * * *

The above discussion was how a real grand piano works.  Note that it is different when one plays a midi keyboard, where no signal will be initiated until the key is at or very close to contacting the keybed.

Now, since a midi keyboard generally sends a note-on midi signal when the key bottoms out (or very nearly so), the "clock" for delay really doesn't start until about 20ms after the key is first pressed.

Let's now go back and review the difference between a real grand piano and a midi keyboard:  You will recall for a note played pianissimo on a real grand, the "latency" between bottoming out the key and hearing the sound might be "negative 10ms", because the hammer contacts the string before the key hits the keybed.  Well, clearly the sound cannot precede the midi keyboard bottoming out, because the nature of an electric keyboard mandates that a switch must be closed in order for a note-on midi message can even be sent.

While I am aware of the so-called latency of X number of buffered samples, say 64, 128, 256, 512, 1024 samples or more in a midi system, we must remember that Pianoteq also requires some time to "compute" the sound in real time.  This is dependent upon the cpu speed as well as the speed of your computer's busses.  There has been some confusion in terms regarding sample rate and samples to be buffered:  Here, we are talking about breathing room for your computer to calculate the sound before sending it on.  By increasing the buffer size, the cpu is given a slight amount of breathing room, but it comes at the expense of taking more time to make the calculations.

Larger sample buffer size = less pops and clicks, but longer latency.
Smaller buffer size = less latency, but at the expense of working your cpu harder.

Once the sound gets calculated, it must also be panned, sent through reverb, and ultimately to the audio amplifier of the sound system.  There are amplifier rise times associated with components of hardware associated with an audio system.

Hopefully you have read this far -- as you know, I tend to become long-winded, especially while typing at nearly 4AM local time on New Year's Day. 

Here is what I believe this latency issue boils down to:
An electronically midi enabled keyboard physically cannot match the latency of a real grand piano --- if one wants to measure the difference in units on the order of milliseconds -- thousandths of a second.  Hopefully the amount of latency / delay you experience is tolerable in the music you perform.  Everything in the electronic chain of depressing a note on a midi keyboard, the midi protocol itself, whether you use a USB connection versus midi in and out, the speed of your computer's processor, the number of processors in your computer, your computer's bus speed, hard drive rotational speed, hard drive data transfer speed, etc, etc, etc, WILL contribute to an amount of delay from the time your finger bottoms out on your midi keyboard and the time you first hear a given sound.  Please be aware that many (most?) laptop computers are equipped with 5400RPM hard drives, when drives ranging from 7200, 10000 and even  15000 rpm's are commercially available, but probably not for laptops.

This response is rambling, especially at nearly 4AM local time on New Years Day 2010.  Your questions and comments are welcomed.

Sincerely,

Joe

P.S.  Happy New Year 2010 to all

Last edited by jcfelice88keys (01-01-2010 19:05)

Re: ?Zero latency?

Joe,
Just a nit, but I am not measuring the bottom-out to sound latency - I am measuring initial key contact with my finger to sound latency, which is also the specification that Guillaume mentioned. I understand & agree that my MIDI controller does not detect the key-press until it has nearly bottomed out though.

Yes, I admitted that I felt that my response was very good before I did any measurements - I personally do not have any problems at all. For me, this has degraded into being interested in determining whether the extra 20ms or so of latency that my USB interface has, compared with a PCI soundcard, is a) normal, and b) able to be improved.   

The original person with the problem (Pianoshrek2) still does not know his total latency (or has not reported it yet). It MAY be excessive.

Also,  I thought that perhaps a 65ms total latency might be enough to cause problems for an accomplished pianist, for fast passages, who is also not used to using a MIDI keyboard.  To me, 65ms seems like a big jump from 30ms - it is like playing the piano from about 10 metres away.  (notwithstanding your comments about the differences between a piano and a MIDI keyboard)

Thanks for your thorough post.

Pianoshrek2: do you have any problems playing a hardware digital piano, such as a Clavinova etc?

Greg.

Last edited by skip (01-01-2010 12:02)

Re: ?Zero latency?

Joe, I just looked on a Yamaha digital keyboard, the depth is 12 mm and the midi on signal is sent when the key is depressed by 9.5 mm, so at 4/5 of its trajectory. If for example it takes 10 ms to depress the key, the midi on signal is sent 8 ms after touching the key and 2 ms before reaching the keybed (supposing a constant key velocity). This would be quite a forte stroke. Playing pianissimo, it could take 100 ms to depress the key, midi on signal 80 ms after touching the key and 20 ms before reaching the keybed. So Yamaha's choice seems to be quite good according to your analysis of what happens on a real grand piano.

Happy New Year to all!

Re: ?Zero latency?

When I play a simple scale as fast as I can, the note duration is about 60 to 70ms. This means that the extra 20ms of delay caused by my USB interface is just under a third of this note duration, which is a significant percentage. (and I'm a hefalump on the keyboard)

Just btw, I also tried using MIDI on the desktop's integrated MPU401 MIDI port, with just the audio going out the Fast Track Ultra - I got the same result - a 60ms total latency. I have started a thread over on the M-Audio forums for this issue: http://forums.m-audio.com/showthread.php?t=15883

I can notice the difference between the PCI card and the USB when playing. It is subtle, but there. (again, I don't think the USB's delay would be a problem for me though, at my level)

Greg.

Last edited by skip (01-01-2010 13:29)

Re: ?Zero latency?

I have a feeling (of course this should be tested) that in Roland RD700GX there is more latency than before and it depends on sounds. I feel that there's more latency in Supernatural E-pianos which are using some kind of modeling. Also if I play with my Roland some audio while I'm practicing this seems to increase the latency. And if the audio file is Mp3 (not Wav) latency is highest. Many players have not liked this RD700GX and maybe it's because of the latency - they may not know what's wrong with it but they can feel it.

One more thing with MIDI-latency. Am I right that MIDI-latency increases if you play chords (=press many keys simultaneously)? This is also something I have felt with my RD700GX. Take some modeled E-piano with effects and play fast things with for example 6-voices chords. Sometimes I have even wondered that there's something broken in my RD. Maybe it's the software+midi latency of this hardware piano...?

Re: ?Zero latency?

to Ecaroh: you might certainly be right. A few years ago (well OK, not so few!), I was working as a journalist for a review called "Meet Music Magazine" here in Belgium, and I was testing all kind of electronic stuff. Included a "new" Roland master keyboard (A90): it had approximately four times the latency of all my other synths at the time! It was even much slower than their older model (A80)! Eventually, Roland did a mod to correct this. Maybe they'll do it again for the RD700GX?

Regarding MIDI latency, I don't think MIDI is the source: it would more probably be the extra maths required to produce the sound...
For example, it happens also with my Alesis Fusion but much less with an (older) Alesis QS8... (with a less sophisticated synth engine)

Last edited by Luc Henrion (01-01-2010 15:51)

Re: ?Zero latency?

Ecaroh wrote:

One more thing with MIDI-latency. Am I right that MIDI-latency increases if you play chords (=press many keys simultaneously)? This is also something I have felt with my RD700GX. Take some modeled E-piano with effects and play fast things with for example 6-voices chords. Sometimes I have even wondered that there's something broken in my RD. Maybe it's the software+midi latency of this hardware piano...?

Hello Ecaroh,

The MIDI protocol (established in 1982 by Roland, Yamaha and Korg) sends all of its data "in series" rather than in parallel.  This means that notes of a chord must wait in line to be sent one-at-a-time from the sending device to the receiving device.  In addition, MIDI was specified to be only in one direction at a time -- this is why one sees the familiar MIDI IN and MIDI OUT ports, especially in pre-USB equipped hardware.

At the time the MIDI protocol was established 28 years ago, the big three companies designed MIDI primarily so one company's hardware device could play another company's hardware device in essentially real time.  Personal computers were in their infancy compared to today's capabilities, and there were few if any sequencing software applications at the time.

My own industrial experience of 1982 recalls desktop computers having no hard drives at the time -- one had to insert a 5-1/4" floppy disk into the A: drive for a given software application; a second 5-1/4" floppy was set into the B: drive to save your word processing data.  If anyone cares, this is why the hard drive of a Windows based PC platform is called the C Drive.

But I digress --

The upshot of the MIDI protocol is latency when lots of information needs to be sent simultaneously.  Although I deleted my Variations on Jingle Bells mp3 from the Pianoteq Files section, the 16 track MIDI performance did contain many places where the notes of chords would not sound together at the same time.  There was so much midi clogging that note clusters had sounded late in relation to the beat.  Even when I resorted to nudging the offending notes forward in time, the effect caused other notes to chime in too late.

Joe

Last edited by jcfelice88keys (01-01-2010 18:50)

Re: ?Zero latency?

I have intermittent arpeggiation when I play internal layered sounds on my Kawai MP9000.  (this is just two sounds together - nothing complicated)

I had worse arpeggiation on an older product - the Kurzweil 1000PX sound module.   At the time, I remember it being generally accepted that the MIDI protocol was not the problem - it was indeed the lack of processing power. (and I also suspect processing is the cause of the MP9000's arpeggiatioin - not MIDI).

I'm a bit surprised, and disappointed, that a product like the RD700GX appears to suffer from this kind of problem as well, in this day & age.

Greg.

Last edited by skip (01-01-2010 19:36)

Re: ?Zero latency?

Very interesting discussion; what really intrigued me was Joe Felice's second post about how midi is sent.  I'd never thought about how it was transmitted.  Then Skip's (Greg's) post got me thinking - I had never experienced arpeggiation (at least I haven't recognized it).

I tried a controlled test to see if Pianoteq suffered from this:

I created a midi file (in a sequencer) that uses 18 simultaneously scored notes ranging from the lowest to the highest C.  Being sequenced rather than played they have identical velocities and durations.  Two chords were used, the first has note lengths of 24 ticks, the second otherwise identical chord lasts 384 ticks.  My sequencer is set up at 120 bpm, with 480 ticks per 4/4 measure (if anyone wishes they can do the math to determine the length of each chord).  No pedal control (damper) is used.

The midi sequence was loaded into Pianoteq, played and recorded with my wave editor (could have used Pianoteq to generate the wave file, but may try this later).

Two "pianos" were used, the sound of the first two chords is from the Erard with two mics set one above and one below the approximate centre of the soundboard.  The third and fourth sounds use Jake's Johnson's Steinway D with mics in the same location.

From my ears it appears that the 18 notes are sounded simultaneously, which suggests that Pianoteq isn't the source of arpeggiation noted by Skip/Greg.

I have uploaded both the midi file and the mp3 wave file.

More and more, I'm concluding that the controller is a very important link in the chain when using Pianoteq.

Comments as always are welcomed (I have thick skin in the pursuit of scientific truth).

Glenn

__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

As I said, these latency problems with RD700GX are not tested yet. But I remember very well that it was kind of first impression when I got it and started to play. Before RD700GX I had RD700SX and before that RD600. After a year of using GX as my main keyboard I've used to play with it and this "latency" (if it exists) is not that much disturbing me anymore. But this supports one of my argument: people can adapt themselves to play with latency and after this adaptation it feels "normal".

I admit that process can be also vica versa: being aware of latency can lead to (over)sensitivity of this dimension of playing. Instead of guessing maybe I should start to measure this latency of my machines. Greg: what software and method do you use for measuring this "time between the click of my fingernail hitting the key, to the first sound"? I can quite easily record this but I'm not sure how to proceed from that.

Re: ?Zero latency?

Glenn NK wrote:

From my ears it appears that the 18 notes are sounded simultaneously, which suggests that Pianoteq isn't the source of arpeggiation noted by Skip/Greg.

I think there is a misunderstanding here - I am not saying Pianoteq is suffering from any arpeggiation whatsoever. The arpeggiation I have witnessed is on the internal sounds of my Kawai MP9000 digital piano, and a Kurzweil 1000PX hardware MIDI sound module. Pianoteq is working great.

Glenn: It would be interesting if you repeated the test, but instead, play the MIDI file to Pianoteq across a MIDI cable.

Note that MIDI isn't really all that slow, despite the fact that it is transmitted serially. To play all 88 notes of a piano simultaneously, it would take 84ms to transmit the MIDI data across the cable. Yes - that's enough delay to cause audible arpeggiation, I think, but it's not drastic. And this is a rather extreme example.

Note that I am not saying that Joe was definitely not suffering from MIDI related problems in his Jingle Bells sequence.

Ecaroh:  To test single note latency, I am recording to a soundfile, and then inspecting the soundfile in a soundfile editor. (I am using Adobe Audition).  I then zoom in, and select the audio with the mouse, starting at the "click" of my fingernail, and ending at the point where the piano sound begins. My software shows me the duration of the audio I have selected very precisely. Thinking about everything I am doing in this process, there is some skill involved. Once you have made the recording, it may be best to just upload it so we can inspect it as well.  There is a very real risk of confusing the sound of one's finger striking the key with the piano sound, for example.

I am not testing arpeggiation yet - this is more complicated. 

Greg.

Re: ?Zero latency?

Greg:

Sorry, I didn't misunderstand - what I should have said is that it appears that we can eliminate Pianoteq as the source of the arpeggiation.

Another aspect in latency is the amount of information that is transmitted via midi; more information takes more time.

I downloaded the "Jazz piano for M3 . . . " midi file that Jake Johnson posted and was quite surprised at the number of midi events it contained (7,561).  By far, most of them were Pitch Bends (I'd estimate at least 75 percent).  By contrast, a midi file of Rhapsody In Blue has just under 13,000 midi events, and runs 16 minutes.

Wouldn't more midi events slow down the interpretation of midi information?

Pianoteq can interpret Pitch Bends if it's activated, but wouldn't it still have to "sort through" them anyway?  If so, this would be another bottleneck.

It may be possible for me to save the file to disk, and insert it in the piano for it to play back to Pianoteq (which could be recorded with my wave editor).  Will try this, but guests are arriving shortly, so it will have to wait.

At this point, it seems to me that the OP's problem isn't with Pianoteq, but the keyboard/controller.

And of course Joe Felice's  remarks about variable latency in a piano due to velocity and note pitch certainly must be considered.  I wonder how long it will take to develop a keyboard that reacts just like the real thing?  I just remembered, it's called a Yamaha Diskclavier.

Glenn

__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

(Off-topic, to some extent: I was surprised by all of the pitch bends. Apparently that's the way the program deals with pitch changes in the Detune pane. Doesn't save them, exactly, except as midi messages that it makes on the fly. Not sure how it handles chords--it can only do one pitch bend at a time, and the notes change pitch to different degrees. Sorry to interrupt.)

Last edited by Jake Johnson (02-01-2010 02:00)

Re: ?Zero latency?

Jake Johnson wrote:

(Off-topic, to some extent: I was surprised by all of the pitch bends. Apparently that's the way the program deals with pitch changes in the Detune pane. Doesn't save them, exactly, except as midi messages that it makes on the fly. Not sure how it handles chords--it can only do one pitch bend at a time, and the notes change pitch to different degrees. Sorry to interrupt.)

Jake - not interrupting at all.  I'm confused, but when I get confused, I have to think.  Not a bad thing.

When you say "that's the way the program deals with pitch changes in Detune pane, etc", what do you mean?

I thought your piano had generated the midi file, you saved it, and posted it.

Does this normally happen when you save a midi file from the keyboard (for example if not using Pianoteq)?


Skip/Greg:  I'll have to do some re-wiring in order to do the test you suggested.  The audio outs of the piano don't transmit the Pianoteq audio - at present only the headphone jack outputs out the Pianoteq sound.

Glenn

Last edited by Glenn NK (02-01-2010 03:02)
__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

Glenn NK wrote:

Sorry, I didn't misunderstand - what I should have said is that it appears that we can eliminate Pianoteq as the source of the arpeggiation.

The O.P did not complain of arpeggiation, though. (not explicitly, anyway)  He complained of latency, which can be totally unrelated to arpeggiation. For example, if you increase your ASIO buffer size by a large amount, causing noticable latency, this will not cause any arpeggiation whatsoever.  Now, if Pianoteq was causing appreciable delays due to it's internal processing, then that could also cause arpeggiation - I agree. 

It may be possible for me to save the file to disk, and insert it in the piano for it to play back to Pianoteq (which could be recorded with my wave editor).  Will try this, but guests are arriving shortly, so it will have to wait.

Just to make sure you understand what I mean - I mean that the MIDI file should be sent to Pianoteq over a MIDI connection. (and not a USB connection - a real, old fashioned MIDI connection). This is to see whether the relatively slow transmission of the MIDI data (circa 1ms per note on) is enough to be audible in your test.

At this point, it seems to me that the OP's problem isn't with Pianoteq, but the keyboard/controller.

I don't completely agree with this. It may also be due to delays caused by the software on the PC - especially the drivers. Refer my test results which show that I have a total latency on my main system that is about 60 to 65ms, for forte playing, which appears to be double that of a real grand piano. Moving to a different system, but retaining my controller, this dropped to 40ms.    I agree that we don't have any reason to suspect Pianoteq at this point though. Anyway, if the O.P also has this kind of latency (60ms), this may be enough to cause problems for him, given that he is probably fairly accomplished. I am very keen to know whether he has any problems playing a normal digital piano. (as I asked earlier)  Note that Steely Dan musicians purportedly could detect differences even for a 5ms increase in latency:
http://www.soundonsound.com/sos/jan05/a...sician.htm

And of course Joe Felice's  remarks about variable latency in a piano due to velocity and note pitch certainly must be considered.  I wonder how long it will take to develop a keyboard that reacts just like the real thing?  I just remembered, it's called a Yamaha Diskclavier.

Their Natural Wood action also takes the variable delay into account, and this appears to be something which differentiates this action wrt their GH3 action:
http://music.yamaha.com/products/highli...AST/8.html

Greg.

Last edited by skip (02-01-2010 03:48)

Re: ?Zero latency?

Apparently I'm confused. There are of course many pitch changes in the Detune pane in that fxp. If I open Options\Midi and use the preset to play the midi file, I see a sequence of Pitch Wheel messages with only a few note-on and note-off messages.

So my assumption, when I just now posted, was that PTeq was making those Detune changes as it played the file. However, I just found that if I use a factory instrument, with no pitch changes in the Detune pane, to play the midi file, the same pitch wheel messages occur. And if I do make changes in the Detune pane, record a quick midi file, and then play it back, I get normal note-on and note-off messages. So I don't know what's going on with the midi file. (I didn't record the midi file myself--I don't play that well. Not sure where I found it--one of too many midi files on drive.) My only guess is that the creator set the keyboard to transpose to a different key, and that got recorded into the midi file as note by note pitch changes. (?)

Last edited by Jake Johnson (02-01-2010 03:33)

Re: ?Zero latency?

Gentlemen,

I must step back and congratulate all of the participants in this thread for your combined intelligence and curiosity.  Regardless of one's personal experience with latency, and regardless of the final outcome of this thread, it is wonderful that we are collectively exploring the finer aspects of real pianos' actions, the MIDI protocol, timing aspects of multi-tone clusters, etc.

None of this type of conversation would have been probable had it not been for the modeling aspects provided by Modartt S.A.S.  True, latency is discussed in other types of forums for samplers and synths, but they do not address it at such a basic level of understanding that is going on here.

Cheers and Happy 2010 to all,

Joe

Re: ?Zero latency?

Jake Johnson wrote:

Apparently I'm confused. There are of course many pitch changes in the Detune pane in that fxp. If I open Options\Midi and use the preset to play the midi file, I see a sequence of Pitch Wheel messages with only a few note-on and note-off messages.

So my assumption, when I just now posted, was that PTeq was making those Detune changes as it played the file. However, I just found that if I use a factory instrument, with no pitch changes in the Detune pane, to play the midi file, the same pitch wheel messages occur. And if I do make changes in the Detune pane, record a quick midi file, and then play it back, I get normal note-on and note-off messages. So I don't know what's going on with the midi file. (I didn't record the midi file myself--I don't play that well. Not sure where I found it--one of too many midi files on drive.) My only guess is that the creator set the keyboard to transpose to a different key, and that got recorded into the midi file as note by note pitch changes. (?)

I have many times transposed midi files in my sequencer program;

http://www.pgmusic.com/powertracks.htm

and have never had any pitch bend commands result (although I have manually inserted them in a file for flutes and saxes).  Whoever recorded the file used a keyboard with some sort of touch control in the keys - something I am not familiar with.  Looking at the score, it's hard to imagine anyone inserting them with a wheel while using two hands on the keys.

Digressing further, I have used only the program noted above, but the number of pitch bends was amazing in the file you posted.  I intend to clean out the PBs and see how many useful commands there are.  My piano and I generate truckloads of pedal control signals, but nothing compared to the number of pitch bends in that file.


Skip/Greg:  I understand about the midi transmission, but in any event I couldn't use USB because my soundcard is PCI.

As for testing through MIDI cables, I can play the keyboard (sending MIDI information to Pianoteq), but can't put the floppy in the piano and have it send the MIDI information to Pianoteq.  Have to think more about this and see if I can find a solution.

With the native ASIO drivers of my soundcard and the latency set to 7.1 ms (buffer 312), I have no latency problem, but at 10 ms or higher I notice it, although I could adapt to it.  Perhaps I've been playing a DP too long and have adjusted to its response.


Joe:  Your post about the variable latency in acoustic grands caused by the hardness of the strike enlightened me.  As soon as I read it, there was an "aha" moment.  Thanks.  Methinks this business is more complex than had seemed at first.

Glenn

Last edited by Glenn NK (02-01-2010 08:58)
__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

Glenn,
Yes, after I made my post, I realised that you were probably going to play the MIDI file from your piano - thanks for clearing that up. Thanks also for looking into doing this extra test - it'd be interesting, if you can manage it.  Can you just play the MIDI file from the PC, and connect the MIDI OUT to the MIDI IN?
EDIT: I just tried this, and it worked - on my system the two MIDI ports can be opened by seperate applications, allowing me to use the Windows Media Player to play a MIDI file, directed to the output. Pianoteq is able to open the MIDI input port, and it plays the received MIDI.

Greg.

Last edited by skip (02-01-2010 10:13)

Re: ?Zero latency?

Glenn NK wrote:

With the native ASIO drivers of my soundcard and the latency set to 7.1 ms (buffer 312), I have no latency problem, but at 10 ms or higher I notice it, although I could adapt to it.  Perhaps I've been playing a DP too long and have adjusted to its response.

Glen: have you used PTQ (or any other softsynth) in Live-performance? In my experience these latency related problems are more disturbing in Live- situations. Earlier I speculated two reasons for this: a) Live-situation is itself different (we need this immediacy and we expect it to be no-latency) and/or b) sound system is causing extra latency.

Anyway this latency-problem is right now for me more interesting curiosity, not the main thing "in the search for perfect piano setup". I can live with my Mac with reported 2.0 ms software-latency (96KHz resolution with 192 buffer).

P.S: my REAL latency-problem is with my desktop PC. In windows vista I have no means to get low-enough latency with Edirol UA-25EX interface. Same interface works perfectly in my Mac Book pro.

Last edited by Ecaroh (02-01-2010 10:38)

Re: ?Zero latency?

Could you try with XP or Win7? Could be very, very interesting...

And, yes, of course, too many pitch bend messages can create very noticeable delays. (poly) Aftertouch is even worse.

Last edited by Luc Henrion (02-01-2010 16:39)

Re: ?Zero latency?

Greg:

My setup is using Pianoteq in stand alone mode; a midi file in Pianoteq never goes through a midi cable because the sound card is internal (PCI).  So once Pianoteq generates the waves, they go directly from the sound card to the headphones/speakers.

When I play the piano, it generates midi information which travels via the midi cable to the computer and to Pianoteq.  I listen to the sound on headphones plugged into the sound card.  Appears to be no way to add another midi route in this system.

Of course I can't possibly play the 18 note chords, so can't conduct further testing.


Ecaroh:

Don't do any live performing, but when playing for myself with headphones, ASIO latency at 7.1 ms is not noticeable.

As far as I know, ASIO is the ideal for minimizing latency in a PC, and sadly many sound cards don't have native ASIO, which means people are stuck with ASIO4ALL.

Glenn

__________________________
Procrastination Week has been postponed.  Again.

Re: ?Zero latency?

Glenn,
You have an external MIDI IN port on your computer, yes? (I think you must have, because you are connecting your digital piano to your computer this way)   Don't you also have a MIDI OUT port on the computer? It would be most unusual to ONLY have a MIDI IN, and no OUT.  If you have an OUT port, all you need to do is connect the two together, and then play the MIDI file using a MIDI file player. (Windows Media Player works fine for this simple task)

If you need to buy something to make the OUT port available as a physical connection, then it's probably not worth the hassle I guess.

I suppose you could just upload the file and I could do the test.   EDIT: I  see that you've already done that - sorry about that.


Maybe it's not worth pursuing at this point, though - I don't think it's particularly relevant to the O.P's issue - just an interesting tangent.

Greg.

Last edited by skip (02-01-2010 22:18)

Re: ?Zero latency?

Ecaroh wrote:

Glen: have you used PTQ (or any other softsynth) in Live-performance? In my experience these latency related problems are more disturbing in Live- situations. Earlier I speculated two reasons for this: a) Live-situation is itself different (we need this immediacy and we expect it to be no-latency) and/or b) sound system is causing extra latency.

Anyway this latency-problem is right now for me more interesting curiosity, not the main thing "in the search for perfect piano setup". I can live with my Mac with reported 2.0 ms software-latency (96KHz resolution with 192 buffer).

P.S: my REAL latency-problem is with my desktop PC. In windows vista I have no means to get low-enough latency with Edirol UA-25EX interface. Same interface works perfectly in my Mac Book pro.


Hello Ecaroh,

The intent of this posting is NOT to "show off" what I have, but to demonstrate the magnitude of latency (or the lack thereof) that does make Pianoteq a viable instrument for live performance, should the need arise.

Upon reading your post, quoted above, I decided to record (via Xmas gift Yamaha Pocketrak digital recorder) my own latency in a "live" situation for everyone to hear, firsthand.  Specifically, I placed the digital recorder's microphones right on top of my Roland A-80 controller, and positioned them midway between where my hands hit the keys, and the computer monitor's speaker.  Please have a listen:

http://www.forum-pianoteq.com/uploads.p...anoteq.mp3

The setup was simple:  I played loud staccato notes, both singly, then chords, then hands-on tone clusters in order for the recorder to pick up the sound of my hands on the keys.  I turned down the speaker's volume so one could hear the clatter of my hands upon the keys more clearly in relation to the piano's audio output.  The total distance between my hands and the computer monitor's speakers were less than two feet (<60cm); the distance between my hands and the recorder ... same as the distance between the speakers and the recorder ... were each less than a foot (<30cm), so any delay time due to the speed of sound in air is less than 1 millisecond between either sound source.

While I do not use Pianoteq in a live professional performance (at which time I usually play a real Steinway grand), I certainly play Pianoteq live in my home studio, all the time, with absolutely no problems regarding latency of any magnitude that would prevent me from playing live.

My computer/Pianoteq/MIDI setup is as follows:
MacPro quad core 2.66 GHz Xeon processors, 13 GB RAM running Snow Leopard OSX 10.6, with an Apogee Duet audio interface module via Firewire connection.

Pianoteq 3.5.2 setup is as follows:
My buffer size is set at 64 samples, maximum polyphony is set at 256, multicore rendering and cpu overload detection are each checked as being "on", and when laying my arms across the A-80, repeatedly pressing as many notes as possible (in succession, and as quickly as possible), with the sustain pedal depressed, the performance index displays between 41 and 43.

My 20-year-old Roland A-80 mother controller keyboard is plugged into (via old style MIDI out) the MIDI In port of a MOTU Midi Timepiece AV, whose USB output goes directly into my MacPro.   Audio output comes out of the computer into the Apogee Duet audio interface module.

As stated earlier, this particular setup allows me to play Pianoteq live without giving a single thought about latency.

Opinions and comments are encouraged.  Perhaps other forum members might like to submit actual recording situations, especially those whose latencies are unplayable in a live situation, wherein the keys are heard and the magnitudes of latency may be heard and discussed.

Cheers,

Joe

Last edited by jcfelice88keys (03-01-2010 09:00)

Re: ?Zero latency?

Joe,
Inspecting two of the single note strikes, I am completely unable to seperate the sound of your finger hitting the key from the sound of Pianoteq.  It might be worth recording some strikes using your fingernail, to cause a distinct "click", which might then make it easier to measure the latency. (I will not be held responsible for any scratching of the key surface that may occur )

At the moment, though, your setup appears to be excellent in terms of latency, you big showoff.

I've uploaded a single note strike:
http://www.forum-pianoteq.com/uploads.p...U_nn60.mp3
This clearly shows the click of my fingernail making contact with the key, and an overall latency of approximately 65ms. I would subtract about 3ms for sound travel from speakers, and probably a few more ms for the initial part of the piano attack, before it has reached sufficient amplitude to reliably detect.  Instrument is C3 Solo.  I say again - I do not have any problems playing with this latency.

Greg.

Last edited by skip (02-01-2010 22:02)

Re: ?Zero latency?

skip wrote:

Glenn,
You have an external MIDI IN port on your computer, yes? (I think you must have, because you are connecting your digital piano to your computer this way)   Don't you also have a MIDI OUT port on the computer? It would be most unusual to ONLY have a MIDI IN, and no OUT.  If you have an OUT port, all you need to do is connect the two together, and then play the MIDI file using a MIDI file player. (Windows Media Player works fine for this simple task)

If you need to buy something to make the OUT port available as a physical connection, then it's probably not worth the hassle I guess.

I suppose you could just upload the file and I could do the test.

Maybe it's not worth pursuing at this point, though - I don't think it's particularly relevant to the O.P's issue - just an interesting tangent.

Greg.

Greg:

Yes of course you are right (and in truth I realized this before).  However the problem is this:

The piano and the computer are located back to back on a common wall in different rooms.  To change the cables would require moving the piano to get at the cable, then pulling it through the wall so as to connect it to the midi on the computer (actually my s/c has an external breakout box).

Then to reconfigure everything, I'd have to repeat the process.

I was hoping someone else would come up with some evidence to same me from going through all this, and it seems that Joe Felice has come to my rescue (actually I'm not surprised at this at all).  Well done Joe.

Another comment - I'm waiting for the OP to return and add any useful comments he may have.


Glenn

__________________________
Procrastination Week has been postponed.  Again.