sigasa wrote:garrettb12 wrote:I have owned pianoteq for many many years. I have never used it in a live capacity because I never got the latency dialed in. I've used desktop PCs with dedicated beefy sound cards, and many different computer's onboard sound card. I've never been able to get the latency down to a minimum. For me, the ultimate test is if I push my keyboards key (it has a hammer action) and the thunk of the hammer and the sound from pianoteq are noticably apart/delayed, then I cannot use it for live playing.
A lot of you posting on the forums here are using Mac computers. Does anybody here own a Windows PC and have achieved unnoticeable latency? What is your setup?
Also, as a laptop user, are there any external sound cards I can use to get zero latency? I have some high end rog strix laptop.
I used to try and get as low latency as possible until I realised that I personally am much happier, and achieve a more realistic playing experience, with a latency that works in harmony with the hammer action. On an acoustic, as I'm sure you can appreciate, there is a delay as the hammer is thrown against the string. I also find that a certain amount of latency makes a difference to the sound with regards to the attack. May I suggest you experiment by varying the amount of latency and listening to (and feeling) the difference, especially over speakers. In my albeit limited experience, less latency isn't always better/desirable.
Warmest regards,
Chris
P.s. What keyboard are you using?
Totally agree on this point:
The search for low latency is useless below a minimum duration. It does not necessarily require the use of an audio interface dedicated to musical instruments. For my part, under Windows 10, I use an external USB DAC that is not specialized for the use of a keyboard: Dac Topping 70 Velvet equipped with the excellent AK4499EX chip providing very low distortion and a good signal-to-noise ratio. With this DAC, under Pianoteq 8.3.2 Pro, I generally use without apparent inertia (in 256 polyphony) the Windows audio output, the sampling frequency 96 khz and an audio buffer size indicating 3.3 ms. In Asio mode, still in 96 Khz, this DAC works very well on my PC with a buffer size indicating 1.3 ms (but with sometimes some crackles if I go down to 0.7 ms). The keyboard: Casio GP-500 is connected directly by USB to the PC (I have not been able to find any figures on the value of the latency induced by this keyboard with its USB driver under Windows 10 (I think that the Casio GP-500 keyboard uses USB 2, not 3, but that should not have an impact on the overall latency).
It seems important to me to put these latency values into perspective with the following elements: The duration of a quadruple eighth note at a tempo of 140 bpm at the quarter note is approximately 27 ms. From a physiological and neurological point of view, it is typically necessary to count between 100 and 200 ms between the intention of movement at the level of the brain and the effective start of mechanical muscular pressure by a finger on a keyboard key: (20 to 30 ms of neuronal transmission between the brain and the muscles + 50 to 100 ms of reaction time of the muscles to the neuronal signal + 30 to 70 ms of coordination time between the muscles of the fingers and wrist ...). At the speed of sound in air, it takes about 1.5 ms to propagate the sound vibration from a loudspeaker to the ear 50 cm away (which is faster than the 300 ms needed to propagate this same vibration from the pipes of the great balcony organ of Notre-Dame Cathedral in Paris to a listener at altar level ...). We must add to the sum of these latency delays about 10 to 20 ms additional for the sound vibrations to pass from the eardrum to the areas of the brain responsible for interpreting the messages transmitted by the auditory nerve.
In conclusion, to achieve zero latency, we will have to ask the Modartt team for an evolution of Pianoteq to integrate a "brain-to-brain" interface, ranging from the mental intention of virtually hitting a key to the interpretation of the virtual restitution signal by the brain centers responsible for interpreting the pseudo-sound signal. It is appropriate to add to this evolution request, a neurosurgery service for the implementation of intracranial electrodes, because the external neural headsets of BCI interfaces are currently unable to transmit or recover a sufficiently powerful and localized signal!
Warmest regards,
Bruno.
