I read an SOS article where they claim that the Korg Wavestation's internal sound engine has 2 ms latency.
For me, this is perfectly fine (I own a Wavestation) since I natively work with sound cards that have 0.7ms latency (2 x 0.7ms latency gives 1.4ms for the entire DAW).
I can't see that Korg has come out with official specs for just latency when it comes to nautilus.
Wondering if anyone has more information on this? I need to know before buying as I am sensitive to latency values higher than 3ms.
To clarify. When I say Internal Latency, I mean the time it takes to hear the sound AFTER I press the key on the keyboard.
Please let me know, strange that there is no info on this.
How long does it take from first finger contact with the keyboard key until the double contact closure completes when playing softly? Guessing it's probably going to be a good deal longer than 2ms. So, right at the start of the chain of events, there is already considerable delay.
Korg didn't release latency figures for the Kronos, as far as I'm aware, so there's no reason to suppose that the Nautilus would be any different.
It might be possible to get some measure of latency without taking the Nautilus apart and invalidating the warranty. A switch could be positioned externally so that it is activated as a key is pressed, and adjusted in such a way that the switch contacts close just as the second key contact closure occurs. The depth of keypress required could initially be established by placing an external mechanical key stop, and gradually adjusting it downwards until a note is heard on the output. The switch could be connected so that it triggers an oscilloscope sweep, whilst monitoring one or both of the main audio output channels on a second scope trace or channel. For the experiment, choose a Program having a fast attack characteristic; ideally zero.
There has been some discussion about sound engine latency in the Kronos forum. The internal architecture of the two keyboards seems to be sufficiently similar for the threads to be likely to be valid for the Nautilus also:
voip wrote:Korg didn't release latency figures for the Kronos, as far as I'm aware, so there's no reason to suppose that the Nautilus would be any different.
It might be possible to get some measure of latency without taking the Nautilus apart and invalidating the warranty. A switch could be positioned externally so that it is activated as a key is pressed, and adjusted in such a way that the switch contacts close just as the second key contact closure occurs. The depth of keypress required could initially be established by placing an external mechanical key stop, and gradually adjusting it downwards until a note is heard on the output. The switch could be connected so that it triggers an oscilloscope sweep, whilst monitoring one or both of the main audio output channels on a second scope trace or channel. For the experiment, choose a Program having a fast attack characteristic; ideally zero.
There has been some discussion about sound engine latency in the Kronos forum. The internal architecture of the two keyboards seems to be sufficiently similar for the threads to be likely to be valid for the Nautilus also:
Thanks for the info, yes, I saw that old post before. My only concern was the latency created when I actually play the Nautilus and then to the sound that comes out directelly, through its analog L/R outputs. Still not a clear answer on that, but I guess it will be lower than 4ms, which is a good trade off, in my opinion.
I investigated this a few years ago and there are many ways of measuring latency. The method I employed at the time was to test the round-trip audio latency of the digital audio input to the digital audio output, under the assumption that the audio engine would perform similarly for audio routing as it would to audio generation.
I don’t remember the exact figures, but without ADC and DAC I seem to recall the round trip latency was well under 2ms. Of course, digital to audio conversion will add to this, as will key scanning and/or MIDI parsing.
To put it another way, I have never felt the latency to be noticeable at all on the Kronos, especially compared to VSTs (even with low buffer sizes). It is one of the key reasons I still prefer the Kronos over computer-based setups. The only time I have noticed latency is when monitoring audio in from a microphone via headphones, where “comb filtering” is slightly evident.
Current Equipment:
Korg Kronos 2 88, Reface CS, Roland JV-1080, TE OP1, Moog Subsequent 37, Korg ARP Odyssey, Allen & Heath Zed 18, Adam F5, MOTU MIDI Express XT, Lexicon MX200 & MPX1, Yamaha QY700, Yamaha AW16G, Tascam DP008ex, Zoom H6, Organelle, Roland J6 & JU06A
Previous: Triton LE 61/Sampling/64MB/4GB SCSI, MS2000BR, Kronos 1 61, Monotribe, NanoKontrol, NanoKeys, Kaossilator II, Casio HT3000, Roland VP-03, Reface DX, Novation Mininova, MPC One
SeedyLee wrote:
To put it another way, I have never felt the latency to be noticeable at all on the Kronos, especially compared to VSTs (even with low buffer sizes). It is one of the key reasons I still prefer the Kronos over computer-based setups.
voip wrote:A switch could be positioned externally so that it is activated as a key is pressed, and adjusted in such a way that the switch contacts close just as the second key contact closure occurs. The depth of keypress required could initially be established by placing an external mechanical key stop, and gradually adjusting it downwards until a note is heard on the output. The switch could be connected so that it triggers an oscilloscope sweep, whilst monitoring one or both of the main audio output channels on a second scope trace or channel. For the experiment, choose a Program having a fast attack characteristic; ideally zero.
It is perhaps easier, and more accurate, to simply put a microphone close to the Nautilus keybed, and record its sound at the same time as the sound coming out through the L/R outputs. This is because the keys emit an acoustically measurable sound when pressed and they come into contact with the rubber contacts under the keyboard.
