Is that normal that when I’m recording vocal and latency is set for a big buffer number (saving processor power) there is a huge delay between recording vocalist and what I here in the speakers and headphones? It disappear after I change latency to 512 for instance. But than my Asio is going nuts. So I don’t have to much room to play with.
Is there any other way to keep latency “big” and no recording delay beside shutting off all FX’s which is bad for the person that is recording because he is recording dry?
Monitor via your interface.
that is what I do. I have delay in monitors as well.
latency doesn’t cause delay, latency is a measure of delay.
Cubase has a direct monitor option, but yes that will be dry. Actually dry is the best way to do vox anyhow. I guess everyone wants to have verb max these days so they don’t know just how horrible they sound and how off key they are singing and then it’s the engineers job to pitch correct everthing.
I’ve been at places who have an independant headphones monitor loop that they have external FX on. They mix the DAW feed into the headphone mix. That way they get the dry vocal on the track, but the singer gets some verb and/or chorus. But, the amount of cleanup you have to do on those vocals is disgusting.
If you´re experiencing latency depending on your buffer settings, that´s probably not what you do, but hard to tell from your problem description.
The MR 816 CSX should let you do hardware monitoring and Reverb with so called “zero latency”
Actually “the best way” is the way that works best depending on the situation, and not some general “that´s how it´s always been” IMHO
True, and it depends on perspective. From the perspective of the engineer, there are very few cases where wet vocals are considered a good idea. But, then you have to deal with the wetness while you are dealing with the vocal. There are exceptions. But they are usually FX based, not “singer needs more verb in their monitor”.
It is a generalization, but in my experience I can say that the worse the singer, the more wetness they want in the cans. The more wetness in the cans, the messier the vocal on the track. Wanting to actually print the FX with the vocal makes it even worse.
However, again the goal here is to let the guy know that plugins have latency that adds up. More plugins = more latency. External monitoring on his audio interface is deffinetley possible. Using external FX to feed the cans can be done as well. But, you should really be driving to get good recordings. Unless for some artistic reason, sloppy, offkey, constantly searching for a note vocals is what is being sought.
I don’t really know what is causing it but when I have buffer set to over 1000 the dealy effect is on.
When I decrease buffer to 512 the delay effect is gone.
I don’t have any FX set right on my audio interface.
I’m recording dry, but I would like to have vocalist to hear the backround with FX’s. Not the vocal it self.
I have Output set up to default, Studio has 2 sets of monitors and headphones are connected straight to the audio card.
No matter how I connect it the delay effect depend of my latency.
It has to be that I’m doing something wrong.
You are doing nothing wrong, the more buffers the bigger the delay, thats how it is.
One way of tracking vocals to a fully laden mix with low latency is to do a quick mixdown of the backing imported back into the project, then switch off all plugins, mute all channels and set the smallest buffer size you can get away with, then track the vocals to the quick mix. Obviously once tracked you can bump the buffer size back up again and unmute track and switch on the plugins.
Well, generally this is a thing that concerns the singer in the first place, not the engineer. If a drummer wants more cowbell in his cans, he´ll get cowbells until his ears bleed. Of course it´s useful if the “engineer” knows how to create a different mix for himself, so he does not have to deal with the “wetness”
Not necessarily. First of all the highest Plugin delay is used for all tracks. Adding Plugins with smaller delay will not increase the overall latency. It will increase the CPU load though, so at a certain point you´ll have to raise the soundcard buffer to avoid dropouts - what will indeed increase latency, but not the plugins themselves.
Well, if that played a major role nowadays, you would not have everyone scream for “Multitrack drum quantising”, Melodyne, and such things, instead of getting people who know what they do…
Like Split already said, the more buffers, the higher latency.
The 2nd option to what split wrote is to keep your mix as is, set the buffers high, and use direct monitoring for the vocal record track.
Yep latency IS delay caused by your machine not being able to process the incoming signal fast enough to be able to return it to the monitor quick enough… yes a VERY good work around is to just run off a quick 2ch mix so you can squeeze the maximum out of your interface and machine… you could also create stems… all the drums grouped to a stereo track, bass to a mono track, guitars to a stereo track etc… so you end up with probably 8 or so channels maximum so at least your vocalist can ask for less guitars or more bass etc… in their mix.
If you use the control room in cubase then it’s DEAD easy to set up a headphone mix separate from the main mix so you can give the talent as much verb as they want and you can hear it dry… you can also toggle between their mix and the main mix… EVERYONE is happy then… of course you do need an interface with more than one stereo output on it… unless of course you are ok with working in mono with mono foldback for the session.
But yeah the monitor mix should be TO THE PERFORMER’S TASTE and NOT to the engineer’s!.. you are trying to capture a performance from someone… MOST vocalists DO like a bit of reverbs in their cans… even one or two of the opera and classical singers i’ve worked with like it! it is THEIR performance and NOT the engineer’s so it is up to the producer/engineer to facilitate the best conditions they can for their performer… of course some performers do take the p@ss but you don’t work with them again.
Have you also tried freezing any audio/instrument channels that have inserts on them? if you are running a lot then this can save a massive amount of ASIO time too… alternatively you could use an external reverb unit and a small mixer that has subgroups on it, run your mic into the mixer… assign it to group 1/2, send the signal from channel one of the group to cubase and then channel 2 to the reverb unit and then the reverb back into the desk so you split the signal in two… exactly the way i used to do it before the control room function just a few ideas for you.
Hello , I’m OK for the direct monitoring aspect , and I have no delay with my new MR816CSX , but is it possible to have the same function than in Logic Pro 9 with " Auto input monitoring " : I can’t do it in Nuendo ?? If I want to hear my playback , in the same time I want to hear my input , on the same track , using recording in lanes , not possible…
Here is the Logic’s documentation for this function :
Using Auto Input Monitoring for Recording
If Auto Input Monitoring is turned on, you will hear the input signal only during the actual recording—before and afterwards you’ll hear the previously recorded audio on the track, while Logic Pro is running. This helps you to judge punch-in and punch-out points when you are punch recording. If Auto Input Monitoring is turned off, you will always hear the input signal.
To turn on auto input monitoring
Your individual recording situations will determine whether or not to use auto input monitoring. Consider the following examples:
Singers cannot hear themselves while Logic Pro is in playback mode—all they can hear is the old recording. In this case, turn off Auto Input Monitoring.
The recording cannot be heard in playback mode. In this case, turn on Auto Input Monitoring to hear audio on the track.
That´s “tape machine style” monitoring in Cubase (Preferences)