Skip to main content

Hi!

I've got some annoying latency problems with my recording system that I can't figure out how to get rid of. I've read a lot and tried different solutions but can't get rid of the problem. It's a matter of old recording stuff that I have had the money to get so I could get things up and running. Aventually things will get upgraded but money don't grow on trees :(

Equipment check:
AMD Athlon 2800+ clocked at 2,19ghz with 2gb of ram
Hercules 16/12fw interface
Samson c-control
Behringer headphone amp
Cubase le4

So here's my deal: If I put a mic directly to one of the hercules preamps and monitor that signal directly from the hercules phones out I get a delay that is annoying. It's like having a dubbler effect that is slightly delayed. In the hercules manual it says that I should be able to monitor the direct signal there without latency? In Cubase it has a input and output latency of 10.667ms and I can't lower this no matter what I do. I currently run all my projects at 96.000khz sample rate with 24bit record format. If I change this to 44.100khz/16bit I get 23.220ms delay on input and output and things get worse in the headphones. In the hercules control panel properties I have the buffer size set at 64.

I've tested the dpc latency checker and with that I get a maximum peak at 900 µs and normally stays at 500-600 µs. I've recorded with people already, everything from drums with 11 mics to guitar and vocals. Guitarists hate this delay and like I said, vocalists hear a doubling effect.

Is there any way that I can get rid of this or am I stuck with this latency until I get some new gear?

Comments

Boswell Mon, 11/05/2012 - 03:58

I've never tried one of those Hercules interfaces, but I thought they had a low-latency (1ms) monitoring route that you could use for any two inputs going through to outputs 1 and 2. It's not going to help when using a whole set of mics on a drumkit, but for a vocal mic and lead guitar it would be a big improvement on going in/out of the DAW buffers.

Other than that, you may have to consider the sort of techniques that are common for live performance. These include splitter cables that split the mic signals between your interface and the inputs of a low-cost mixer solely used for headphone monitoring. The mixer is not in the recording chain so its audio quality does not affect the recorded results.

bouldersound Mon, 11/05/2012 - 07:33

Nutti, post: 395701 wrote: In cubase it has a input and output latency of...

Don't monitor through Cubase, use the interface's control panel to get low latency hardware input monitoring. This bypasses the long trip through the computer and sends input signal right back out the main (1/2) outputs with almost no noticeable latency. See section 5.4 of [[url=http://[/URL]="ftp://ftp.hercules…"]your manual[/]="ftp://ftp.hercules…"]your manual[/].

Nutti Tue, 11/06/2012 - 00:40

well this was embarrissing :D I've had this interface for about a year now without any luck in direct monitoring...been searching the net for hours and hours without finding that in the manual. +5 points to me!

It was easy to monitor now, but the direct monitoring is panned hard left/right for the two channels beeing monitored. Do you guys know how to get each channel centered? It's probably as easy as the last problem I had I guess...

Boswell Tue, 11/06/2012 - 03:09

It's not stereo - they are separate mono channels going to outputs 1 and 2. Use a simple external mixer on the outputs if you want to be able to pan each channel's position in the L-R field. Something like [[url=http://[/URL]="http://www.musician…"]this [/]="http://www.musician…"]this [/]$50 mixer that has individual pan controls on two inputs would do the job.

e-mixmaster Tue, 11/06/2012 - 07:19

Boswell is right...

May be you want to invest in a decent console so you can monitor and distribute signals form there.

The only thing that you should be aware of is the added electronics the recorded signal passes thru, so you may wanna split the signal from a pre-amp if you can: one signal going to track and the other for monitoring, in this case a console.

.. so the deal is to have your daw signals and recording signals all together on the same place.

The other option for you is a more sophisticated system, but that requires some investment.

Best !

[[url=http://[/URL]="http://e-mixmaster…"]Home - e-mixmaster.com[/]="http://e-mixmaster…"]Home - e-mixmaster.com[/]

bouldersound Tue, 11/06/2012 - 07:20

The pictures in the manual are too blurry to tell if the control panel can put a single input into both ears, but most interfaces have input monitoring that can do that. If yours doesn't then a small mixer would work, and some headphone amps can as well. It's also possible to build a simple summing network with a couple dollars worth of parts.

bouldersound Tue, 11/06/2012 - 07:25

e-mixmaster, post: 395764 wrote: Boswell is right...

May be you want to invest in a decent console so you can monitor and distribute signals form there.

The only thing that you should be aware of is the added electronics the recorded signal passes thru, so you may wanna split the signal from a pre-amp if you can: one signal going to track and the other for monitoring, in this case a console.

.. so the deal is to have your daw signals and recording signals all together on the same place.

The other option for you is a more sophisticated system, but that requires some investment.

Best !

[[url=http://[/URL]="http://e-mixmaster…"]Home - e-mixmaster.com[/]="http://e-mixmaster…"]Home - e-mixmaster.com[/]

No need to over think this. Boswell's suggestion was to use a mixer on the interface's monitor output so inputs on one channel are heard in both ears. It's not in the record path. And I still suspect the input monitoring has some way to do this without more hardware.

e-mixmaster Tue, 11/06/2012 - 07:42

Well..

if he sets his buffer to 64 and his getting 20 milliseconds of latency...some thing is wrong here.

Anyway.. if he is getting that latency, I don't think the problem only lies on the monitoring path, it looks like the daw is not capable of such task. He, the OP, said he's got a year with this system with out any luck.

cheers and luck !

[[url=http://[/URL]="http://e-mixmaster…"]Home - e-mixmaster.com[/]="http://e-mixmaster…"]Home - e-mixmaster.com[/]

bouldersound Tue, 11/06/2012 - 07:56

e-mixmaster, post: 395770 wrote: Well..

if he sets his buffer to 64 and his getting 20 milliseconds of latency...some thing is wrong here.

Anyway.. if he is getting that latency, I don't think the problem only lies on the monitoring path, it looks like the daw is not capable of such task.

The latency problem has already been solved by using the interface's hardware input monitoring feature.

Boswell Tue, 11/06/2012 - 08:07

I didn't examine that Hercules interface in great detail, but my impression was that the low-latency monitoring route is a bit half-baked. At least it allows a pair of channels to come out for monitoring without their having to go via the computer, but to assume that they are an incoming stereo pair and not to allow centred imaging of separate mono channels is a real shortcoming.

It could be that it's not worth spending even $50 on a 2-input panning mixer just for the purpose of monitoring only two channels, but instead put the money towards a better unit that can mix multiple channels for monitoring at the same time as passing them on to the interface for recording. The alternative is to consider upgrading from the Hercules to a more versatile computer interface that has configurable internal monitoring on all channels as well as better-quality pre-amps.