The Song Fight Bulletin Board

Hey all. I'm using an Alesis Multimix 8 Firewire with a Macbook (Tiger, 2 GB RAM) and Cubase LE. I have a huge latency problem. I've tried changing the buffer size but nothing seems to improve. Any tried and true tricks (either w/ Cubase or something specific to the Mac OS X Audio Midi Setup app) that people know and that I'm missing would be greatly appreciated. It's kinda upsetting because when I was using a crappier USB unit with a computer with less RAM, I had no latency problems, and only after moving onto Firewire and more RAM does latency bother me.

Are you using the WDM or ASIO drivers? Have you tried both?

Ken

Is there such thing as WDM drivers for Mac?

Have you checked if your hardware supports zero latency monitoring? Since it's a mixer, I'd be surprised if it didn't. What you want is to monitor the signal going in -before- it goes to the computer, mixed with the playback audio coming out of the computer. Perhaps the manual has some info about that?

oh,oh oh... I'm sorry I'm late...

Have you checked if your hardware supports zero latency monitoring? Since it's a mixer, I'd be surprised if it didn't. What you want is to monitor the signal going in -before- it goes to the computer, mixed with the playback audio coming out of the computer. Perhaps the manual has some info about that?

Ok, so I'm not that funny.

The only time I've ever experienced a latency issue was when I had the song I was working on stored on an external USB hard drive. As long as the song I'm working on is on the same machine as my Cubase program, I never have issues.
Not sure if this applies to you, but it drove me nuts until I figured out why it was happening.

Also, you might have to keep your multimix 8 program open and running while using Cubase. I know that I have to keep my preSonus Firebox control panel open while using Cubase. In fact, if I don't open the the PreSonus FireBox mixer, I can only hear what I'm playing when I click the monitor, even though I don't use the FireBox mixer.

My setting on my FireBox control panel:
Sample Rate: 44.1 kHz
Latency: 8.0 ms
Clock Source: internal

I've got something you can stick in your FireBox...
oops sorry, not helping!

BLT..

8ms is that the lowest latency available on ur box. Or is that the happy medium between latency and crackling.

And do you up it for mixing, cos if you do the cpu can handle more.


I use as low as I can for recording, then pop it up for mixing. Makes sense to me. I usually record at 2-4ms.

No, I never change it. And yes, my CPU gets taxed hard at times. It really sucks to have the computer freeze while I'm in the zone. I've started exporting tracks with the FX I want, then bring it back so I can turn off the VST plug in. It's been working fine, but way more time.
So you're saying that I can minimize my computer freezing by changing my latency?

I would rather DIE than record with buffer crackling. Yuck!

I set my Firepod to:
1.5ms latency
48 khz
CPU: medium

whether it actually pays attention to my settings is unknown, but it works.

We try to find creative and artistic uses for buffer crackling...

Show of hands. Who exports at 16 bit or 24 bit?
16 bit at 44.1, here.

Billy's Little Trip wrote:Show of hands. Who exports at 16 bit or 24 bit?
16 bit at 44.1, here.

That's an extremely interesting (to me, at least) question. Out of habit I dither to 16 bit (44.1khz, natch. No other way to go if you're at all interested in compatibility), which is fine if I'm putting the songs on a CD, but typically my target medium is mp3. mp3s can be encoded from a 24bit source. I have truly no idea which sounds better in theory, a dithered 16bit mp3 or an undithered 24bit mp3.

As for the question at hand, on my Windows box I had a lot of latency and dropouts with my firewire card at first, but it turned out to be a hardware interrupt issue. Disabling my wireless networking card solved the problem, and now I rock the 3ms. Very few people seem to have had the same problem as me, but it might be worth a try.

Billy's Little Trip wrote:So you're saying that I can minimize my computer freezing by changing my latency?

Yes. Thats exactly what Im saying, just not as clearly as that. Damn my inability to comunicate effectively.


But er, yeh.... The higher the latency the less the CPU is taxed, I tend to record at low latencies, without buffer crackle obviously. And add in all the heavy VSTs at mix stage, with a higher latency. I might use a few lightweight VSTs during recording just to get a rought sound of what Im looking for, but these will be replaced by the heavy ones later.

Tho I must also point out I use very few VSTs, due to my arsenal of DSPs

Also the point of exporting WAVs with the VST on and then removing the VST from the project is also a great idea.

I always turn all my VSTis into WAVs before starting a mix. I never mix with VSTis.

Me$$iah wrote:I always turn all my VSTis into WAVs before starting a mix. I never mix with VSTis.

So I am thinking in the right direction.

OK, I feel like I'm hijackingoff this thread, but one last thing I'd like to throw out here for a thought or two.
If I have a left and right rhythm guitar running on separate tracks that I'm panning hard right and left, but export the two together as a single stereo track, am I losing OR gaining anything as apposed to keeping them separate left and right mono tracks? Do you know what I'm saying? I personally can't hear a difference, and it really cuts down on my final tracks and vst's for CPU purposes.

...oh, and what's a dsp? Sounds like I need to put a condom on my computer.

that makes sense... kess track less CPU tax



However I say if its mono .... keep it mono.....


I use left and right tracks... sometimes I add different fx to both channels... this is harder with a stereo file.
Tho the lower track count may be an effective CPU trick, I would personally keep em as mono tracks panned.
This also gives you more options during mixtime. the ability to move guitars round in the mix.... bring in FX on one side not the other etcetcetc..I think the cost outweighs the benefiet.



oh yeh... DSP (DigitalSignalProcessor)..... dsps free up cpu.... they move the wotk from the computer to its on dedicated 'comuter' within the DAW....

I use Sonic-core (formerly Creamware) Ken swears by UAD some use TC or you can spend 40000 on a real protools HD system...... its all good


DSPs rock

OK, I remember talking about DSP in the past. Seems logical for a computer daw.

As far as exporting a left and right guitar track as one stereo track. I only do this after I've decided on the correct panning and fx, etc, to work with the whole mix. Once I've made a decision on these things, I rarely change them later. If I need changes, I actually go back to my original separate mono tracks and re-export them with any changes needed.

By the way, I should mention that I always start a new project called "the mixing tracks". Each song I do now has the recording tracks, which are all of my original tracks that often number in the thirties and forties, and the mixing tracks which are the condensed exported tracks from my original recorded material. Track number 13 is never used and deleted.

Also, I'm not 100% convinced that one stereo track is any lighter than two mono tracks. But I do know that five or six tracks condensed to a stereo track sure is.

I have songs where I condensed all of the drum tracks to a single stereo track, and some where I kept all of the drum tracks separate. I guess I'd have to post each version to see if people can tell the difference.

Yeah, this thread has been hijacked, but that is what happens when you pose a question and then don't come back to check in with the answers.

I don't want to get you too far away from your work flow, but have you tried using group tracks? Here is something I do regularly: If I have two guitar tracks (typically a double) that I am panning left and right, I create a group track and send them to it. This way I can use ONE stereo compressor/EQ/Effect for both tracks instead of one on EACH track. It saves a bit of processing power. I also do this for multiple back up vox.

I do this for drums as well, but it is a bit different as I tend to treat each drum track individually and then use the group track to effect the entire kit at once.

This is all pretty easy stuff in Cubase. PM me if you need some specific directions.

Honestly though, if your computer can't handle your mixes, it is time to upgrade.

Be well,
Ken

yep... what Ken said about group tracks is spot on...

and dont forget AUXbusses also.






(you know! a comp and reverb on an FX track with all Vox going through it for example..)

I never used group tracks, nor any more than bus 1 and 2. Guess it's high time I start.

what's an aux bus?