Hello all – I’m a newbie (apologies up front ;.) and am wondering if the following can be done.
I have a 5 piece band and would like to use NTrack and our mixing board to create 6-track recordings (vocals, drums, bass, Guitar1, Guitar2, keyboards).
First, I would buy a soundcard with 2 analog inputs (e-mu or m-audio – advice appreciated!)
- vocals: mixing board aux1 bus to the left side of analog input 1
- drums: mixing board aux2 bus to the right side of analog input 1
- bass and keyboards are sent from their individual mixing board channels to the left and right side of the 2nd analog input respectively
- both guitarists use a Line6 POD XT which has a USB out
So, that’s recording 6 simultaneous tracks (4 analog tracks through the 2 inputs of the sound card and 2 seperate digital tracks through USB).
Is this feasible? Is there a better way?
What issues should I expect to face?
Thanks in advance!
You are thinking along the right lines. Only problem is you will have heck of a time trying to get the POD’s audio stream and the stream from your soundcard to stay synchronized and thats IF you get all the I/O drivers to live happily together. Your USB stream (guitars) and the stream from your sound card will most likely drift. This will make your guitars lead or lag the other tracks and it will get worse with longer songs.
Since you already have a Mixer, what about an M-Audio Delta 1010LT. They can be had for about $250 new and much less via EvilBay. That card will give you 8 analog inputs and 1 stereo SPDIF digital input. Use the analog ins for your mixing console and pipe one of your POD’s through the SPDIF. I think PODs have a SPDIF out…right?
Anyway, one Delta driver to run, n-Track, a little cabling + your mixer and bingo! 8 analog tracks and 1 stereo POD track simultaneous recording.
See you at the Grammy’s… 
TG
PS Welcome to the forum dude. (or dudette?)
Hmm. You have described a sound card with 4 inputs - two stereo inputs, actually, which would be rather odd, wouldn’t it? The cost of multi input cards is coming down - so why not go for 8 as TG suggested? There are a number of good options - from “all in one” solutions from Aardvark like the Q10 or the Presonus stuff I keep seeing in the catalogs, to cards that require separate mixer/preamps.
Thanks for the feedback!
I performed a quick test using my current soundcard and the POD XT USB input that seemed to work well.
In the NTrack “Settings–>Audio devices”, I selected the MME inputs for both the soundcard line in and the POD XT digital input. I then recorded a couple minutes worth of guitar and one vocal and it seemed to work fine.
Since NTrack is recording all of the tracks, is Synchronization an issue? (i.e. can you help me understand the synchronization problem? ;.)
Thanks again
TomS,
Thanks for your feedback as well. I have seen the e-mu and m-audio soundcards that have 2 (stereo) analog inputs. Since our instruments/vocals are mono, my thought was to record a seperate track per stereo channel.
But before making any investment, I was looking for feedback from the NTrack group here. The 10 input soundcards I have seen run about $250. That’s more than we wanted to invest. However, I also don’t want to invest in an idea that won’t work!
Which soundcards would ya’ll recommend I check out?
The Delta 1010LT is $219.95 @ http://www.zzounds.com/item–MDOD1010LT
Don’t cheap out unless you absolutely have to. I think you will find that you will very quickly outgrow anything less than the 1010 or something equivalent, if you are regularly recording a 5 piece band. Just the ability to have more tracks available for drums will improve your mixing options a ton.
I have an Audigy I and 2 SBLives installed on my computer. They are linked with SPDIF cables and all ASIO inputs end up on 1 card so they stay in Sync. This is done using the kX drivers.
With another trick suggested by a kX user, I added one of those adaptor cards made for onboard sound, and with a command line entry I now have 8 Mono inputs.
The 3 sound cards cost me around $70 U.S. from Ebay.
You could also achieve the 8 Mono inputs with 2 Audigy sound cards hooking in 2 adaptors.
http://come.to/sblive
Right now I am using ntrack V 3.3, PIII 800 Mhz, 384 Meg RAM.
Here is my current favorite song done by a kX user. Not sure which sound card (s) he is using but it would be some flavor of SBLive or Audigy:
http://www.driverheaven.net/showthread.php?t=58340
Note: you have to right click on his link and save the shortcut to your hard drive then double click SON.asx to play the tune:
| Quote (guitarzan @ Oct. 14 2004,07:42) |
| Since NTrack is recording all of the tracks, is Synchronization an issue? (i.e. can you help me understand the synchronization problem? ;.) |
I'll try. I may screw this up so somebody else jump in here if I'm wrong!
Let's assume we are talking about 16 bit 44.1 khz analog to digital conversion for simplicity. What this means is the card is grabbing a sample of the audio at 16 bit precision at a rate of 44,100 samples per second. The soundcard (or cards) have an internal oscillator to derive that clock frequency of 44.1khz. No two cards, even the same brand and model will have EXACTLY the same clock frequency. Thus, UNLESS they are somehow synchronized to a common clock, drift will occur. Besides your tracks being slightly "off" timing wise, you may hear strange phasing or flanging sounds in your audio. Many cards, even some of the lower cost SoundBaster variety, CAN be synch'ed together via the SPDIF connections. Of course, you will lose that SPDIF port for carrying audio.
I hope that makes sense. You are doing the right thing trying to "root out" possible snafus before you get started. Research pays off when undertaking any project.
TG
Doug, that is very clever. I wouldn’t have the courage to try that!
I’m with Phillip - a few hundreed dollars may seem like a lot right now, but you should always buy as much power as you can afford, even it if means waiting a month or two - you will soon find that you will have pushed the envleope on whatever you have. And the 1010 is just plain dirt cheap, and is very good. Drums alone, for example, are really hard to do on one track. Two tracks will get you separate snare and kick which will help immensely, four will get you snare, kick, and 2 overheads, and at that point things get a whole lot easier. And a good drum sound is key - perhaps the key - to rock and roll. IMHO.
I have successfully used 2 different sound cards made by two different card manufacturers to successfully capture audio from 4-track tape.
I think that this is a “try it out and see if it works” situation rather than a “don’t bother” situation. The only thing that could go wrong with it is that there would be significant drift and then you are out only $29 and a couple hours of practice time.
I do not buy gtr4him’s argument about the clocks being that far out of synch. Almost certainly, the clocks are phase locked to the bus speed of your computer (otherwise they would have significant problems transferring data to memory). This is likely to induce some timing jitter, some drift on the order of a few samles (=flanging), and some annoying problems with differing latencies.
Personally, I have been very pleased with the changes Line6 made to their USB communication in the last patch. I owned a guitar port and before that time, I had all kinds of problems with the recorded guitar part not being in synch with the othe rtracks. The effect was random, and was the fault of Line6, but like I said, after the last patch everything was very fine.
I say go for it! The worst that can happen is some wasted time and a couple of bucks down the drain… 
By all means, try it. It’s worth an hour and the outlay for a SBLive or two… My own take is that if this requires more than a couple hours of downloading and changing drivers, tweaking the BIOS, making little jumper cables etc, then it’s more cost-effective to just cough up the $250 for something that will work right the first time and will have guaranteed sync.
I think you’ll find that using independent cards will work for the normal under 5 minute pop tune, but the clock differences would become an issue in longer recordings - eg 30 min or more live recording.
| Quote (Ben Birdsey @ Oct. 14 2004,12:02) |
| I do not buy gtr4him’s argument about the clocks being that far out of synch. Almost certainly, the clocks are phase locked to the bus speed of your computer (otherwise they would have significant problems transferring data to memory). This is likely to induce some timing jitter, some drift on the order of a few samles (=flanging), and some annoying problems with differing latencies. |
This is wrong. The soundcard’s sample rate is determined by either an internal on-board clock with its own crystal, or in the case of some better cards, by syncing to an external digital audio stream or ext word clock. I’ve never seen a soundcard without a crystal. (Could happen, I suppose, but I’ve never seen one)
Data bus transfers don’t have to be in sync with word clock, they just have to happen often enough to keep up with the sound card, so that its buffers aren’t starved or overflow.
| Quote (archimedes @ Oct. 14 2004,12:25) |
| Data bus transfers don't have to be in sync with word clock, they just have to happen often enough to keep up with the sound card, so that its buffers aren't starved or overflow. |
Right. The data buss basically has NOTHING to do with AD/DA circuits of the card. Different animals entirely.
What I said before was that no two cards will have EXACTLY the same clock frequency. Sure I'm talking about small differences. But they do add up and on longer recordings can cause problems. Why do you think professional studios spend massive amounts of money on a primo master clock and distribution systems?? Every piece of gear synched to ONE clock. Thats the ticket.
TG
Yeah, Ben – you were lucky. This forum has had plenty of posts from people wondering why their two soundcards drift out of sync as a song progresses. That’s why you always want to get a soundcard that accepts a sync signal from another soundcard.
Most soundcards that have a S/PDIF input will sync off that. In that case, if you get a second card, just make sure it has a S/PDIF output. I don’t know whether you can take the S/PDIF output on a Pod or something and Y-cable it to two soundcards, but it might be worth a try.
Sync issues are bad enough when using isolation booths. But you’d probably be recording in a situation where there will be a lot of bleed, so sync has to be even more accurate to avoid problems.
BTW, while it’s worthwhile pursuing what you’re thinking (recording “live in the studio”), I bet that eventually you’ll find that you get the best results just recording the minimum rhythm section first, and then overdubbing the vocals and other instruments. This way you get the best of both worlds: you get a nice “groove” going in the rhythm section, and you get the flexibility and control during the overdubbing of the other parts.
First, check out the proposed setup
<!–QuoteBegin>
| Quote |
|
- vocals: mixing board aux1 bus to the left side of analog input 1 - drums: mixing board aux2 bus to the right side of analog input 1 - bass and keyboards are sent from their individual mixing board channels to the left and right side of the 2nd analog input respectively - both guitarists use a Line6 POD XT which has a USB out |
The only two instruments on this list that need to be recorded acoustically are the vox and the drum overheads. If the rest are recorded as line-ins and everyone uses a headphone mix, there may be very minimal bled through. This would require the singer to be in another room and/or a improvised isolation booth.
Second, how much drift is the maximum that has been experienced? By my rough calculation, the absolute maximum you are likely to see is about 500ms over a 4 minute song (544ms = (441500Hz/44050Hz - 1) * 240s ), which IS a complete deal breaker. It seems much more likely that the uncertainty is at least a factor of ten below this (54ms = (44105Hz/44095Hz - 1) * 240s ), which is also a deal breaker for rhythm instruments.
If this is true, you’ll probably do best to record drums and bass on one card, and keys and vox on the other card. You can always cut and paste to resync the vox to the backing track, as the singer probably takes breaks to breathe. In any case, most singers do not have a precise sense of rhythm and will probably have to be aligned by hand anyway…
Third, the problems are probably MUCH less severe than those outlined above, as I never had any “time streaching” problems between the guitar port and various audio cards that were used to cap vox.
In fact, I used to use digital monitors as output, with the soundcard, and the guitarport as input, and there never seemed to be any significant “time streaching” issues even though all three digital devices are working off different clocks.
Again, what was the maximum measured error between the cards?
I’ve seen 50 to 100 msec by the end of a 3:30 song. That’s not just a dealbreaker for rhythm instruments, it also generally ruins piano and guitar parts (acoustic or electric). Hammond organ parts can usually tolerate it, especially if percussion is turned off. Pads & strings, no problem. Like Ben said, vocals are pretty easy to tweak if necessary.
The worst cases are generally when one of the clocks involved is the computer’s clock. Be sure to select “wave timer” rather than “system timer” – but you might try both settings just in case.
But save yourself the hassle and use gear that can sync up. Many soundcards can do this, even cheap ones. Generally not builtins, though.