Behringer USB guitar

[Steve Conner]

I guess I'm getting old, but this just seems WRONG

http://www.behringer.com/IAXE393/

They completely forgot the "Big wooden crate covered in tatty beer stained vinyl that gives out a lovely smell of hot tubes" part of the instrument

At least they left a proper 1/4" jack outlet and didn't put the USB socket in the actual Fender jackplate thingy.

[Mark Hammer]

It's not THAT wrong. Lots of folks want to record their guitar direct to digital and a USB port on the guitar itself helps in doing so. The trouble is that, unlike having a separate interface box for analog (I've got one of these myself: http://www.roland.com/products/en/UA-30/index.html ), where the user can stick whatever the heck they want in between the signal source and input jack on the adapter/convertor box, the Behringer requires the user to rely on software FX exclusively....unless they want to spring for their own separate convertor box and use THAT. That means that you BETTER like the software emulation of a Fuzz Face because you won't be able to use your own vintage one and simply rely on the digital EQ and reverb.

Certainly, the headphone jack is a nice addition though, and the price is a painless way to get into software guitar processing and modelling. I'm curious about whether the guitar would permit replacement of pickups or even whether the mag pickups serve as the signal source.

[Sock Puppet]

Well it looks like a good idea...

I've tried the NI combos and their Guitar Rig, they're lovely to play with for a time - so's the DSP stuff on my E-MU sound card but having spent the day with my 15 year old nephew setting his guitar up and listening to his 'sound bites' through his new modelling amplifier - yes it had a valve in! - I have a headache and a craving for a clean guitar sound (a touch of reverb) or a power cut.

The computer processing completely strips character from the guitar and with it pickup tone et al. I'm not knocking effects at all, I've built many (didn't box many:-)); the ancient pedals had the effect of adding character or uniqueness and required a bit of forethought to chase a sound that was in your head.

O.K. it sounds like I'm ranting but you weren't there. I'll stop.

S.

[Enzo]

MSRP only $149???

I looked on my dealer site, and the tech data for this is not up - no schematics.

[David Schwab]

I was just looking in Musician's Friend at this last night. I also thought it was a dumb idea, if only because even though I usually do plug my guitar and bass directly into my mixer when recording, I really don't think I want to use that USB interface... how good could it be on a $150 guitar?

I think USB is a bad idea for audio anyway.

Regarding those modeling amps... I think the big problem is people end up using the factory presets it came with, and that's never a good thing!

I've played around with a Pod a little, and once I got past the factory presets, it had some potential.

I actually use a Roland VM-3100Pro digital mixer to interface to my Mac (via S/PDIF), and that has built in effects, and amp modeling.

It took me a while to find it, since the documentation is so bad, but it actually has some decent amp models, such as a Jazz Chorus, Clean Twin, Matchless, 6160, etc. You can tweak the hell out of them, and I actually have gotten very good tones.

This surprised me because I dislike those Pod things as a general rule... just makes everything sound too processed.

I was recently playing in a band, and one of the guitar players used a Variaxe and a Pod, and ran the whole thing into a Powerbook running Guitar Rig. I could never hear him... everything was always so distant sounding.

[Mark Lavelle]

Originally Posted by David Schwab

I think USB is a bad idea for audio anyway.

Why? All the USB does is move the data once it's digitized.

Any reasonably up-to-date host system with USB 2.0 has more than enough bandwidth to handle stereo audio at any common data rate & sample size (but I'll grant that serious multitracking isn't likely to be well supported).

[David Schwab]

Originally Posted by Mark Lavelle

Why? All the USB does is move the data once it's digitized.

Any reasonably up-to-date host system with USB 2.0 has more than enough bandwidth to handle stereo audio at any common data rate & sample size (but I'll grant that serious multitracking isn't likely to be well supported).

Well you answered the question. Even with USB 2.0, you really don't have that much sustained bandwidth as you have with Firewire 400. And Firewire 800 is even faster.

USB can get bogged down with audio pretty quickly.

[Mark Lavelle]

Originally Posted by David Schwab

USB can get bogged down with audio pretty quickly.

If you want to send high-quality multitrack data, but I thought we were talking about a single guitar...

[David Schwab]

Originally Posted by Mark Lavelle

If you want to send high-quality multitrack data, but I thought we were talking about a single guitar...

Right, but if you are using USB for audio, chances are you have an USB audio interface to compete with.

I'm sure one guitar via USB is fine... there sure are a lot of USB audio interfaces out there, I just wouldn't use any.

[Steve Conner]

I have some experience with digital multitracking, and I kind of agree with David. USB doesn't seem that well "sorted" for multitrack audio.

For ages, I used a 600MHz PIII with a M-Audio Delta 66 for my audio work. The Delta 66 is a PCI card, and so can more or less just suck the audio data out of the PC's memory when it needs it, while the CPU is getting on with something else. This system worked great with low latency. I recently upgraded to a 3GHz P4 with a Delta 1010, and it more or less just works hassle-free.

I also played around with a Roland UA-1EX USB audio dongle (with its "Advanced Driver") on an oldish 550MHz PIII laptop with the same BX chipset (so USB 1.1 only) as my old desktop. I had to use a lot more latency, getting on for 100ms, and even so I got occasional clicks and pops. However I have managed to use it for live mixing with an old version of Traktor DJ. I found if I play arty enough electronic music, people think the glitches and pops are part of the tune.

I assumed that the extra latency I needed for stable operation was down to the overhead of the USB connection.

[David Schwab]

Originally Posted by Steve Conner

For ages, I used a 600MHz PIII with a M-Audio Delta 66 for my audio work. The Delta 66 is a PCI card, and so can more or less just suck the audio data out of the PC's memory when it needs it, while the CPU is getting on with something else. This system worked great with low latency. I recently upgraded to a 3GHz P4 with a Delta 1010, and it more or less just works hassle-free.

I use an M-Audio Audiophile 24/96 PCI card. I had the Delta 66 once, but at the time it didn't work well with the computer I had.

I only play one track at a time (I multi track everything) so the 24/96 works fine.

Originally Posted by Steve Conner

I found if I play arty enough electronic music, people think the glitches and pops are part of the tune.

I've gotten some of the coolest noises out of my Mac when I've pushed things to the point where Cubase has become locked up and has this glitchy loop going. I'm going to have to try and record some of that stuff one day.

Originally Posted by Steve Conner

I assumed that the extra latency I needed for stable operation was down to the overhead of the USB connection.

USB was made for mice and stuff. My very first audio interface was a USB device (because oddly enough my G4 has no audio inputs, even though it's the "Digital Audio" model! Go figure!). The latency was really bad, and it had no headroom, since it was powered off the USB buss.

I feel safest with PCI cards, and after that Firewire, but even that can get glitchy if you are running a hard drive on the same Firewire channel.

My latency is so low that I can monitor everything through Cubase if I want, and you can hardly hear a delay. I monitor off a external mixer though.

[Steve Conner]

Yay, another M-Audio Believer! I got the Delta 66 because it was the only multi-channel soundboard I could afford at the time, and the 1010 because I liked the 66 so much, and my local music store was selling an ex-demo unit cheap. I guess the only problem is that it doesn't have any ADAT I/O or anything, but it's not like I have anything else with an ADAT or Tascam connection to plug it into.

I agree USB was made for mice (didn't it start out as a souped-up version of the Apple Desktop Bus?) but they did a pretty good job turbocharging it to similar speeds as Firewire with USB 2.0. I don't know if they upgraded the suspension and brakes to cope with the extra power though

It would be interesting to know how much latency the Behringer USB thingy has. I'd also love to know latency figures for the Variax's onboard electronics and the Line6 modelling amps. Sound travels about 1 foot per millisecond, so 10 ms is like playing from the end of a 10 foot guitar cord, 100ms is like playing it from half a block away. You can imagine that ramming the guitar into the speaker to get feedback isn't going to work the same as with an analog amp.

The band I play in recently went to a local studio to record an EP. The studio boasted of their vintage analog desk but admitted they used PC-based hard disk recording. I got talking to the engineer about the monitoring setup, and he admitted that all the foldback we heard (our own amps and suchlike coming back through our headphones or the control room monitors when tracking) was going through the digital gear with 64 samples of latency. That's only 1.5ms though and nobody noticed the extra 18 "inches" of delay.

I see a few makers have brought out USB condenser mics now, so it'll only be a matter of time before some marketing weenie says "Wouldn't it be great if we made a USB mixing desk". And you thought the pro audio industry had hit rock bottom already!

[Mark Lavelle]

Originally Posted by David Schwab

USB was made for mice and stuff. My very first audio interface was a USB device (because oddly enough my G4 has no audio inputs, even though it's the "Digital Audio" model! Go figure!). The latency was really bad, and it had no headroom, since it was powered off the USB buss.

Originally Posted by Steve Conner

I agree USB was made for mice (didn't it start out as a souped-up version of the Apple Desktop Bus?) but they did a pretty good job turbocharging it to similar speeds as Firewire with USB 2.0. I don't know if they upgraded the suspension and brakes to cope with the extra power though

Excuse me while I fly my USB geek flag:

I've been working on USB products (and some of the standards committees) since before you could buy it, and I can state categorically that it was not "made for mice!" We (Logitech [where I work], Microsoft, Intel, Compaq, HP & others) were talking about audio, video and all kinds of heavy lifting from the very beginning. The original goal was really to replace nearly all the connectors on the back of your PC with just the one kind (and we may yet get there).

The USB protocol and USB hardware (at least some) has been good enough for glitchless high quality audio from Day 1, with the only real limitation being bandwidth (12Mbps in the beginning, now 480) - and that only limits the number of simultaneous channels. Unfortunately, the Windows driver architecture at the time (Win98) was not up to the task, and didn't become sufficient until XP came out, so a lot of people got a bad taste before then. With the advent of USB 2.0 and the newer high speed root hubs, there is really no good technological reason why USB audio today should be any worse than Firewire or even bus cards.

I'm not going to try to tell you that every USB audio product out there lives up to the promise (duh!), but I'll bet you will see very good multitrack USB mixers (or at least 8-in/8-out I/O interfaces) within a few years...

[Steve Conner]

Yay, a real USB geek! I have a question for you that I always wanted to know. Do USB controllers support DMA? That is, can the USB host controller stream incoming audio data straight into the computer's memory (and the other way too I guess) without using any CPU power? I always got the impression that they didn't, or maybe older chipsets didn't, or only USB2.0 does it, or maybe they all do but older versions of Windows don't support it or whatever.

[Mark Lavelle]

Originally Posted by Steve Conner

Do USB controllers support DMA? That is, can the USB host controller stream incoming audio data straight into the computer's memory (and the other way too I guess) without using any CPU power?

All USB host controllers (even 1.0) use DMA.

I don't know all the details (I'm more of a peripheral firmware & comm protocol guy), but the problem with streaming data (i.e., audio & video) in Windows before XP had to do with the way the various software components (multiple levels of drivers, plus the apps) communicated with each other. The hardware has certainly improved since 1996, but the really huge difference is that WinXP uses it much more efficiently than the older versions...

[To pick a nit: While individual DMA transfers can occur without intervention from the CPU, continuous data streams still need to be managed by software.]