|
posted January 7, 2001.... updated April 30, 2008
For a long time I'd wanted to have a high quality microphone preamp box for USB. It seems like such an obvious idea... a high quality two channel mic preamp followed by with a 24 bit A/D converter and a USB interface for direct to disk recording. When Sound Devices released their USB Pre in early 2001, it seemed to be exactly what I had been looking for, so I bought one of the very first production models.
As is to be expected with a new product, there were a few little annoyances. Initially I was very impressed that Sound Devices acted very quickly and competently to resolve the issues that arose. And indeed, they did resolve the hardware problems, but unfortunately they seem either unwilling or unable to resolve some other software/firmware annoyances. But, above all, the USBPre sounds great! At 44.1 kHz sample rate, 16 bit stereo uses about 540 MB per hour, and 24 bit stereo uses a little over 800 MB per hour. Using Adobe Audition, saving the audio in 32 bit stereo for maximum precision gobbles up a little over 1 GB per hour, but with ever-falling price of hard drive storage the disk space is cheap. The USBPre mechanical and electronic design is really great. The USBPre controls, indicators, switches and jacks are all nicely arranged and easy to use. The aluminum knobs have a great feel. The sturdy aluminum case seems bullet proof. The analog specs for the mic preamp are very good. And it draws power from the USB port of the computer so there are no extra batteries to worry about. Sound Devices released an updated 1.5 version of the USBPre that includes coaxial S/PDIF input and output. For a modest upgrade price, they kindly took my original unit in trade and sent me a shiny new version 1.5. The S/PDIF is a great feature, now I can use my USBPre to download S/PDIF audio from my Sony PCM-1 portable DAT machine. Following are brief descriptions of some issues that I encountered while using the Sound Devices USBPre. 1.) Windows - older operating system limitations: Apparently Windows 98, 98SE and ME all have errors that prevent 24 and 32 bit audio data transfers via the USB. Microsoft made a half-hearted attempt at fixing the problem in Windows 98SE with a special QFE patch that you must beg for, and which could possibly cause other operating system problems. For all the details, see the Microsoft web page at:
http://support.microsoft.com/support/kb/articles/Q242/9/37.ASP Microsoft acknowledges that Windows ME cannot handle 24 or 32 bit USB audio, but they are apparently doing nothing about it. Arrrggg. Fortunately, the USBPre seems to work fine with Windows XP and Vista.
2.) Audio whine: There is something in the USBPre running at 1 kHz and you can clearly hear an annoying whine above the noise floor. Sound Devices acknowledges this problem and says that it is caused by the 1ms isochronous data transfer in the USB stream which is leaking into the audio. They do expect to have it fixed in all future deliveries, and they will upgrade the units that have been shipped. After upgrading to version 1.5, I'm happy to report that the 1 kHz whine has indeed been removed from the recorded audio, but unfortunately the 1 kHz whine is still present in the audio playback.
3.) 16 bit audio output: The Sound Devices ads frequently mention 24 bits, so that is what I naively expected everywhere. But alas, it is not to be. When you read the specifications carefully, they do mention that the USBPre is only delivering 16 bit resolution for audio out. It would have been really neat to have full 24 bit audio output. As it is, I can record 24 bit audio, but I have no way to hear the quality of the full 24 bit audio, since the USBPre D/A converter has only 16 bit resolution. That's a pity.
In some applications, I often use microphones that are not designed to withstand phantom power, such as the Audio Technica AT-822 stereo mic. With such mics, accidentally bumping the very easily pressed phantom power pushbutton on the front panel will instantly destroy the mics and put an end to the recording session. Such an accidental bump of the phantom power switch could easily occur in a typical live on-location field recording session. The USBPre switch settings for phantom power, input source and mono headphone mix can all be set using the front panel pushbutton switches or via a software control panel. So why not use the software control panel to prevent this problem? Or at least recess the switch to prevent accidentally applying phantom power unexpectedly? It would be nice if the USBPre software control panel included a provision to lock out any changes via the front panel pushbutton switch. It would be nice if each switch setting could be individually locked, but even a global lockout would be better than the present situation. There just needs to be some way that the appropriate phantom power setting can be established and then accidental changes locked out. In version 1.5, Sound Devices have added the front panel lockout to the USBPre software control panel, but unfortunately the lockout disappears off every time the computer is restarted... making the software lockout rather pointless.
When doing multitrack recording, it is important to get the new track properly aligned time-wise with the old tracks. Since the USB audio is only transmitted in bursts, and the computer has many tasks to perform, it is logical to expect that there will be some delay. To test the latency, I recorded a click (or hand clap), and then connected the output of the USPPre back into its input, played back the click track through the USBPre and recorded it as a second track in Adobe Audition's multi-track editor. On my old 450 MHz Pentium III machine, there was typically 18 to 20 msec of latency between the original test signal and the newly recorded track. On my current Dell 8400 computer (3.2GHz) , the latency is only about 3msec. With Adobe Audition, there is an input setup menu where the 3msec latency can be entered, and then Audition automatically adjusts the tracks to be synchronized. Sound Devices now has an ASIO driver to further reduce the latency, but I've not tested the ASIO driver.
The mechanical and electronic design of the USBPre are very nicely done. The USBPre is very rugged, and the controls have a very nice feel. Unfortunately, from my experience, the USBPre software is not in the same league. I had several problems with the USBPre software on two older computers, but on my current Dell 8400 the following mysterious problems are now resolved... so maybe the problems were not entirely due to the USBPre software. With two of my older Dell computers:
The USBPre hardware and electronic design is masterfully done, and
fortunately their software seems to be compatible with my current Dell 8400, so
there are no complaints about the software when using this particular computer. 7.) Capacitor Failures: After about a year and a half, the phantom power blocking capacitors failed in both channels of my USBPre. Sound Devices repaired the USBPre for free and and it continues to work fine. It seems like a flaw in design or component selection when capacitors
fail in both channels at about the same time. I asked if they
had taken any measures to avoid such failures in the future, but they
never replied. 8.) Newer Products: There is no doubt that the USBPre is a great product. The USBPre is sturdily built like a tank, and has great quality mic preamps, but you may find it useful to look at some some of the newer products from well respected companies such as M-Audio, Echo, Focusrite and PreSonus (see also my review of the PreSonus AudioBox) However, if you simply want something that is simple, reliable and rugged, the USBPre is
a winner... although it is rather expensive. The Sound Devices USBPre is a great product, the hardware is well designed, the product is well built and most importantly it sounds great.
|
