Korg MR-2000S Review

Publish date:
Social count:
Image placeholder title
Image placeholder title

Korg's new 1-bit stereo recorder, the MR-2000S, looks almost as good as it sounds.

In the majority of current studio setups, the concept of the mixdown deck has fallen by the wayside. Whether we're summing within a DAW or mixing through analog gear and printing the mix back into the same session, most of us find ourselves operating without a dedicated device with which to mix. Despite that trend, Korg offers the MR-2000S ($2,499), a 2-track digital recorder with an 80GB internal hard drive and a variety of format options, including the increasingly popular 1-bit (DSD) format.

A Panel Study

The MR-2000S is housed in a single-rackspace chassis. The display section has an easy-to-read 160×104 backlit LCD, offering visual feedback for menus and file status. Small square LEDs indicate the sample rate (from 44.1 kHz to 5.6 MHz),clock source (internal, S/PDIF or word), analog reference level (-12 dB to -20 dB), and internal disk activity. Two bright 23-segment LED bar graph meters indicate left/right signal levels, from -54 dB to +3 dB, with a peak LED beyond that. (The +3dB spec may seem odd; I'll explain it later.) Controls include a rotary encoder and nine buttons for menu navigation, transport control, changing the menu display and peak/hold clearing. A standby/on switch and headphone jack/volume pot round out the front panel.

Among the rear-panel connections are balanced analog XLR I/Os, RCAs for unbalanced analog I/O, coaxial S/PDIF I/O and word clock I/O on BNCs. The S/PDIF jacks double as ports for linking multiple units for multitrack DSD recording. A USB 2 port connects to a Mac or PC for easy file transfers to/from the unit, with the MR-2000S showing up on your computer desktop like any other FAT32 hard drive for drag-and-drop file copying.

Less is More

The MR-2000S can record/play quite a few audio formats. It can record “regular” PCM files in the BWF (Broadcast WAV) format from 44.1 kHz to 192 kHz, and at the lowest two sample rates it's 16/24-bit-selectable. Above 48 kHz, 24-bit is the only option. As for 1-bit recording, I'll explain it briefly now (and at emusician.com/online_exclusive/korg_mr2000s_review_bonus) for those of you new to the concept. Most of us have been trained to think that higher bit rates are better, yet if the sampling rate is high enough (in the millions of samples per second range), that one bit can make accurate calculations when turning an analog input voltage into a digital datastream.

Three 1-bit formats are available on the MR-2000S: DSDIFF, DSF and WSD. All are slight variations on the DSD (Direct Stream Digital) theme, and each can run at the standard 2.8MHz or the doubled 5.6MHz sample rate. If you wondered why the meters go to +3 dB, it's because DSD doesn't hard-clip the way PCM does when it reaches 0 dBfs. DSD exhibits a softer clipping that sounds more similar to the way analog tape saturates, so the meters reflect this saturation region with an extra few LED steps.

Call Me a Convert

Image placeholder title

FIG. 1: AudioGate's user interface is simple and effective. You can use your mouse's scroll-wheel to scrub your audio.

Now, what to do with those 1-bit files once you record them? Well, unless you are authoring SACDs yourself or sending the 1-bit files directly to a mastering engineer (more mastering engineers are accepting DSD files all the time), you'll convert them to PCM at some point. Korg includes the handy and easy-to-use AudioGate application, which doubles as an audio player interface, for this conversion process (see Fig. 1). AudioGate lets you play any 1-bit PCM or MP3 file from your Mac or PC, as well as from the MR-2000S directly when the latter is mounted on your computer as an external hard drive.

Eds. Note: According to Korg, a recent AudioGate update also lets you create 2.8MHz DSD disks on standard DVDs that can play on some Sony Vaio PCs, PlayStation 3s and a few high-end DVD players, such as Sony's SCD-XA5400ES.

Blind Leading the Blind

Of course, there is real-time conversion happening if you play a 1-bit file through your normal computer DAC. Therefore, when listening this way, the quality will be less pristine than when doing a nonreal-time conversion export.

I was curious to hear how the sound of the MR-2000S would compare to that of my standard setup. My normal mixdown method is to use Pro Tools like two tape machines — I split its outputs across my console and then run the stereo bus of the console back into Pro Tools via Lynx Aurora converters, which are clocked to an Antelope Audio OCX. For these tests, I clocked the MR-2000S internally, although it can be clocked to the OCX or any other word-clock source. I recorded the mixes off the 2-bus of the console into Pro Tools and the MR-2000S in 5.6MHz DSDIFF (.dff) mode simultaneously, and then I ran the same material into the MR-2000S in 24-bit PCM (.wav) mode at the same sample rate as the original session file (44.1 kHzor 96 kHz). I then used AudioGate to convert the 1-bit file to 24 bits at the appropriate sample rate and visually lined all three stereo files up to the sample in a single Pro Tools session for easy A/B/C comparisons. I sent the session to a group of bat-eared engineer friends, and I listened myself in a double-blind test so I wouldn't know which file was which. The results were quite interesting.

At 44.1 kHz, the file from the MR-2000S was similar to the one from the Aurora, with the latter having a slightly tighter bass response; the Aurora's version sounded a little more mastered in the low end. The difference between the 1-bit file and the two PCM samples was more evident: There was general agreement that the 1-bit had more high and high-midrange energy, and was less thick in the low-midrange. There was also more audible decay on the cymbals and snare drum in the 1-bit version, almost as if it were slightly compressed. Opinions varied as to which sounded better, based on the listener's taste, but overall the 1-bit file garnered more votes because of its slightly more finished sound. Very little difference in the stereo imaging was detected.

Interestingly, at 96 kHz the differences between each of the two PCM versions were less detectable, but the differences between 1-bit and PCM were more pronounced, with the DFF file exhibiting a slightly wider stereo field, as well as a more open, airy sound. In a word: more analog-sounding. I contribute this partly to the fact that conversion to 96 kHz retains more of the resolution of the 5.6MHz/1-bit file. Also, the program material was less dense than the 44.1kHz tests (spacious, dynamic acoustic music vs. loud rock music), so the expanded dynamic range of the material let us hear more subtleties of variation between the differing capture formats. If you're working with more dynamic music at higher sample rates, you should benefit more from the MR-2000S' 1-bit conversion. But either way, capturing your mixes at the highest possible resolution has its benefits, archiving them for a time when some form of 1-bit delivery medium becomes more widespread. In the meantime, if the downsampled PCM files sound as good or better than your current capture process, what do you have to lose besides a little hard drive space?

Not 1-Bit Disappointed

In all, I found the build quality and sonics of the MR-2000S to be of the highest caliber; even the integrated headphone amp has a clear, rich sound. Although I would prefer more choices in organizing the files on the hard drive, the process of accessing/renaming files through the menus on the display is quick and easy. In designing the MR-2000S, Korg upgraded the analog signal path of its own MR-1000, which is a cheaper, more mobile 1-bit option. If you're ready to add 1-bit recording to your studio, the MR-2000S is a solid choice.

Eli Crews operates New, Improved Recording (www.newimprovedrecording.com), a studio in Oakland, Calif. Special thanks to Myles Boisen, Christian Hanlon and Jay Pellicci for their help with this review.

Image placeholder title