I've used my R-4 twice now. But a bit of background. I've got many years of recording experience, always with two mics (and long before digital!). Have gone from Sony PCMF1 through DAT, Alesis Masterlink (ML9600), and now also have an R-4. I used to use Schoeps mics, now use Earthworks for about 7 years (incredible mics). I have recorded many concerts (classical/choral type, not rock) and have released about 30 CDs (see
http://www.dtrmusic.com for a list and audio samples).
I firmly believe that the most important quality aspect is to record at 24 bits. Great improvement in quality at any sampling rate and when you get to the final (CD) stage at 44.1/16 bit you are only at that quality level as the very last step.
The first time with the R-4 I recorded at 44.1/24 bit and had no problems. It was a quiet concert (16 voice madrigal choir, no instruments), so the levels were at about 12:30. Ran on AC power because the 1.5 or so hours off of the AA batteries doesn't cut it. The sound on the recording was very good.
The second time was actually two performances on the same day of the same concert, a continuous 1 1/4 hour performance. I was recording these at 96kHz/24 bit and was recording 4 channels in a surround sound setup with 4 phantom powered Earthworks mics. The first concert recording was a disaster because of Edirol R-4 power limitations. Between setting levels at the pre-concert practice and the start of the concert the external power plug had fallen out of the jack on the R-4, so I was (unknown to me) running on batteries. After about 18 minutes I saw the "low battery warning" flash. An odd message on AC power until I discovered the "non connection". I had read in the manual that you can't switch power sources, but took a chance (unfortunately) and plugged in the AC power. The R-4 immediately shut down and lost all of the 18 minutes. I then had to restart it and recorded the rest of the concert without incident.
I consider the fact that the R-4 does not seamlessly switch between battery (internal) and AC (external) power sources inexcuseable. Laptops have done that for many, many years. If the R-4 worked as it should then you could have batteries in the unit as a backup power source, so if the AC/external was lost it would work from the internal batteries, and vice versa. Obviously simple technology these days.
After being distressed by the loss of part of the first performance (the first time in many, many years I ever lost part of a concert, and this was a paid job), I was in a major panic during the second performance when after just over one hour I noticed that the big time display said about 3 minutes, rather than 1 hour plus. Then after the performance was over I tried playing back (keeping my fingers crossed) and could only play the last 10 or so minutes of the concert. Found out that the first part was saved in separate files, but for some reason they were not initially displayed in the find function. (I think I corrected later this through a system playback option.) I've been told it's somewhere in the manual, but I hadn't seen it, that the max file size is only 2 gig (which is just over an hour at 2496 stereo) and that the R-4 automatically switches to a new set of files. I was relieved when I found the recording was complete, but I think that users should be made very aware that this will happen and isn't a cause for panic.
Nick (nickspicks) has mentioned that the quality of the mic preamps leaves much to be desired. I think that while they can be improved, it is critical to avoid recording with the level controls set too low. I had to set the levels for recording the two performances, which involved a 150 voice choir, pipe organ, and brass group, to below 9 o'clock (below the 2nd tick mark) and I also used the limiter in the R-4. Later I learned that the limiter acts separately on each channel and therefore should be avoided. Nick mentioned that the line/mic switches just insert a 23 db pad early in the input chain and will work even when using phantom powered mics. It would have been better to select line on the input switches and then set the levels on the pots more in the 11 o'clock to 3 o'clock positions to be in the midrange of the preamplifier's amplification range (where the quality should be better and you have more control over the actual level).
Sometime I will try to record using external Earthworks mic preamps that truly are top notch and use both the R-4 and an Alesis Masterlink for the recording to compare quality.
I also think that Alesis needs an additional display mode where the entire width of the display is used for the 4 input meters (and, obviously, there are no output level meters) and the meters then use the added width to display more detailed levels. They currently are marginally useful for setting levels.
Be aware that Alesis reverses the polarity of the DC input jack from the "standard" if you get an external battery pack.
I consider the carry case useless except to transport the R-4 itself (with no AC power adapter, spare batteries, or anything else) and not something where you can run the R-4 while it's in the case.
The R-4 provides a lot in a small and inexpensive package. The inability to switch power sources automatically is a big failure. The meter displays need the additional option I described.
Bob