"is the additional 22 dB dynamic range offered by the SD702 going to make a significant difference while recording anything but the most soft sources?"

Perhaps not. What it helps in is allowing you to do two things:
- capture things that vary in volume (soft things will not disappear in the noise).
- lessen the need to be as exact setting volumes (more room between noise floor and clipping).

If at all possible I think you should try out the units you are considering before buying, with the actual microphone you are planning to use. It is quite a difference, but we react differently to it. Personally I am sensitive to noise, after trying everything then on the market, the SD 722 was my choice. My ears are satisfied, but there was quite a bit of bill to pay.

Rumours I hear has it that the fr2le is a good, solid useable unit. The mr1 to me looks in a different league, below my cutoff point, the big-brother mr1000 I have seen and I believe I could live with that one (suspicious about the file format though). Long time since I saw the R4 though, remember it as useable.

Sorry if this does not help in any way in your selection process.

Gunnar

I'm in dispute of their spec's on this one. On the MR1000, they quote 96dB SNR. My own measurements do not bear this out. I'm only basing this on a 1 kHz tone produced at the input at great enough gain to cause a level of 0dBFS. Upon doing this at the lowest gain setting, I'm getting S/N ratios that greatly surpass 100dB:1.

It's only when I greatly increase the gain on the line inputs that I get a higher number, but then I'm clipping by several dB which would mean...I should lower my gain again and that would put me back in the better figures again.

I've got several questions for the folks at Korg and hope to talk to them soon.

ghellquist wrote: the big-brother mr1000 I have seen and I believe I could live with that one (suspicious about the file format though).

What are you suspicious about?

Perhaps I can help.

the big-brother mr1000 I have seen and I believe I could live with that one (suspicious about the file format though).

actually the most important reason why i'm even thinking about the korg unit is the sound format. I'm very curious to fing out whats behind so many people saying that korg's implementation of DSD sounds way more like "analog" than PCM... in general, i'm too jaded to believe anything (thanks to years of preamp hype at gearslutz), but this "analog" connection makes me weak. I think that music itself has entered its most steep downward spiral thanks to the advent of digital, but thats a whole different can of worms.

If at all possible I think you should try out the units you are considering before buying, with the actual microphone you are planning to use. It is quite a difference, but we react differently to it.

ofcourse you're absolutely right, but the closest i can get to actually auditioning any of this stuff is at pune, where the only other indian contributor to this forum resides, and pune is way across the country.

though the whole thing is a bit hypothetical at this stage (i'm currently firmly rooted in "commercial" studio work right now), the sd302+mr-1 offers me extreme portability, relative economy, no redundancy and the all critical (for me) ability to use three mics. I have no doubt that the 702 is a better built, more solid, robust and serious recorder, and that 112 dB is certainly *better* than 90 dB... a car that can do 120 mph stably is most likely a *better* car than one which can do only 80 mph stably, but here in channai, and for my driving style, i have never exceeded 60 mph... :)

i'm glad that so far, nobody has said that "90 dB dynamic range is totally unuseable", which leaves me with some hope for an economical solution :) lastly, and tho' not in context, i must say that if the 744t had 4 pres instead of two, i'd have given up the search long ago...

thanks for the responses,

90dB is about the ideal max for 16 bit recording - there has been plenty of commercial stuff recorded using 16 bit converters. I don't know fer sher (and I'm too lazy to look it up), but I would expect the s/n of professional analog tape to be significantly less than 90dB. Granted, with tape you can record hotter than 0dB.

Again...I have to stress -
With the gain set at the appropriate level for +4 input (which was a difficult task to begin with - Shame on Korg for not marking some indicator as to what is nominal for a balanced input!), the noise level was well below -96dBFS. My meters barely registered noise at all (-120 dBFS). Only when I cranked the gain did this change.

I think they measured this incorrectly - perhaps they simply recorded the inputs at full gain with no input signal and used that as the basis for their measurement (which is around or a little higher than -90 dBFS).

By the way -
I just recorded a Beethoven Festival with the Korg MR1000 and nothing more than the Royer SF12 and the Millennia HV3D and I'm LOVING the sound. I'll see if I can bounce some of the Octet down to 24/192 and post it. It will obviously be a VERY large file size, so I'll keep the length short.

It really does sound that good.

thanks cucco, but you are talking about the mr-1000 which has a quoted dynamic range of 96 dB, while the mr-1 has a quoted dynamic range of 90 dB. I hope that you are right and that the same is true of the mr-1.

BTW, i read that proaudioreview has "bench tested" the mr-1000 at 96 dB, but i have not seen the review personally.

i must say tho' that the 22 dB dynamic range difference between the mr-1 and the sd702 is still a scary number :?

I'd be curious to see PAR's review. I just re-ran the test (set the input level to the level which would equate to unity based on the output of a +4 signal along balanced lines equalling 0dBFS on both the MR-1000 and the PC) and the recorded gain with signal present was -0.0 dBFS, without signal present (and a LPF set at 99kHz) the meters would only show me an RMS of -120 to -118 dBFS over the average of 60 seconds of audio.

I know we're talking two different machines, but I would assume they'd both be similar.

Cucco wrote: [quote=ghellquist]the big-brother mr1000 I have seen and I believe I could live with that one (suspicious about the file format though).

What are you suspicious about?
Perhaps I can help.

Well, I can probably borrow MR1000 from a friend and test it side by side with my 722. Come to think of it, maybe I´ll make some three way XLR-s and test both next to the HV3D/Lavry Blue combo. Might test each with its own preamp or all through the same preamp.

Anyway, I am a lazy person and expect it to be quite a bit of extra hazzle with the DSD files. Difficult to do the postprocessing on them. And not worth it once processed down to 44.1/16 on a CD.

I also like the monitoring options on the 722, MS decoding as one example, that are not on the Korg.

Gunnar

In answer to the question posed as the subject of this thread, I'll go out on a limb and say that Mr One's 90dB s/n ratio is *sufficient*, but I use that word in the same context that I use the word *acceptable* - yes, it will do the job with no serious complaints, but you could do better.

Certainly, Mr One represents good value for the money. There are few acoustic spaces to make and/or playback recordings that can match that kind of S/N ratio, and even fewer styles of music that require it. So, if you got your recording levels right, it wouldn't be a problem.

I'm more worried about the complete system because a system is, of course, the sum of its parts. The SD302 and SD702 will integrate together beautifully, outputs to inputs, nominal levels and so on. You'll have a very professional kit that is rugged, sounds good, is designed to work together (one sits on top of the other, bringing all controls and meters to one common front panel surface with one common operating philosophy) will probably never let you down, and (when you've recovered from the initial outlay) will be something you'll be very pleased to own. Also, the 702 will allow you to plug two microphones directly into it, so you end up with five balanced microphone inputs, if necessary. Or, for simple direct-to-stereo jobs, using it alone will suffice.

But integrating the SD302 mixer with Mr One worries me a bit. Actually, it worries me a lot. The SD302 has balanced XLR outputs, while Mr One appears to have only minijack inputs. I would expect all kinds of troubles with overloading the analog inputs of Mr One on a regular basis - you may find that you have to run the SD mixer below its nominal level to avoid giving Mr One a level hernia, possibly resulting in more noise and certainly constraining the quality the SD302 is capable of. (One work-around would be to use in-line pads between the SD302 and Mr One...)

I also think that you'll eventually tire of Mr One; you'll be using a very professional piece of mixer, that does everything it is supposed to do in the manner you expect it to do it, with good solid and reassuring XLR inputs and outputs, nice LED metering that can be seen in any light, and all the necesary controls readily available on the front panel. Alongside that, Mr One is going to look and feel like a toy.

Simmo, two long replies of mine have been timed out to this slow bloody wireless laptop connection.

In the meanwhile, for a musician-engineer like me who does all of his work indoors, everything has changed. Logic 8 is here and its $500. This is a landmark day in the history of music recording and production, in the digital domain. Apple is going to own the market from now on.

Macbook + Apogee Duet + Logic 8 for a two preamp, two channel solution (US$2.1K)
Macbook + Apogee Ensemble + Logic 8 for a 4 preamp, 16 channel solution (US$3.25K)

The only thing to be worked out is powering the Ensemble in remote locations, like that church I mentioned a while ago, but i will find a way to do it :)

In case you did'nt know, Logic has long since incorporated such luxuries as linear phase equalization, multiband compression and impulse response reverberation into its feature set. Only, until now, for a multitude of reasons, it did'nt make sense for me to switch.

there is no other way to go for a poor boy like me who wants it all and wants it cheap :) Luxuries like the beautiful 702 and the apparently "analog sounding" korg units are for rich white people or those who record outdoors. (no offense intended, please).

damn. i just invested in a decked out quadcore PC, 'else would have hopped to Logic right away!

anyway, i'm taking the evening off too drool over the new possibilities

:D

Well...I've done a little more testing and have discovered why the disparity in the measurements.

It is true that the total band measurement is in fact 96dB SNR. However, I think this is selling this unit short since most every single bit of that noise is well outside the audible band. At higher Sample Rates (such as 192kHz or DSD), the noise is not even captured by Sequoia's excellent FIR filters (but it is captured in Algorithmix's ReNOVAtor - only shown as a line of signal across the top.)

At lower sample rates, (specifically those putting the noise in audible range such as 44.1 and 48), the noise is more of a dither pattern spread out across the spectrum but mostly concentrated at the extreme low end (sub sonic) or near the top end.

When a good filter is used, you can bring this noise level (specifically on the higher sample rates) down to that of other similar gear (in the 110-120 dB SNR range).

On a side note, I ran duplicate rigs the other night splitting the feed from my preamp - one portion went straight to DSD (double rate) and the other to PCM 96kHz. I've listened to the two and while I was very pleased with the sound of the 96kHz recording, it's plainly obvious that the DSD recording is simply easier to listen to.

Where the differences are most obvious are in the decay of reverbs and resonances of instruments. Whereas in PCM, the decays tend to have what I call a zipper effect (almost like the sound is moving down in amplitude at marked intervals) there is no such effect through the DSD. The marvelous thing is that much to all of this is translated into the PCM bounce down (though sadly not MP3 which happily mangles the signal no matter what the source).

I've noticed this most on the following instruments (specifically their releases) in order of obviousness:

Triangle
Suspended Cymbal
Trumpet
Violin
Piano
Trombone
Oboe

These are the ones where it's so silly obvious, it's as if you were getting slapped in the face with the differences.

Other differences seem to include a faster attack on low mid frequencies. Timpani seem so much more present without the need for extraneous spot mics.

The mic preamps on the MR1000 are usable for high-sensitivity mics, but I wouldn't use them for anything where absolute precision was necessary. I could very easily see myself using them for recording interviews, seminars, on-location sound FX, etc.

I do agree with Simmo - the MR-1, to me, seems more "toy like" and I too fear the minijack for anything serious. (First of all, it's a pain in the ass to make any cables that terminate in minijack and I don't know of any that are commercially available that I'd really trust.)

Anyway...I'm done now (for the moment.)

Cheers -

J

rfreez wrote: Macbook + Apogee Ensemble + Logic 8 for a 4 preamp, 16 channel solution (US$3.25K)

Now you're talking about a decent and versatile system that is well suited to your needs. Add the Milabs to that rig, with their 12dBA self-noise and 21.5mv sensitivity, and you're doing very well.

Hopefully, the laptop's physical noise (fan, hard drive) won't be an issue.

rfreez wrote: The only thing to be worked out is powering the Ensemble in remote locations, like that church I mentioned a while ago, but i will find a way to do it :)

That won't be too hard; a 12V sealed lead acid battery, a sine wave inverter, and a mains-powered battery charger will do the job. The battery only needs enough Ampere/hour capacity to power the Ensemble for as long as the battery on your laptop lasts - no point going beyond that. (Er, unless you want to work for longer than the laptop battery, in which case you're going to need a bigger battery capable of powering both systems.)

In my opinion Logic has always been a fine program, with very good sound.

rfreez wrote: Luxuries like the beautiful 702 and the apparently "analog sounding" korg units are for rich white people or those who record outdoors.

I resemble that statement...

But seriously, I wouldn't be considering those things as 'luxuries' in comparison to the system you're now talking about. More convenient, yes, but in terms of sound quality and what they'll let you do, I think they're quite comparable. Also, the laptop-based system will let you do multitrack overdubbing and so on, which the other rig won't, so it opens up many more possibilities.

Simmosonic wrote: [quote=rfreez]Macbook + Apogee Ensemble + Logic 8 for a 4 preamp, 16 channel solution (US$3.25K)

[...]In my opinion Logic has always been a fine program, with very good sound.
Appologies if this is thread hijacking, but the remark about Logic cought my attention.

How is Logic for the purposes most often presented in this forum where we record a few, up to many simultaneous tracks, often a few takes and then want cut and splice bits and pieces from the takes as if they were virtual multitrack tapes?

L

larsfarm wrote: How is Logic for the purposes most often presented in this forum...

I have never operated it before, but this is what I can say:

1) I have worked on multitrack projects alongside other engineers who have used Logic (typically, I do the miking and so on, and they do the recording, overdubbing and editing), and have always been impressed with what it could do and how it sounded.

2) I know of two engineers who use Logic with Digidesign hardware, claiming that it sounds much better than Protools (hence the change to Logic software - they already had the Protools software, of course, because it came with the hardware).

3) There used to be a lot of very loyal Logic fans, before Apple bought it from Emagic. I can remember some of them, PC users, expressing utter outrage when they learnt that Apple had purchased the program and future releases would only run on Macintosh. Some of these guys were devastated, and felt that Emagic had betrayed them and Apple was forcing them to move to the Macintosh platform. That's business, of course, but it's hard to convince a loyal PC-based Logic user of that. It's interesting that a program could inspire such devotion.

4) I don't know if this is applicable to the latest versions, but the earlier versions had some kind of programmable user interface, which allowed you to design your own interface depending on what you did: direct-to-stereo, multitrack, mastering, etc. The learning curve was steeper, but the end result was something highly customised to your needs. One of the guys mentioned in point (1) above is/was a master of programming Logic, and created (what looked like) entirely different programs for his multitrack and mastering work.

That's about all I can tell you. It's not much, but I hope it will be enough to draw any Logic users out there into the conversation - even if it's just to correct and/or contradict me. I have no problem with that...

Jeremy,

What do you mean by "PCM bounce down?" Is this a Korg internal process, or are you able to edit DSD?

srs

Cucco wrote: The marvelous thing is that much to all of this is translated into the PCM bounce down (though sadly not MP3 which happily mangles the signal no matter what the source).

So, a good way to perform AD-conversion is to do it in two steps. 1) One-bit DSD followed by 2) DSD-PCM-conversion. If this is sonically superior to what ordinary, or for that matter quite expensive AD-converters do, why don't they do that directly in the box in order to achieve this better quality PCM?

L

srs wrote: Jeremy,

What do you mean by "PCM bounce down?" Is this a Korg internal process, or are you able to edit DSD?

srs

Yes and no. In a different post (the one about micing an orchestra) I went into a little detail. I've written some software which will allow me to perform edits that "appear" and are monitored via the PCM track but take place on the DSD track (allowing me to do basic crossfades and trims without having to upgrade to all DSD converters but still avoids downsampling to edit and then re-upsampling and re-downsampling again).

If I can get all of the bugs worked out, I'll share as freeware, but I'm having a b*tch of a time right now getting it to be stable. However, in the example of the Beethoven, yes, I've done just that.

I use the Korg software to perform the downsampling from DSD to PCM though as I'm not even about to try to get that to work! What kills me is that the AudioGate software actually does have a DSD editing function, but they don't capitalize on its abilities. It can already split and combine DSD files, therefore, it should be able to do this editing.

larsfarm wrote:
So, a good way to perform AD-conversion is to do it in two steps. 1) One-bit DSD followed by 2) DSD-PCM-conversion. If this is sonically superior to what ordinary, or for that matter quite expensive AD-converters do, why don't they do that directly in the box in order to achieve this better quality PCM?

Yes and no.

You'd have to compensate for the extra latency if doing it live. Also, you'd have to build in some kind of filter which people would get all uppity about. I don't think manufacturers would touch this with a ten foot pole.

Although, many folks do use a 1 bit Delta Sigma converter anyway, so it's not too far of a stretch to imagine keeping it in the single bit domain until the very end...I suppose.

Part of the problem is, there's some demystifying which needs to take place before the manufacturers and the general public would buy into it. Also, even if conditions allowed DSD for everyone, it certainly isn't the best choice for every project. I'd shutter to think of 64 tracks of DSD with comps/limiters/EQ/reverb all going at the same time to record a crappy garage band. I'd think that CPUs would begin contributing to global warming then...

Forget about the MR-1 and go straight to the MR-1000.

The MR-1 has no functionality with its 1/8" inch inputs and it's compromised analog section.

I have found that the MR-1000 is suitable for high quality work and the DSD sounds very good. It does show a high "pleasant-ness quotient" in its sound.

I would not describe it as more "analog-like," but edges and harshness are shown less than in cheap digital pcm. There is very little difference, however, between the MR-1000 and high quality pcm digital.
(+$6000-8000 a/d cost.)

The MR-1000 is a great machine.

Plush wrote: There is very little difference, however, between the MR-1000 and high quality pcm digital. (+$6000-8000 a/d cost.) The MR-1000 is a great machine.

I'll take that as high praise indeed...

Do you know how much recording time it offers when running from batteries?

I have received about 4 hours running time on batteriesin the MR-1000.

Continuous running time is under 3 hours.

It uses rather inexpensive AA batteries, so it's no big deal.

At Costco, I can buy 48 AA batteries for $9.50.

I run the MR-1000 alongside my Nagra V and then I compare which I like better. It's not always the Nagra and it's not always the Korg.

Both are high quality.

8-)

Plush wrote: I run the MR-1000 alongside my Nagra V and then I compare which I like better. It's not always the Nagra and it's not always the Korg.

Very interesting, thanks!

Plush, what is your take on the quality of the MR-1000's preamps as compared to the Nagra V?

Duckman wrote: Plush, what is your take on the quality of the MR-1000's preamps as compared to the Nagra V?

And continuing the Plush 'Mr One Thousand' questions:

When you run the Korg MR1000 and the Nagra V together, how do you feed signals into them to ensure they're both getting the same signal to make a valid comparison? Are you using mic splitters? Or driving both from a mixer, line level?

Are you able to do a side-by-side comparison of their preamps?

The recording to both units is made from the line output of a Neve console. The same analog out signal is going to both recorders. Then the comparison and judgement about fidelity is made two ways. I listen on headphones switching back and forth between the Nagra and the Korg. Nagra is 24/96 and the Korg is the 5.8Mhz. DSD format. Often the two units sound about the same. Sometimes I prefer the Nagra and sometimes the Korg.

I have also compared the Korg to the Nagra at 24/44.1.
I play both through our SADiE here. The DSD files from the Korg are converted to 24/44.1 by the Audiogate soft. Then I switch back and forth from the Nagra and the Korg file.
The 5.8 Mhz. DSD file converted to 24/44.1 performs very well indeed. The DSD format is particularly well suited to string ensemble recordings.

The mic amps on the two machines are totally non comparable. The Nagra has Nagra mic amps and the Korg has Korg mic amps. The Nagra V mic amps are very good indeed---in the tradition of Kudelski.

The Korg mic amps are somewhat limited in their dynamic range and are noisier than the Nagra by a good margin. I feel that the Korg box is made to be used line in. There is nothing more to say. You get what you pay for.

The Nagra V remains our main two track machine. That will not be changing here. However, a laudable DSD machine for
$1200 is a no brainer!

Run it along side your set-up and send in your comparisons.

Plush wrote: The recording to both units is made from the line output of a Neve console. The same analog out signal is going to both recorders.

Thanks for that information, Plush. It answers all of my questions while also confirming my suspicions about the mic preamps. The world is round again!

If I get a chance to play with Mr One Thousand, I'll report here...