[We received our first shipment of M-Audio MicroTrack 24/96 flash memory recorders this past Monday. I've been posting my impressions to the PenComputingAudio group on Yahoo, and thought that this groups might appreciate them too. -- LM]
As I work my way through what the MicroTrack 24/96 can do, I've found some more interesting and unexpected things:
First, as I noted before, its phantom power voltage is 30 Volts and not the usual 48. I haven't measured how many milliAmps per channel it can supply.
When recording an S/PDIF data stream, MicroTrack 24/96 is currently limited to making 16-bit recordings. That really clinches my initial impression that it's really a 16-bit recorder. Its Mic and Line inputs are limited to roughly 16-bit dynamic range (despite the option to record that to file in a 24-bit format), and the digital input is also 16-bit only. Perhaps the next firmware release will extend it to 24-bit?
Also, while the level meters operate when recording from the S/PDIF input, the headphone output (and presumably the Line outputs too, since they are probably driven from the same analog signal source) does not. So you can't monitor on headphones while recording from a S/PDIF source. (You can monitor on 'phones when recording from the TRS and 1/8-inch analog inputs.)
Another surprise is that MicroTrack 24/96 is currently a stereo-only recorder. It can not record in mono, though there are indications that it will be added in a future release of the firmware.
There are a fair number of bugs, most of which do not "sink the ship", but are pretty weird and annoying. Like what? Like imagine you recorded three files from the S/PDIF input. You simply can't get it to play back the first file in the file system. If you try to play back the first file, it simply won't play. If that weren't enough, it also causes a condition that you then can't play *any* of the files. But if you first play back the second or third file, it works fine unless you then try to play the first: that breaks it and you have to cycle power to get it back.
The workaround (not surprisingly) is to record a few short files and then delete the first file -- then you'll always be able to play the rest of the files.
Weird.
There are others like that. And then there are the ones that can sink the ship. Like the one that I haven't quite figured out yet. It makes the file system act like the CF card has no more space for recording. If you check the time remaining for recording, it's a negative number. But when you check the directory to list the files that are there, there aren't any files! If you've made a recording, it's gone and in some cases unsalvageable. The only way I've found to get back into business is to reformat the card in a PDA or a PC and start over.
Thanks a ton for this review. I've been waiting for months for this exact type of info.
Personally, I almost expected firmware bugs, that can be fixed in later revisions. What I'm mostly worried about is the sound quality itself. Any impressions on the overall sound quality as compared to the R1 or PMD660?
One major flaw seems to be the weak phantom power. IMHO, this could be grounds for a recall....while I realize that phantom power specs are often overrated by a few volts, there is a huge difference between 30v and 48v. I really doubt 30v would provide enough capacitance to properly drive most condenser mics.
Anyhow, any word on the overall sound quality, or better yet, some .wav or flac clips would be great! Thanks a ton Len!
> I tested a few production preamps for phantom power when > trying to make my own supply. They produced in the > neighborhood of 24 - 50v depending on what mic you > plugged in.
The spec for phantom power specifies a fairly-large current-limiting resistor - 5,600 ohms as I recall.
That means that a mic that draws say 1 milliamp, will have a phantom voltage at its input, that is 5.6 volts less than the open-circuit phantom voltage from an ideal phantom supply.
IOW, the range of voltages you found does not necessarily reflect on the mic preamps - they could all be stone perfect and simply performing within spec.
Well, the flash card was pretty much of a killer for me except as "one more recorder that I don't really need but it looks pretty cool" but it sounds like it's still a fur peace away from being ready for all but the most horny.
The phantom power voltage question came up on the Ampex list and someone reported talking to a product manager who confirmed that it was 48V. M-Audio has offered up other units with non-standard phantom powering. I wonder if it even powers their own condenser mics?
Oh, well. Another new product that I can ignore for a while.
Jonny Durango wrote: > Personally, I almost expected firmware bugs, that can be fixed in later > revisions.
Has our existence come to this, that we EXPECT bugs? Shame on you, Johnny! I guess we've pretty much come to expect bugs in software, particularly when that software isn't sure what environment it will be running in, but this is a piece of hardware that functions as a piece of hardware. It can, and should, be completely predictable. There's no excuse for any of the things that Len reported, except perhaps for the low voltage phantom power (clearly not a bug) which was cleverly absent from any of the advance literature.
Oops, there's a little software bug in your car. The anti-lock brakes lock up when you press Radio Button 5 when moving at speeds between 37 and 42 mph. How would that be?
spud wrote: > I tested a few production preamps for phantom power when trying to > make my own supply. They produced in the neighborhood of 24 - 50v > depending on what mic you plugged in. One mic runs off two 9v > batteries for many hours so I'm not sure how critical phantom voltage > really is. Be nice to know.
spud, as Arny Krueger pointed out there are dropping resistors in series with the phantom voltage supply, so it is normal for the voltage at the socked to drop a certain amount when a microphone is connected.
Since those resistors are supposed to be 6.8 kOhms and each one conducts half the current, the voltage drop should be 3.4 Volts for each mA that the microphone draws. Meanwhile, the overall tolerance limit is plus or minus 4 Volts. So for example, if a microphone draws 4 mA from the supply, 3.4 times 4 gives a 13.6 Volt drop across the supply resistors, and 48 minus 13.6 is 34.4 Volts. The voltage at the XLR socket should be within four Volts of that amount when the microphone is connected. (When the microphone isn't connected, it should be between 44 and 52 Volts, of course.)
If a microphone is plugged in and the resulting voltage falls below that range, the microphone's performance, especially its overload limit but also other parameters, could well be affected--in some cases quite severely. A supply that's only 20% below the tolerance range, for example, could cause some microphones to start clipping 10 dB sooner than their specs would indicate. That isn't something you want to find out during a live recording.
Certain types of microphone can tolerate a rather wide range of supply voltages. But most traditional, externally-polarized condensers can't, and that includes most of the really good ones.
>> I tested a few production preamps for phantom power when >> trying to make my own supply. They produced in the >> neighborhood of 24 - 50v depending on what mic you >> plugged in. One mic runs off two 9v batteries for many >> hours so I'm not sure how critical phantom voltage >> really is. Be nice to know.
> spud, as Arny Krueger pointed out there are dropping > resistors in series with the phantom voltage supply, so > it is normal for the voltage at the socked to drop a > certain amount when a microphone is connected.
> Since those resistors are supposed to be 6.8 kOhms and > each one conducts half the current, the voltage drop > should be 3.4 Volts for each mA that the microphone > draws.
Ahh yes, my error. The two resistors are in parallel from the standpoint of the phantom current.
>> I tested a few production preamps for phantom power when >> trying to make my own supply. They produced in the >> neighborhood of 24 - 50v depending on what mic you >> plugged in.
>The spec for phantom power specifies a fairly-large >current-limiting resistor - 5,600 ohms as I recall.
>>> I tested a few production preamps for phantom power when >>> trying to make my own supply. They produced in the >>> neighborhood of 24 - 50v depending on what mic you >>> plugged in.
>> The spec for phantom power specifies a fairly-large >> current-limiting resistor - 5,600 ohms as I recall.
Chel, by coincidence you examined two brands of microphone which allow for non-standard phantom powering. Please don't conclude, or lead others to conclude, that many other high-quality microphone manufacturers are following a similar approach. Some are, especially in electret microphones. But in general, most serious international manufacturers of traditional condenser microphones assume standard phantom powering according to DIN EN 61938 in their designs.
For example, in transformerless Neumann microphones such as the KM 180 series, a DC/DC converter is used to obtain the polarization voltage for the capsule. This circuit uses an oscillator whose frequency varies with the supply voltage. If the supply voltage is too low, the converter won't reach its intended output voltage. The sensitivity of the microphone will decrease along with the maximum SPL. Meanwhile the equivalent noise level rises, and the frequency of the oscillator itself can even fall into the audible range, or perhaps just into the range of some people's preamps and A/D converters if they go for extended frequency response and high sampling rates.
This isn't something to fart around with. If a microphone clips or is noisy during a recording, you're in serious trouble; why invite that? Why spend thousands of dollars on microphones and undermine them to save a lousy few bucks in the power supply? It's nice if a microphone can tolerate wider deviations than the standard allows, but only a power supply which conforms to the standard can be presumed to support all 48-Volt phantom-powered microphones correctly.
> I tested a few production preamps for phantom power when trying to > make my own supply. They produced in the neighborhood of 24 - 50v > depending on what mic you plugged in. One mic runs off two 9v > batteries for many hours so I'm not sure how critical phantom voltage > really is. Be nice to know.
Electret condenser mics use the phantom voltage only to run the impedance converter circuit, so most of them can tollerate a very wide range of voltage supply.
OTOH, the more "traditional" condenser mics use the phantom supply to polarize the microphone capsule. Lower phantom voltage directly translates to significant changes in how the microphone performs.
In rec.audio.pro Mike Rivers <mriv...@d-and-d.com> wrote:
> Jonny Durango wrote:
>> Personally, I almost expected firmware bugs, that can be fixed in later >> revisions.
> Has our existence come to this, that we EXPECT bugs? Shame on you, > Johnny! I guess we've pretty much come to expect bugs in software, > particularly when that software isn't sure what environment it will be > running in, but this is a piece of hardware that functions as a piece > of hardware.
> Oops, there's a little software bug in your car. The anti-lock brakes > lock up when you press Radio Button 5 when moving at speeds between 37 > and 42 mph. How would that be?
Better:
Vendor: "Hello ... is that Tim ... we're having a problem with your card payment, it seemed to go through, so we hiopped the goods; but when we look in out account, the money isn't there."
Tim: "Yes, that's a bug in my card payment software. The first payment gets lost; the second or third payment go through OK, unless you try to obtain the first payment, when *none* of the payments will go through ... oh, you know already?"
Mike Rivers wrote: > Jonny Durango wrote: > > Personally, I almost expected firmware bugs, that can be fixed in later > > revisions. > Has our existence come to this, that we EXPECT bugs? Shame on you, > Johnny! I guess we've pretty much come to expect bugs in software, > particularly when that software isn't sure what environment it will be > running in, but this is a piece of hardware that functions as a piece > of hardware. It can, and should, be completely predictable. There's no > excuse for any of the things that Len reported, except perhaps for the > low voltage phantom power (clearly not a bug) which was cleverly absent > from any of the advance literature.
Manufacturers bring this shit to market well ahead of when it's actually a finished product, and they rely on salivating indiscriminate purchasers to fund their bug-fix cycle. That's after failing to bother with proper phantom power supply design and implementation, or to have 24 bits in the device instead of just in the marketing literature.
Economies of scale suggest that nowadaze sucker births exceed one per minute, by a substantial margin; correspondingly, the price of sucker bait is falling precipitously as companies ramp-up to meet demand.
> Oops, there's a little software bug in your car. The anti-lock brakes > lock up when you press Radio Button 5 when moving at speeds between 37 > and 42 mph. How would that be?
Tim Martin wrote: > "Mike Rivers" wrote... > > Oops, there's a little software bug in your car. The anti-lock brakes > > lock up when you press Radio Button 5 when moving at speeds between 37 > > and 42 mph. How would that be? > Better: > Vendor: "Hello ... is that Tim ... we're having a problem with your card > payment, it seemed to go through, so we hiopped the goods; but when we look > in out account, the money isn't there." > Tim: "Yes, that's a bug in my card payment software. The first payment gets > lost; the second or third payment go through OK, unless you try to obtain > the first payment, when *none* of the payments will go through ... oh, you > know already?"
David Satz wrote: > Chel, by coincidence you examined two brands of microphone which allow > for non-standard phantom powering. Please don't conclude, or lead > others to conclude, that many other high-quality microphone > manufacturers are following a similar approach. Some are, especially in > electret microphones. But in general, most serious international > manufacturers of traditional condenser microphones assume standard > phantom powering according to DIN EN 61938 in their designs. > For example, in transformerless Neumann microphones such as the KM 180 > series, a DC/DC converter is used to obtain the polarization voltage > for the capsule. This circuit uses an oscillator whose frequency varies > with the supply voltage. If the supply voltage is too low, the > converter won't reach its intended output voltage. The sensitivity of > the microphone will decrease along with the maximum SPL. Meanwhile the > equivalent noise level rises, and the frequency of the oscillator > itself can even fall into the audible range, or perhaps just into the > range of some people's preamps and A/D converters if they go for > extended frequency response and high sampling rates. > This isn't something to fart around with. If a microphone clips or is > noisy during a recording, you're in serious trouble; why invite that? > Why spend thousands of dollars on microphones and undermine them to > save a lousy few bucks in the power supply? It's nice if a microphone > can tolerate wider deviations than the standard allows, but only a > power supply which conforms to the standard can be presumed to support > all 48-Volt phantom-powered microphones correctly.
Thanks, David. When one's recording fails, excuses for mediocrity in kit won't fix it.
And there's likely more to excellent phatom power supplying, as I heard David Jospehson at AES in SF last year ask Grant Carpenter about the noise floor of the Gordon preamp's phantom supply. Aaaaah, the details!
Chel, there is a subset of phantom-powered microphones which will work properly when the phantom supply voltages are non-standard. And I don't mean to downgrade the importance of AKG or Shure. Your microphones may indeed work with the piece of recording equipment which is the subject of this thread. If so, you may consider yourself lucky, if you like.
But then again, they may not. We have mainly been talking about open-circuit voltage, but some AKG mikes with the 9 - 52 Volt arrangement (such as the old, original C 451) draw as much as 6 mA apiece. I wouldn't assume that so much current is necessarily available for each microphone from this new device. I've seen plenty of portable preamps that can't even put out 4 mA per microphone input--the supply drops out of regulation first.
When the standard is followed, P48 microphones are correctly powered by P48 inputs, end of story. It is a very bad idea to release new equipment with non-standard phantom powering, because so many people have the types of microphone which depend on conformance to the standard.
but then we are talking about 350$ box with allot of compromise in between to build better powering circuit might be one of these things( with allot others ) which bring not develop the product at all might be that the mkh or sanken users wouldn't love it ,but if they can afford the mikes- they obviously can afford the dienekey ps-2 :-)
> When the standard is followed, P48 microphones are correctly powered by > P48 inputs, end of story. It is a very bad idea to release new > equipment with non-standard phantom powering, because so many people > have the types of microphone which depend on conformance to the > standard.
Chel van Gennip wrote: > This design, together with AKG's statement "many AKG condenser microphones > can be powered anywhere from 9 volts to 52 volts", make me think that the > 30V will not give problems for me. I do think Shure and AKG are "serious > international manufacturers of traditional condenser microphones"
Well, let us know what you think of your Micro-Trak when you decide to upgrad to something like a Shoeppes. But then there isn't much point to that if you're only doing 16-bit recording. I'm sure your AKG mic will do fine.
I'm not knocking your choice of mics, I'm knocking your calm acceptance of a device that limits your choice of mics. But then I guess I'm happy with a PC even though it won't run Mac software.
>> This design, together with AKG's statement "many AKG condenser >> microphones >> can be powered anywhere from 9 volts to 52 volts", make me think that the >> 30V will not give problems for me. I do think Shure and AKG are "serious >> international manufacturers of traditional condenser microphones"
> Well, let us know what you think of your Micro-Trak when you decide to > upgrad to something like a Shoeppes. But then there isn't much point to > that if you're only doing 16-bit recording. I'm sure your AKG mic will > do fine.
> I'm not knocking your choice of mics, I'm knocking your calm acceptance > of a device that limits your choice of mics. But then I guess I'm happy > with a PC even though it won't run Mac software.
>>>Personally, I almost expected firmware bugs, that can be fixed in later >>>revisions.
>>Has our existence come to this, that we EXPECT bugs? Shame on you, >>Johnny! I guess we've pretty much come to expect bugs in software, >>particularly when that software isn't sure what environment it will be >>running in, but this is a piece of hardware that functions as a piece >>of hardware. It can, and should, be completely predictable. There's no >>excuse for any of the things that Len reported, except perhaps for the >>low voltage phantom power (clearly not a bug) which was cleverly absent >>from any of the advance literature.
> Manufacturers bring this shit to market well ahead of when it's actually > a finished product, and they rely on salivating indiscriminate > purchasers to fund their bug-fix cycle. That's after failing to bother > with proper phantom power supply design and implementation, or to have > 24 bits in the device instead of just in the marketing literature.
> Economies of scale suggest that nowadaze sucker births exceed one per > minute, by a substantial margin; correspondingly, the price of sucker > bait is falling precipitously as companies ramp-up to meet demand.
Yep that pretty much sums it up....I'm afraid this is nothing more than a low quality-control consumer device marketted as a semi-pro piece of equipment. Sorry M-Audio, but my $ is going for an Edirol R1 or Oade PMD660!
Here are some noise specs....not very impressive and the left channel 1/4" TRS input behaves very strangely (again, poor quality control):
Yes, but some of the good ones have a DC-DC converter in there to step up the voltage on the capsule, so depending on how they are built a lower phantom voltage may cause anything from no difference to no sound.
|
