18 replies to this topic

[Juan22]

Hello everyone, to keep it short I need to know if my current md gear is suitable for the technical demands of bioacoustics analysis. Frog calls (my subject of study) are usually in the range of the 20 Hz - 20.000 Hz. I was offered a tape recorder but doing technical research I found it to have a frequency response of 250 Hz - 6.300 Hz. I've downloaded the service manual of my Sony MZ-R70 and its frequency response equals that of the suggested range (+/- 3 dB) HOWEVER, I would like to know what range of frequencies are being manipulated to achieve the 292 kbps compression of the atrac format. Some frogs call in ultrasound, mind you.

thanks

JD

[Juan22]

anyone? at all?

[sfbp]

Sorry, but I think the problem is with the way you asked the question. It's not true to say that a "range of frequencies are being manipulated".

As long as you stick to SP (no choice with an R70), you should equal or beat the quality on a standard CD. The more likely problems are:

1. What microphone you use (many will cut out at 30, 40, 50 Hz), and whether the fancy mics will work in the microphone socket of the MZ-R70. You may need a device called a battery box to transform the signal from the fancy microphones to line in voltage levels, instead of the microphone input, due to the signal levels involved.
2. How you plan to get the results digitally if you need to do scientific analysis. Just about the only reasonable way to get the recorded sounds from an SP minidisc is to buy or rent an MZ-RH1 and upload the SP minidisc to a WAV (16 bits, Fs=44100) file. Anything else you are likely (but not guaranteed) to lose the upper and lower ends just by transferring it (you can test this if your equipment is good enough, on a scope or whatever) on an analogue cable.

All in all, you're the scientist, and will have to establish your own means and methodology.

The one thing you do NOT have to worry about is the compression - ATRAC does a very good job on that score. What it lacks in overall bits it makes up in bit depth. ATRAC is a 24-bit format with a logarithmic exponent (floating point) whereas Linear PCM is a 16-bit fixed point format. So for very loud and very quiet sounds, ATRAC will probably work better than one of the devices that merely records 16-bit "LPCM".

Hope this helps.

[Juan22]

Thanks sfpb, first of all I am not a scientist just yet! but your last paragraph alone answers my question. I don't use a fancy microphone (sony ecm-ms907 or something like that) and I digitize the analog minidisc recordings using audacity at 16 bit 44.1 kHz and then analyse the calls using fancy software from Cornell University's ornitology lab. which lets me calculate all the meassurements I need (peak frequency, begin time, end time, entropy etc.) I might add my recordings with the mz-r70 have turned out very 'scientificly acceptable' and i'm writing a paper regarding neotropical frog sistematics which is soon to be published.

thanks again for your input.

JD

[A440]

MS907 is not a great microphone.

Its frequency response is very deficient in bass and superhigh treble. T he bass is more serious. There are two octaves of human hearing below 100 Hz.


http://reviews.cnet....7-20037945.html

The R70 will grab good sound, but you ned a good source.

[Juan22]

I am having some problems when I switch the recording mode from SP stereo to mono on the MZ-R70. There are times when the recorder seems to identify silence while in reality the source is emiting sound. When I'm digitizing the recording, whose length could be 10 minutes, those sound blanks make it shorter, some times loosing up to 2 minutes of sound because it pauses constantly. This doesn't happen when I record mono on my other MD, MZ-N707.


 

[sfbp]

Looks like an intermittent microphone socket to me.....

[Juan22]

Do you think dirt inside the socket could cause this? I no longer have the unit so I can't check but I would just like to identify the source of the problem to avoid it in the future.

[sfbp]

Jim Hoggarth probably has an opinion. I will second guess him and suppose that you have a dry joint on the 3.5mm jack socket... that's quite easy to happen on any of the 3.5mm sockets. More likely than dirt.

[Juan22]

ok, how are dry joints fixed?

[sfbp]

With a soldering iron (and in this case, if possible, a microscope), and a LOT of skill.

[jim.hoggarth]

Yes, it could be a bad contact or cracked solder joint around the microphone input. Dirty contacts is fairly common, bad joints not so but if the input has had a lot of use? That said, I would have expected a more chaotic signal than clean on/off as the graph. Unfortunately there is no indication of time duration on the x-axis. Also seems odd that the problem is reported in Mono, not SP Stereo.

 

A staccato signal would be the result of the laser not being able to read the track reliably. A constant signal would be played back with intermittent blank sections, which could range from milliseconds to several seconds each time. But then, the playback time would be extended to include the blanks, rather than reduced. We can presumably rule this out.

 

Could it be the output of the R70 that is intermittent? Have you tried both earphone outputs? I am assuming you are taking the recording from one or other of these sockets, into the line input of your PC card. Have you tried the same recorded disc in the N707, does that produce the same inconsistency? In which case it is definitely the recording at fault, not the playback. Can you be sure the microphone does not have a bad connection in the cabling? Are the discs you are using good quality and in good condition, ie not dusty and scratched?

 

Final question: Are you absolutely positive the [problem never happens in SP, only Mono? If the latter is true, I would reckon the problem is in or around the encoder, which means a main board swap.

 

Sorry for the ream of questions, it's just easier for diagnosis if all the possibilities are dealt with in one go.

Jim

[sfbp]

Comprehensive analysis, Jim - I would expect no less! You're right, if OP doesn't report whether "normal" recordings have the same fault, we have no idea if it's a playback issue.

Also I know that cables for PLAYBACK with non-gold contacts (eg many headphones) can be really awful. So maybe the dirt theory isn't so far wrong in the first place, if one thinks non-gold==dirt What sort of cable are you using, Juan?

Edit: I just thought of something. This could easily be a Windows problem. If you look at your volume control in Windows (for recording inputs) you may find there is (or can be - see advanced properties on the mixer) an input called "Mono In". It might be be that you need to select this to prevent the input from going crazy when it gets no signal from one channel.

[Juan22]

Hello Jim, I've uploaded a new screen capture with the time line on the x axis. Unfortunately I didn't try the second headphone socket and since that unit was stolen from me I can't check. I'm sure the microphone is working properly as I constantly use it with the N707 (the connection is in deed gold). The disc was in good shape, HOWEVER It did combine both stereo and mono recordings, something I stopped doing after this incident. I never had this problem using the R70 on SP Stereo (using the same mic, disc, battery etc.) so I reckon this would only happen on mono. I can't try to digitize that recording on  another unit because the disc was in the recorder at the time I was robbed.

Hi sfbp, I digitize all my recordings using a standard 3.5mm audio cable (non-gold) but this has never been a problem in the past (nor now).

thank you both for you input!

JD

[sfbp]

I'd still be inclined to suspect the line cable. One way to resolve this would be to see if you can hear with your own ears the "bad" sound from the mono recordings, using the headphones and using the line cable routed to an amplifier. I would always avoid a PC for making direct recordings, personally. Latency is always an issue, as well as pickup electronic noise from the other components plugged into the (PCI) bus. Here's the picture of my Volume Control:

 volcontrol.JPG   42.5K   6 downloads

[Juan22]

In deed sfbp, I can hear the 'blanks' sending the sound from the computer to my stereo's line in and connecting my headphones to the computer as well.

[sfbp]

In deed sfbp, I can hear the 'blanks' sending the sound from the computer to my stereo's line in and connecting my headphones to the computer as well.

But that is AFTER you transferred to the PC.

Can you hear the blanks with headphones plugged in the recorder?

[Juan22]

You mean I should record the recording using the MZ-N707? I no longer have the original disc in which the recording was made, so the only way to listen to it is via a computer. I guess I could re-record using the 707 but I'm guessing the blanks will be recorded as well.

[sfbp]

I mean, you should take headphones and plug into whatever machine produces these blanks and see if there is a problem with the recorded sound, straight from MD -> headphones.