Topic: Preamp noise comparison of field recorders (Zoom H4/H4n and Edirol R-4)

[Walstib62]

I've had the H4n for about 6 months. I haven't had any noticeable issues with noise or distortion. I do have to wonder if there isn't some sort of defect with the unit being tested.

[refrain]

I bought my H4n in July, and i've recording all sorts of ambiences (some with very low amplitudes), a lot of concerts (most of them acoustic and very near silence), sound effects, mostly indoors, all that with the onboard mics or 4 channel recordings with a MS pair of AKG blue lines using the onboard matrix and P48V. And I haven't found any of the troubles you're describing... i never used the TSR line inputs, always the XLR inputs... I usually record with volume rec set at 30/40, sometimes even lower, 15/25...
I only have to say that going from the MT2496, a Canford audio MS preamp and a pair of AKG in MS style is a big plus, very small setup, I used the sony pcm D50 a couple of times, but i really prefer the H4n onboard mics and the XLR/4 channel set...
Batteries is another department, using 4 Channel or even XLR/48v is very demanding for a couple of AA batteries...

CS

[chris319]

A 10kHz sine wave at 44.1kHz is always going to "look" jagged

Ach! I misread 10kHz as 1kHz. Yes, it will look jagged at that frequency.

I would suggest running Rightmark Audio Analyzer on this unit to get some meaningful numbers.

[mdopp]

Martin -

WRT your distortion tests, there appears to be something fundamentally wrong with your methodology. I examined the wav file of the H4n you posted and those are not sine waves at all. See for yourself. Visually they are jagged and show a great deal of quantization error. Properly-recorded sine waves shouldn't look like that.

You say "rec-level 40"; if you are using the H4n's continuously-variable gain setting, I believe that reduces the gain in the digital domain only and may account for the quantization error you are seeing. Try recording at the maximum (digital) level setting. On the H4 this was 127 IIRC. The H4 had a slide switch which I believe was a three-position analog attentuator. Set that for maximum attenuation/minimum sensitivity, then bring down the level of your CD player accordingly. Examine your recordings using just about any audio editor to make sure the sine waves themselves are unclipped and undistorted.

What can be wrong of recording a pure sine tone from a CD-player into the line inputs ?
I have done the same with the Edirol and the Sony and the results were excellent.
And yes, the recorded signal of the Zoom is not a pure sine-wave. It is horrible. That is my point ;-)
BTW, it stayed horrible even when I reduced the output level of my CD-player to the lowest setting (way below normal line-level) ...

I assume (!) the rec-level setting of the H4n reduces the signal in the analog domain. If it wouldn't work analog and just digital (as in the H4) the distortions would have an obvious explanation. But at the same time that would mean the Zoom H4n is completely useless as you could not reduce the (analog) input sensitivity (the H4n does not have attenuator switches like the L/M/H-switches of the H4). I don't believe this is the case.
Furthermore, you also have the blinking red clipping warning lights. They turn off when reducing rec-level. This is another hint that the rec-level works on the analog signal before digitalization.

Do I have a faulty unit? Maybe, but it doesn't seem likely to me. The distortions of this device creep in at higher frequencies. A casual listener will not notice them - at least not instantly. I did some music recordings and they sounded quite normal at first. The difference only became apparent when comparing to the original (CD-recording). Suddenly the sound was harsh and unpleasant.
This is just like stuck pixels on a computer monitor. You can use it for a long time without ever seeing it. But when you noticed it just once your eyes will always be drawn to the defect.

[fotoralf.be]

Do I have a faulty unit? Maybe, but it doesn't seem likely to me.

So why not?

I think that, pending further clarification, you should limit your verdict to the only established fact and that is that *your* H4n might be having a problem.

Ralf