dmp3 vu meters - is it really peaking?

[cellardweller]

Okay, I got my DMP3 last week, and it seems that my VU meters are kinda erratic.
I have it going into my fostex MR8, and others who have used this combination have not had to use the MR8 pre at all.
When recording a single 12, close miced with a 57 and an e 609s, the VU meters peg out, but the peek light only blinks and does not remain lit.
So is it truly peaking, or is this the 4dB “window” prior to clipping?

Maybe the simplest way to phrase it, is there a “standard” method of optimizing the signal coming out of the dmp3?
Is it possible I need to engage the low-cut filter *shudder*?

 

[distortedrumble]

i'm thinking that the peak light is the amout of input on the signal and the vu meter is the level of output.

 

[MadAudio]

It's not the low cut filter you should worry about, but the button above that one - the gain range. What do you have that set for now? In or out? I use my DMP3 with a pair of MXL 603's as drum overheads, and that output is hot hot hot! I actually have to use a Rane ME15 to attenuate the signal.

More importantly, though, how does the recording sound? If it sounds good, it is good!

 

[scrubs]

Meters and lights are only part of the picture. Your ears are the other. If the clip light only pops on occasionally, it's probably o.k. I don't even pay attention to the VUs, because they seem pretty erratic on mine, too. Listen back to the track and check for any nastiness. If you are getting noticeable clipping, even with the gain knob down, then engage the Hi/Lo gain button, as madaudio suggested.

 

[cellardweller]

If you are getting noticeable clipping, even with the gain knob down, then engage the Hi/Lo gain button, as madaudio suggested.

I've only really used the high gain setting (button out), because I could not get the signal hot enough without blasting myself out of the room on low gain setting...
mics are approx 1-2 inches from speaker.... It'd be my luck I got a lemon. It certainly doesn't seem as hot as I've heard madaudio and others say... but maybe I was just being "VU paranoid" and not actually getting it hot enough..

argghhhhhhhhhhh

 

[cellardweller]

More importantly, though, how does the recording sound? If it sounds good, it is good!

There is no noticable clipping, but my concern is having to use (still) the MR8's pre. I was hoping to get away from that, since by all accounts the dmp3 is "hot".

I abused the guitar up to the point of the peak led coming on, and then cut back. I was trying to emulate the hardest I would be picking...
but still the input meter on the MR8 was approx 75%...

I just hate bumping up those nasty mr8 pres.....

 

[baekgaard]

Well...

This sounds rather strange. I have a DMP3 as well, and it is certainly a hot thing...

I'm running it into my sound card, which is at nominal -10 dBv level, and on my setup, I can hardly approach any levels on the VU meters before it severely overloads the inputs.

If I take my SM58, put the gain on hi (out position) and the knob at around 3 o'clock, I get around 0 dB on the VU meters at normal speech levels close-miced (like you normally use a SM58 on stage). If I raise my voice, the VU meters peak, and the clip light comes on from time to time, but only momentarily.

You should get similar results with your SM57 -- which I think is what you say also?

However, if I run that into my sound card, it overloads drastically. I need to back down to the 12 o'clock position, at which the VU levels read around -10 dB. I now get peak levels just below 0 dBFS (i.e. I'm close to clipping on the input converters in my sound card), but the reading of e.g. wavelabs VU meters now also hoover around -10 dB.

This is what I would expect also. The DMP3 is calibrated to output +12 dBv (RMS, I think) when showing 0 on the VU meter (for a 1 kHz sine tone) -- which is very hot! It means the peak level will be at +15 dBV, and then the overload margin is still up to 7 dB before maxing out at the specified +22 dBv (18 Volt!). This is balanced output, and since I'm running it unbalanced, I loose 6 dB. But still, to get it down to the max input level of my soundcard, I need to back down further to avoid overloading.


Note that there is a difference in the way a VU meter and a peak meter works. A VU meter averages the music, and the reading you get is more related to the loudness of the music. A peak meter (here: the LED that flashes) shows you when even a single peak of the waveform is getting closer to clipping. You can have one take sound at e.g. -8 dB VU without peaking and another sounding at -10 dB VU but peaking from time to time, and they may both be roughly equally loud sounding.

So, peak levels often read more than 10 dB hotter than the VU reading.

Furthermore, the DMP3 VU meters are output meters, and the clip LED reflects input level (after the preamp). So, you should make sure the LED only lits up momentarily (which it does 4 dB before "soft" clipping in the intermediate gain stage), and use the VU meters to check the overall levels.

If your recording device is spec'ed at +4 dBu, then you should still have plenty of headroom, but I guess you would need to have the gain knob set at 3 o'clock to make it hot enough.


Here's some reading to do on a long and sleepless night

http://www.digido.com/portal/pmodule_id=11/pmdmode=fullscreen/pageadder_page_id=36/
http://www.digido.com/portal/pmodule_id=11/pmdmode=fullscreen/pageadder_page_id=59
http://www.soundonsound.com/sos/1996_articles/may96/metering.html


So much for the long winded theory...

I assume you have tried this with only one mic at a time also?

And it is a problem with both of your mics?

Are you running balanced on unbalanced from the DMP3 to your Fostex? I assume you're using a line input. If there is a gain stage/knob to twist, make sure it is set at line level (which it normally is if set around 2 o'clock on many devices).


Hope this helps explain more than it confused,


-- Per.

 

[cellardweller]

If I take my SM58, put the gain on hi (out position) and the knob at around 3 o'clock, I get around 0 dB on the VU meters at normal speech levels close-miced (like you normally use a SM58 on stage). If I raise my voice, the VU meters peak, and the clip light comes on from time to time, but only momentarily.

You should get similar results with your SM57 -- which I think is what you say also?

I haven't tried vocals with my SM58 yet, but if I remember correctly I was in the 3 o'clock area with a SPB1 on vocal... I will check it with the SM58. Hopefully I will have an opportunity tonight.

If your recording device is spec'ed at +4 dBu, then you should still have plenty of headroom, but I guess you would need to have the gain knob set at 3 o'clock to make it hot enough.

I assume you have tried this with only one mic at a time also?

And it is a problem with both of your mics?

The two mics (sm57, e609s) were very similar in relation to the position of the gain knob on the dmp3. The 57 was a bit hotter than the e609s. Both VU meters exhibited the same "spastic" behavior. I'm really beginning to think I should pretty much ignore them when recording distorted guitars, and mind only the peak...

Yes +4 dBu. Read response below from fostex tech.

Maximum Input level of the MR-8 is +2dBV which is +4dBu.

So you should be able to connect the Mic Pre to MR-8's input, You just turn
down the TRIM knob all the way down to LINE.

Best Regards,
Matt Suzuki
Fostex America Product Support.

Are you running balanced on unbalanced from the DMP3 to your Fostex? I assume you're using a line input. If there is a gain stage/knob to twist, make sure it is set at line level (which it normally is if set around 2 o'clock on many devices).

Unbalanced, as the MR8's 1/4 inputs are TR/unbalanced. It does have XLR inputs... should I purchase a TRS and an XLR adapter? Could this make a difference (+6dB???)

Thanks for the info and the reading, I've saved it for a long boring night, like tonight may end up being

 

[cellardweller]

TRS to XLR cables made a world of difference.

Now if I could just make the guitars sound as good on my monitors as they do in the cans when I'm tracking...

 

[baekgaard]

TRS to XLR cables made a world of difference.

Now if I could just make the guitars sound as good on my monitors as they do in the cans when I'm tracking...

Good to hear you fixed the issue.

Yes, everything usually sounds nice in cans and less so in monitors. But I guess that when you're finally happy with the sound in the monitors, it will sound good in most people's speakers, and even more so in your cans ;-)

Rgds,


-- Per.