The steps in which the volume can be adjusted are rather strange. Here is a graph:

The x coordinate is the volume, the y coordinate is the difference between this volume and the next step up.

The right half and the left quarter of the graph are fairly regular, but the interval between -45 and -26 looks completely random.

Why can't we have 0.5 increments all the way?

Borislav

There was a thread about this a few months ago. The volume steps reflect a rough scale of human perception of sound levels. If you ignore the numbers, do you have a problem with how the volume scale sounds? You should find that it sounds rather linear and predictable.

Rob

Hi Rob.

I do agree with Borislav that the volume adjustment graph he posted shows what I hear when adjusting the volume on my empeg: The volume adjustment is not linear in terms of dB nor in terms of human audibility. If it where, the graph Borislav posted would either be flat (linear on dB) or had a log/exp scale (I just can´t concentrate enough right now to tell which one).
The human audibility is roughly a doubling of the "felt / heard" volume on an advancement of 10dB. So the volume I hear (y) is about 1/2^(-/10), or the volume setting for any given factor y is about -ld(1/y)*10.
As we should all know, ld (log(x)/log(2)) is a continuously ascending function, so Borislavs graph should also have some continuity, no changes from ascending to descending and back.

cu,
sven
(MkII 12GB blue, #080000113)

The volume steps reflect a rough scale of human perception of sound levels.

The comment about the seemingly random fluctuation is fair, though. Supposing (just supposing) the human perceptual response DID look like that. The exact positions of the little peaks as heard by the listener would depend on the gain of the amplifier. If I chose to have my amp gain a bit different, then the peaks of the empeg would line up with the troughs in my ear/brain combination. This would equate to me pressing volume up a lot of times for no perceived increase in loudness, and then all of a sudden I'd press it once and the music would seem much louder to me.

Now, I don't believe the ear's response does have all those sharp peaks, but I'm happy to be proven wrong. But either way, because of the unknown amplifier / listening environment gain, it really only makes sense to have a very smooth curve, I reckon.

Richard.

If you ignore the numbers, do you have a problem with how the volume scale sounds?

No, I don't, although the steps are small enough that it's kinda hard to tell.

I had an email response from john@empeg that mentioned "lack of coefficient accuracy in the DSP at low volume levels" which would explain the steep part of the curve. That's fair enough. Otherwise, I agree with Richard and Sven that a monotonic curve is the only thing that makes sense.

Borislav

would there be any thing wrong with making it a log response so that one full turn of the dial goes from off to 0db?

the remote and panel buttons may be stepped differently however.

i find that even with the acceleration, it still takes 3 fast turns of the dial to turn the sound right down (not off).


____________________
Murray 06000047

I had a bit of a play around with a very high quality sound level meter recently. My mother is an audiologist, so I get nagged to have a hearing test every now and then. Needless to say I have a slight hearing loss around 4-6K, I can now adjust the EQ to cope not only with the acoustics of the car but my own hearing response, but I digress…

The meter measured both dBA and dBC. As I understand it dBA is the standard for measuring loudness with respect to the human ear. dBC is similar but used for high levels i.e. 90-140dB.

Anyway I found several tracks that I though were representative and tried to correlate the measured level and the –dB level displayed on the empeg. The results where interesting, basically as close as I could tell the scale on the empeg maps directly with the dBA scale.

I ended up using 3 tracks and did the test a few times. It’s not that accurate because the reading fluctuates up to 10dB while you are watching it but you good at averaging it. Also I didn’t go into the overdrive range because I didn’t want to damage my equipment, (or myself). Here are the, averaged, results.

empeg (-dB), level (dBA), permanent hearing damage after (min)
0, 114, 1
-10, 103, 15
-20, 93, 150
-30, 84
-40, 73
-50, 59

It drops off a bit at the low end, but that might be speaker response.

Another interesting observation was that bass thumps didn’t influence the level as much as I thought it would. The most representative and loudest track had a frequency range that is all over the spectrum (Killing Heidi – Class Celebrities). There’s one to chew on for all the SPL lunatics with a dozen 15” subs. Actually I don’t know how they measure level in SPL competitions.

For what it’s worth I like the empeg scale as it is.

Actually I don’t know how they measure level in SPL competitions.

I can only speak to IASCA competitions. Other sanctioning bodies may do things differently.

The microphone is placed in a designated location, something like 4" from the windshield post on the passenger side and 1" away from the glass. From that position it can be oriented in any direction. The microphone is filtered such that only frequencies of 90Hz and lower are received.

Specified tracks from the official IASCA disk are played, and highest decibels win. Interestingly enough, it isn't a matter of just cranking up the gains until you either get loud or blow something up. Your maximum decibels will be reached somewhere well short of maximum gain on most systems. Far before you reach maximum gain, you start "clipping" or distorting and even though you can continue turning up the gain, your decibels start going down. Shortly after that, well... electronic equipment runs on smoke. The proof is, once the smoke leaks out, it stops working.

tanstaafl.

"There Ain't No Such Thing As A Free Lunch"

The microphone is filtered such that only frequencies of 90Hz and lower are received.

Hmmm, that's an interesting definition of loud. I guess it's just not as impressive to watch a tweeter exploding

It's a lot easier to generate ridiculously loud audio at higher frequencies, where as bass is something more of a challenge.

Rob