[bigshot]

Yep. That's EQing subtractively. And then you just turn up the volume on the amp and use a little of the 10dB headroom. In practice, most EQing for tonal balance doesn't exceed +/-3-4dB. 10db allows for a lot of leeway.

 

[ADUHF]

Totally depends on what you mean by "headroom". Because that term can mean different things in different contexts (as this discussion has amply demonstrated).

 

[ADUHF]

amply demonstrated

A little bit of "amp" humor there for yer Monday night.

Those DBX EQ's are designed primarily with pro audio in mind btw. So I wouldn't generally recommend them for home use.

 

[bigshot]

Totally depends on what you mean by "headroom". Because that term can mean different things in different contexts (as this discussion as amply demonstrated).

Wow. I've reminded this thread of the context a few times now and it keeps getting forgotten. The topic is as hard to follow as grabbing a handful of earthworms! There are serious communication problems here.

The question that started this discussion was how much extra volume do you need in an amp beyond the ability to play music at the loudest normal listening volume. I said that studios allow 20dB headroom when they record, but for playback 10dB is plenty and you would never need any more power or volume with normal use.

In this case, I'm saying that an amp with 10dB of headroom is more than enough to raise the volume of music that has been subtractively equalized.

I have a Rane 2 channel quarter octave equalizer. It's very handy if you are equalizing analogue signals, like reel to reel or records.

 

[ADUHF]

The question that started this discussion was how much extra volume do you need in an amp beyond the ability to play music at the loudest normal listening volume. I said that studios allow 20dB headroom when they record, but for playback 10dB is plenty and you would never need any more power or volume with normal use.

When you say something like this, then your comment starts to make a little more sense. There is still plenty of room for interpretation (or misinterpretation) in your above statement though.

Fwiw, I think these were the original questions that started this off...

Also, would ~1/3 of a watt generally provide enough headroom for enjoying both loudness-boosted, and also full dynamic range recordings (assuming the user wants to be able to enjoy both)? And is there a general range that's considered optimal or minimal for this kind of thing?

I can also see some potential advantages in the rendering of transients, at the extremes of the volume range, with an amp that provides a bit more headroom. But I don't really know what is a reasonable range for this kind of thing with higher quality recordings. And whether 1/3 of a watt is really sufficient for enjoying full dynamic range recordings, for example. Because most of the stuff I listen to is heavily compressed in terms of its dynamic range, in order to boost its loudness. So it requires less amplification.

So from my perspective, it was more about the amount of overall amplification needed for enjoying full dynamic range recordings vs. recordings where the dynamic range has been more compressed in order to boost loudness. And I think we have that pretty much sorted at this point.

The comments/questions re EQ-ing seem to have re-opened the subject of exactly what "headroom" means though. And it doesn't seem as though we've established any sort of a universal definition for that. It seems to me that it can mean different things to different people, depending on the context.

If you say it's the number of dB above your loudest normal listening volume, then I can gauge a little better what you mean by that. If I'm looking at the specs on different amps though, and trying to gauge if amp A has enough amplification for my needs vs. amp B. And I have no idea what my loudest normal listening volume is in decibels to begin with. Then simply saying that it's 10, or 20, or 30 dB "above that" is not that helpful.

 

[castleofargh]

Between EQ and replaygain, I'm often near -15db on the digital out with modern tracks. If we add my lazy habit of using foobar’s volume instead of the amp for small adjustments, it's pretty relevant to consider the necessary headroom. In my case I still wont end up above 115db total(spl output + compensating digital attenuation). But some people might.

 

[castleofargh]

Know thyself.
Nothing in excess.

And the last one doesn't help picking an amp.

 

[VNandor]

The comments/questions re EQ-ing seem to have re-opened the subject of exactly what "headroom" means though. And it doesn't seem as though we've established any sort of a universal definition for that. It seems to me that it can mean different things to different people, depending on the context.

Here's the reason why I brought up EQ. If someone wanted to calculate how much power they need from the amp, at some point they would have to consider the differences between the average levels (be it RMS or sustained peak, doesn't matter as long as they know what they mean) and the peak levels of their music. Since the EQ not only changes both the peak and average levels but also the differences between them, whatever power requirement they might calculated from their pre EQd music won't hold true for the EQd music and the amp obviously have to reproduce the EQd signal.

Edit: I guess all the other DSPs that one might use would have to be taken into account as well. I just singled out EQ because that's the only thing I use for listening.

This distinction ultimately wouldn't matter if the amp is powerful but EQd music will need more power to reach adequate levels compared to what it would need if it weren't EQd. I happen to use an interface that only provides a couple of mW to my headphones so in my case every dB matters and I had to turn off my EQ to reach my usual listening levels with quiet music in some cases.

That is what I don't understand. If that second waveform is normalized up to zero, there has to be a spike in there somewhere that pushes up to the top. That would be a pretty big spike compared with the overall level of the music. My question is, where is it and what caused it? Are you sure that second waveform was normalized up? I don't see anything in there up near zero.

Look more closely. It is normalized to 0dB. I think the biggest peak is between 1:30 and 1:45 but if you look at the whole piece, you can see there are several peaks throughout the music that would be clipped if I applied more gain and tried to get back to the same RMS as before the EQ.
It's of course caused by the EQ. This is how this music ends up if I EQ it with my usual curve I use to "correct" my headphones. So I guess it's caused by the music as well?

 

[HiFiHawaii808]

Yesterday was a great example. I went to a hifi store because I wanted to audition a whole host of products so that I could form an opinion of them. What I discovered in that listening session is that the Focal Clear OG and the ifi Diablo is a perfect match. No amount of measurements would have predicted that a priori. So, the answer to the original question for this thread is this. No, matching headphones to amps is a discovery process just like fitting eyeglass prescriptions to people. You can get in the ball park with science, but at the end of the day, subjective preferences need to be field tested.

 

[bigshot]

I think the biggest peak is between 1:30 and 1:45 but if you look at the whole piece, you can see there are several peaks throughout the music that would be clipped if I applied more gain and tried to get back to the same RMS as before the EQ.
It's of course caused by the EQ. This is how this music ends up if I EQ it with my usual curve I use to "correct" my headphones. So I guess it's caused by the music as well?

I suspect it is an interaction between the tonal quality of the music combined with the EQ curve you are using. I don't think you could predict whether it makes the music quieter or louder without knowing how the EQ is going to affect the music. I think I understand it now, even though I wouldn't be able to predict the effect it would have on the perceived volume of the track.

I suppose if you are going to stack up layers of signal processing, you might need a bigger headroom. Then you are getting into the territory of sweetening, rather than just playing back. But I suppose if you are applying a lot of processing, you probably aren't using a tiny dongle amp anyway.

 

[bigshot]

Yesterday was a great example. I went to a hifi store because I wanted to audition a whole host of products so that I could form an opinion of them. What I discovered in that listening session is that the Focal Clear OG and the ifi Diablo is a perfect match. No amount of measurements would have predicted that a priori.

Now that you know it works, make a note of the specs on that combination. You know it works, and likely there are other products with the same specs that would work the same. You might very well find something that performs just as well for a lower price that way.

 

[ADUHF]

Here's the reason why I brought up EQ. If someone wanted to calculate how much power they need from the amp, at some point they would have to consider the differences between the average levels (be it RMS or sustained peak, doesn't matter as long as they know what they mean) and the peak levels of their music. Since the EQ not only changes both the peak and average levels but also the differences between them, whatever power requirement they might calculated from their pre EQd music won't hold true for the EQd music and the amp obviously have to reproduce the EQd signal...

...This distinction ultimately wouldn't matter if the amp is powerful but EQd music will need more power to reach adequate levels compared to what it would need if it weren't EQd. I happen to use an interface that only provides a couple of mW to my headphones so in my case every dB matters and I had to turn off my EQ to reach my usual listening levels with quiet music in some cases.

If you are close to your system's limits in terms of volume, then I can see how the changes in volume with EQ could make a difference. When using EQ on my system, the overall volume is generally reduced by anywhere between about 3 to 6 dB. So I have to boost the volume on my amp by roughly that amount to match the pre-EQ volume level.

When I'm creating a new EQ curve for my headphones though, in Equalizer APO's Configuration Editor, I will usually create more than one project. And one of those projects will contain just a simple preamp control, with a negative gain in the above range (so usually around -4 or -5 dB). This allows me to do a more fair comparison between the sound of my headphones before, and after a new equalization curve has been applied, with the levels more appropriately matched. So I don't have to make the adjustments for this on the amp.

Edit: I guess all the other DSPs that one might use would have to be taken into account as well. I just singled out EQ because that's the only thing I use for listening.

Other DSP's could also be a factor. If your amp already has enough volume (I'm deliberately avoiding the term "headroom" now to hopefully reduce some of the confusion) to comfortably play back full dynamic range recordings, then you really shouldn't need any additional volume on top of that if all you are doing is re-expanding the dynamic range on a more heavily compressed recording... At least in theory anyway.

 

[ADUHF]

Yesterday was a great example. I went to a hifi store because I wanted to audition a whole host of products so that I could form an opinion of them. What I discovered in that listening session is that the Focal Clear OG and the ifi Diablo is a perfect match. No amount of measurements would have predicted that a priori. So, the answer to the original question for this thread is this. No, matching headphones to amps is a discovery process just like fitting eyeglass prescriptions to people. You can get in the ball park with science, but at the end of the day, subjective preferences need to be field tested.

Howdy, HiFiHawaii808. Are you sure there wasn't just a little bit of science involved in your choice? Maybe just a smidge?

 

[ADUHF]

I won't get into the volts and watts again, until I understand all that stuff a little better. From what I think I'm able to understand though, it seems as though the ADI-2 was designed with accuracy, neutrality, precision, and also flexibility in mind. As opposed to a certain sound signature.

 

[bigshot]

When using EQ on my system, the overall volume is generally reduced by anywhere between about 3 to 6 dB. So I have to boost the volume on my amp by roughly that amount to match the pre-EQ volume level.

That is pretty typical for most people who EQ. I don't think many people are doing gross corrections with EQ unless they are creating an RIAA curve or something like that. Most headphones are in a general enough ballpark to not need huge corrections.

From what I think I'm able to understand though, it seems as though the ADI-2 was designed with accuracy, neutrality, precision, and also flexibility in mind. As opposed to a certain sound signature.

That could describe just about any DAC.