Headphone amps, external power supplies: what difference do they make?, was: "First post here, Hi." Options

[skamp]

Thanks for doing this so I don’t have to.

I blame head-fi prose for making me nauseous.

Of interest, is Ethan Winer's description of 4 parameters that affect sound quality: "noise, frequency response, distortion, and time-based errors.". Incidentally, those are measurable. I don't know if it has been done already, but I wonder if anyone's ever conducted a battery of double-blind tests to match their thresholds of audibility with measurements. That would be very interesting (though I can't be arsed to do it myself).

[Arnold B. Kruege...]

Thanks for doing this so I don’t have to.

I blame head-fi prose for making me nauseous.

Of interest, is Ethan Winer's description of 4 parameters that affect sound quality: "noise, frequency response, distortion, and time-based errors.". Incidentally, those are measurable. I don't know if it has been done already, but I wonder if anyone's ever conducted a battery of double-blind tests to match their thresholds of audibility with measurements. That would be very interesting (though I can't be arsed to do it myself).

While Ethan's list lacks some rigor (the better words are IMO noise, nonlinear distortion, and linear distortion), the idea is very good.

The problem with testing the thresholds of hearing for those things is that things can get real complicated fast as soon as your tests engage things like masking.

Arguably the first well-known tests for such things resulted in the Fletcher Munson curves which are interesting because they were done in such as way as to pretty much avoid masking. The general rule is that since masking is so endemic, Fletcher and Munson's numbers are usually highly optimistic, IOW too sensitive. But they are indicative. Perceptual coders would have never worked out if we thought that Fletcher Munson were the be all and end all.

One thing that can be said is that if you get all forms of noise, nonlinear distortion, and linear distortion more than 100 dB down, you need to move on for sure.

The actual thresholds for specific things are mostly in the 30-60 dB range, depending. There are a few cases where 20 dB down can suffice.

[psgarcha92]

Respected Moderators,
I am sorry that i did not read the TOS at all. Now that i have, i will be taking care of all that i have been notified about in the preceding posts and TOS.
Thanks for being patient.

@skamp,
Thanks for your replies. You guys are way more experienced about this stuff than i am, i will be taking extreme care of what i post here from now on.

@saratoga
In my reply to the OP, i thought i would just mention my Rig in-case he wonders. I wasn't asking him to use an amp, i just mentioned my Rig. My reply was about trying another set of Earphones and see how it goes for him. It was you who pointed out that an adding an amp makes zero sense. Which i found insulting, and that's why i replied to it. I happen to like the coloration that i end up with, nothing else. The post was about changing Headphones, not adding an amp. I am sorry if i offended you in any way.

Regards

[saratoga]

^^ unless something is really screwy with that amp you are not getting any coloration. That's my point, you took one good output and substituted it for another equivalent one. That doesn't do anything . Then you posted to tell us about it.

I am not offended and I didn't mean to insult you but I was trying to suggest to you that you may not understand what you are doing and might want to read a bit more.

[Arnold B. Kruege...]

I don't remember if there is +/- 12V on the PCI bus. I would assume so because they used to have +/-12V RS-232 ports on I/O cards, but I don't know for sure.

You don't need to remember, you just need to know where the spec can be found:

http://en.wikipedia.org/wiki/Conventional_PCI

Pins 1 and 2 of a conventional PCI slot are for + and - 12 volt power.

[saratoga]

Any sound card will use a regulated voltage anyway (or at least it'd better or its quality is going to suffer), so they can generate any voltage level they want from the usual 3.3v rail. This is very uncommon though because most headphones have moved towards lower impedance designs that are meant to be driven by relatively low voltages. Even modern high impedance headphones are actually of relatively moderate impedance by historical standards. Its fairly rare to see headphones that can be safely (for the users ears!) driven at 2 Vrms these days.