I'm working on a project involving online awareness for Noise Induced Hearing Loss. It's in the very early stages, but I was thinking of different ideas which would help people understand more personally what "too loud" is for listening to music. I have had some ideas brainstorming that were out of my depth in terms of theory, so I was hoping some head-fi users might help 
 
 
The particular idea I have is to create a website where a user is asked to set their volume on their computer to maximum, then adjust an in-page volume slider for a music sample to about their normal listening volume for music. The website would then inform them whether the level was safe, or tell them the amount of time that they could listen to the music at that level before hearing loss could occur.
 
 
The more abstract theory question I'm looking to ask you guys is this:
 
If you took a few common computer listening setups, for example the headphone out of a Macbook Pro hooked up to white iPod headphones, or other common audio devices without amplifiers or other technology. Could you determine without external equipment, and within a reasonable margin (lets say 10dB), the sound output of the setup at max volume from the spec sheets?
 
If this isn't possible, would there be any other "good enough" methodologies for determining the approximate decibel range of a pc/mac volume control? Say adjusting the volume until an audio source is about as loud as a friend talking to you in a normal voice, then recording the audio slider level and multiplying out to figure out the maximum output in decibel from that reference point?
 
I realize that most audiophiles have fairly complex PC setups, but my hope is that the typical online browser has a fairly common setup. Either one of a few onboard motherboard sound with readily available spec sheets online, or a few common consumer grade soundcards, that with some research I could figure out the output range in decibel.  My hope would be that with this information this kind of project could be accomplished. But if its a pipe dream, I'd love for you guys to let me know.
 
Edit: I naively missed this thread which covers a similar issue, which makes it seem like a calculation based on mW output, impedance and sensitivity of headphones which might be too difficult without some serious data collection, but maybe theres a "good enough" methodology with comparison to external sound, i.e. talking in a normal voice as an approximation of 70 dB.
 

If you know the power output of the device into the impedance of the transducer you are driving (eg 15mw from an MP3 player headphone output into 16ohm iPod headphones) and you also know the sensitivity of the headphones, you can work out the maximum attainable volume mathematically: and there is a reasonable amount of data on this sort of thing, especially as regards the headphones, which normally do give their sensitivities. Even when they don't, there are various spreadsheets of that sort of thing you can find by searching the forums.
 
There is an excellent guide to working it all out on a certain NwAvGuy's blog, but I am unable to link it. As for some starting figures, the iPod manages about 15mw into 16 ohms and about half that into 32 ohms. Combine that with a lot of headphone sensitivities and that should get you going.

Edit:
 
oops...I just noticed you have already seen this thread:
 
http://www.head-fi.org/t/586490/approximating-headphone-volume-output-db