RMS and PMPO, and implications for "loudness race", no need to squash for cheap equipment? Options

[MugFunky]

just had a thought about the usual speil about needing to squash records for optimal playback on cheap equipment.

if you've ever browsed a department store's entertainment section, you'll be infuriated to see all those awful all-in-one units have power output measured in some wacky "PMPO" measure instead of the _actually meaningful_ RMS value.

a search for "PMPO" yields many conflicting explanations, but i've managed to summarise it as meaning "dynamic headroom" from a mailing list somewhere.

this basically means the RMS can be tiny, but the PMPO can be very large depending on how much headroom the system has.

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

here's the interesting part:

my interpretation of this is that a cheap stereo manufacturer is very much interested in getting the maximum PMPO for the least money (thus lower RMS).

this means a cheap hifi system will have low average power, but a very high transient response. in theory it should handle very dynamic recordings well, and possibly could suffer badly with squashed records with the highest RMS that can fit on the CD.

if that is so, then why the loudness race at all? the only thing i can see is the CD changer argument, where marketing dicks want their disc to be the loudest in any given changer. no other argument works at all, when you consider a squashed song will come out quieter compared to a dynamic one on the radio.

any thoughts?

This post has been edited by MugFunky: Apr 5 2004, 07:55

[cabbagerat]

I don't think the PMPO rating is as scientific as that article implies:

This one is closer. http://sound.westhost.com/power.htm

[sven_Bent]

PMPO is mathematically expressed as

PMPO = Actual_Power * k
where Actual_Power is calculated by the formula below, and k is a constant whose value is approximately equal to one's grandmother's age, divided by the square root of the distance from the office to the advertisement writer's normal place of abode - measured in millimetres, inches, furlongs, statute miles or pounds per square inch - as appropriate, to provide the number you first thought of.

LOL

[2Bdecided]

I always thought it was the maximum amount of power that could be provided for 10ms without destroying the amplifier.


However, you're right MugFunky (blimey - I've typed that twice today! steady on MugFunky!) - dynamic tracks can sound louder on poor equipment than squashed tracks.

At a give perceived loudness, the squashed track will distort constantly, before the peak distortion on the dynamic tracks begins to sound too nasty.

As that article suggests, it does depend on the design of the power supply.

Cheers,
David.

[MugFunky]

hehe. yes, well obvioulsy i wasn't advocating PMPO as a scientific measure, but it's got to come from somewhere or consumers would sue (maybe they should).

my point was that by a reasonably cynical definition of PMPO (max transient response without damage), it can be assumed that dynamic tracks can be handled without damaging the equipment, and (depending on power supply, i'd read that but didn't feel it necessary to mention) would sound pretty good too.

considering that at your average "18th party" where these stereos are used at their maximum, they are driven to insane clipping anyway, there's not much to lose. my old all-in-one suffered years of abuse by me and still works to an acceptable level (acceptable to my brother who i gave it to... personally i think it sounds like poo), and i used it out-of-spec for several years.

i drove 3 pairs of car speakers (shudder) in parallel with that thing, often at 2/3 full volume (after that the clipping way too obvious even to an adolescent metal-head such as i was then).