Old power specs vs new ones

[PJL]

Audio specifications are not fit for purpose. I am really not interested in knowing my amp can deliver 650W of 1kHz tone for 1 second into a 1 ohm load with 1% distortion. When will the audio industry come up with a sensible audio specification based on a standard sound track with realistic content.

[stevehertz]

My thoughts exactly, and why I started the thread. In my mind, the old way of speccing both (all) channels driven, 20 to 20KHz, at 0.1% (or whatever) distortion is as good a way as any. But as has been the case for decades, manufacturers don't like that because it doesn't always make their products look good (indeed they probably aren't).

[mhennessy]

Originally Posted by PJL

Audio specifications are not fit for purpose. I am really not interested in knowing my amp can deliver 650W of 1kHz tone for 1 second into a 1 ohm load with 1% distortion. When will the audio industry come up with a sensible audio specification based on a standard sound track with realistic content.

But what's the alternative?

Amplifier manufacturers can't know how sensitive your speakers are, how far you sit from them, how dead or lively the acoustics are in your listening room, what sort of music you play and how much dynamic range it has, or even how loud you can play, considering neighbours, etc.

Given a knowledge of those variables, an amplifier manufacturer could suggest the maximum replay level from a given amplifier in dBSPL. Would that figure - even if approximately accurate - be any more meaningful to the average consumer than watts? I'm not convinced...

Perhaps that would be better for a more technical person, but such a person would surely have no trouble converting watts "RMS" into dBW and adding that to the sensitivity of their speakers, while factoring in the other variables, no?

Here's what I've been recommending for years, perhaps decades. It's really simple. Put clipping LEDs on amplifiers - just like they do with pro amps. Then, if the LED flashes when you're playing your chosen music at your chosen level, you know that you're at the limits of what the amp could provide, so you need to get a bigger one.

Apart from cost, why don't hi-fi manufacturers fit these? Because hi-fi amps clip rather more than we'd like to admit, especially when we're playing very dynamic, well-recorded material. If you take a bunch of modern semiconductor amplifiers that have been competently designed and audition them all, the differences between them all vanish when you do 2 things: precisely match the replay levels (within 0.1 of a dB), and ensure none of them are clipping under any circumstances. If you ignore that last point, then the group test is nothing more than a comparison of the different clipping behaviour of the amps (funny, I though it was about hi-fi, not guitar peddles!). Hence designers ensure their amps sound reasonable (as far as one can) when clipping, in full knowledge of this.

It all comes back to the peak-to-mean ratio of real music that I keep going on about...

I lack the time/energy to do it all again, but a few years back I spent a few days convincing the owner of a certain UK brand of loudspeakers about all this (on his discussion forum, compete with pictures and audio clips). He went from total disbelieve to complete understanding, and has now revised his advice for amplifier selection as a result. Assuming he's not removed it from his forum, it's all out there - perhaps I'll take a look and find a link when I get home.

get a CE mark

You don't get one, you apply it yourself presumably after making sure it will meet the requirements. Well done Linn.

[sp10mk11]

As Paul said some time ago this is an absolute minefield and in some ways I just ignore such specs and use my ears to decide, there will always be anomalies with any system of measuring performance of an amplifier.
Though as Steve says the old way of measuring output power, both channels 20-20Khz into 8 and 4 ohms worked reasonably well the, 4 ohm measurement I always thought was a good guide as to how up to the job the power supply was, all other things being correct.
Ah Well good luck debating this one
Gary

[stevehertz]

Originally Posted by merlinmaxwell

You don't get one, you apply it yourself presumably after making sure it will meet the requirements. Well done Linn.

Or if you're Chinese, you just put a CE sticker on it anyway; the customer wants CE approval (and a sticker to confirm that), so we put a sticker on. Means zilch, but who cares, least of all the Chinese government who welcome the revenue, honesty be damned. UK buyer beware of CE marked Chinese goods...

[RojDW48]

An example from about 10 years ago - subjective, of course, but relevant I hope. I had a B&W Aura amp - ostensibly 200W p/c and I got the opportunity to buy a Radford STA25 very reasonably - ostensibly 25W p/c. An old friend and I compared the 2 extensively - same sources into a pair of LNB Paralab 20 loudspeakers. Both sounded very good of course but the 'little' Radford had an effortless quality about it compared to the 'big' B&W. in other words it seemed to be able to deliver the big complex transients better than the 'more powerful' Aura. It still doesn't make much sense, though this thread has gone a little way to clarify. Needless to add, I kept the old Radford and sold the Aura!