$50 gadget claims to "clean the sound waves of MP3"

Well, the first obvious point is that it doesn't work on digital audio: it's processing the analog signal after the D/A converter.

Looks like little more than a frequency-dependent compressor, smoothing out the peaks at high frequencies. A crossover splits the signal in to high and low, and it compresses the high part. I suppose it might make bog-standard 128kb/s MP3s sound "nicer", but if you encode your MP3s at a higher rate with a good encoder, and play them through decent headphones, you don't need to add anything else to the signal path.
 

I was about to post the same thing as Brian but he's pretty much nailed it form my point of view, so I'll, errr, shut up.

They don't mention that the resulting signal strength will be reduced. For those who like listening to ear-bleedingly loud music on the bus, they'll be disappointed.

Does anyone else feel that $50 is a lot for a dynamic filter?
 

It doesn't mention any sort of power supply, so I don't think it's possible for it to be an actual active compressor. A multiband compressor/limiter would more or less do what they're claiming - it turns everything into LOUDER IS BETTER music, but most MP3-player-listeners will probably think that sounds better, and it genuinely is better if you want your music to be audible over background noise on the bus or while jogging or whatever without turning the volume way up.

But I don't think it can be that, because there's no battery to power it.

So I guess it could be some kind of filter, possibly as little as a cap across the signal.

Or it could be a zero-ohm resistor connected to nothing.
 

I'm with Dan. Since it's probably passive there's really only so much you can do.

Also, they say it contains a custom semiconductor they designed, but there's obviously no digital logic here (otherwise they'd have to include both an ADC and a DAC, not to mention actually power the damn thing) so my guess is that it's nothing but fluff.

So who's buying one and cranking it open?
 

When it comes to products like this, the old saying "you can't polish a turd" springs to mind.

Compressed audio is compressed audio. You're not going to make it sound 'better' by passing it through a little box. If anything, further processing will move the result farther away from the uncompressed original.
 

sounds like a good ol' BYPASS box to me....
 

From their site it looks like it's some kind of passive compressor. It's not like most of the music people have on their iPods needs any additional compression what with modern trends in audio production.
 

My bet's on a scam.
 

"Compare the sound of an FM station to the sound of most CD players. The same song usually sounds more vibrant and natural on the FM station."
from the ClariFi site

That's a deal breaker for me...
 

The chap got in touch. He's going to send me one. If it arrives, I'm going to be asking for experiments to do with it, before cracking it open to see what's inside.
 

clari-fi can go plug it's 1/8" stereo connector into it's own jack.

(if you know what i mean)
 

Apparently the "good high frequencies" are the ones above 300 Hz and not peaking.

I think it's a crappy passive 2-way cap limiter.
 

I wouldn't have spent any time at all with this, except that it was Mr Whitwell who put out the call. But frankly the smell here is overpowering. I'll look at the semiconductor side of it, since I'm an electrical engineer and I've worked in that business:

* 'Passive semiconductor chip'? Of course that's technically possible, but completely dotty. It's almost certainly vastly cheaper, and better, to do a passive circuit with discretes. Anyway, I can't believe that anybody with 'thirty years experience', and enough money to do an IC, would even think of doing something like this.

* I'm sure there's some 4-inch IC equipment still around, and maybe even a fab or two, but they aren't wasting their time making 'passive semiconductors' (diode manufacturers perhaps excepted).

* If the passive element in question is voltage-variable, it almost certainly isn't passive, varactors excepted.

* If the passive element is a varactor, then it's a huge varactor of a previously unseen class, being large enough to affect very low frequencies; not an IC; potentially an advance for which there ought to be journal papers and patents; and makes basically no sense in the context of 'passive compression'.

* If the element in question is voltage-variable, and has three terminals, then it's almost certainly a transistor, and therefore not passive.

* If the element in question is voltage-variable, has three or more terminals, and is not a transistor (or something optical), then we have an advance in semiconductor devices, and I expect to see journal papers, announcements, patents, and applications far beyond cleaning up the sound of MP3 players.

* What are the 'voltage and impedance constraints' in the MP3 player market? Who sets them? IEC? ISO? AES?

* Can't resist pointing this one out: 'Parasitic compression function'? Most evocative ..

* From another page: 'The more audio files sizes are reduced, the more sound waves become jagged ..' This statement is nonsense. (I'm happy to elaborate on request, but this comment is already too long.)

If there's a product behind this, I'm guessing that it's a box with some random junk wired up inside, or maybe an RC filter. Of course I'll be glad to be shown otherwise.
 

the site says it's powered by the audio... I bet they're using the power of the low frequencies to compress and boost the upper frequencies...

while it's a quantifiable decrease in volume, it seems louder because of the compression and the fletcher munsen curves (our ears are, on average, more sensive to higher frequency sounds)

anyway... FM radio does NOT sound better than a cd... and anyone who claims it does is an idiot. Go listen to the radio and then play the same song back from a store bought cd, you'll see what I mean. Burning an mp3 on a cd doesn't get the same effect.

also... it's a proven fact that compression is one of the main CAUSES of listening fatigue, so I don't see how more compression would solve that problem.

all in all this product seems like a farse to me
 

I finished trying to read the WWW site, and it is complete and utter rubbish. It would take too much effort to critique all the errors, because, basically, NOTHING MAKES SENSE.

If you want to see what it really takes to clean up degraded or compressed audio, look up "digital audio restoration" or "bandwidth extension". Bandwidth extension (or replication) attempts to fill in the high frequencies that can get lost in audio compression. Acronymns include SBR and AACplus.

I was fortunate enough to meet some of the inventors of audio codecs like MP3, AAC, and ATRAC, and learn from them. I'm not a world authority, but I know how these codecs work - enough to know that the "invention" here is nonsense.
 

I'm surprised to see so many people so pleased with the absolutely horid sound of the ipod headphone amplifier. It is terribly difficult to listen to music on an ipod, and I know what it means to experience ear fatigue when using one. Me personally, I'm interrested in trying out one of these gadgets. From posts on other blogs, it sounds like a giant leap forward in silicon design, and not just some caps and resistors wired up as a sort of gimmick. I mean, how expensive would it be for a small company to try to deceive people in such a ridiculous way. Get real folks. No wonder there is so little audio inovation these days. So quick to critique without even trying.

I suppose the people who comment on blogs are perhaps those same people who hang out at music stores to demo equipment, and never buy anything.........