So I was shopping around for the DAC 3 from Bel Canto after reading how excellent it was on Stereophile magazine. I was at the showroom that displayed Bel Canto front ends paired with Thiel CS3.7 flagship loudspeakers. And to my amazement, the top end sounded very hard, harder than the Beolab 9s. The overall presentation was quite coherent, neutral and the bass was tight and the midrange was transparent but I just couldn't settle down to listen to them because of how much the top end was hurting my ears. The Thiels were paired with the Bel Canto e.Ref 1000 amplifiers which utilize 1000 watts Icepower amplifier modules. It was not long before I went to the salesman and got my DAC/preamp. This came as a surprise for me because it was the industry acclaimed icepower amplifiers paired together with the award-winning Thiels. Maybe it was my 21 yr old ears being overly sensitive to distortion and the fact that dreadful music was being played through those speakers, but it seems that Bang & Olufsen's Icepower still has a thing or two to be sorted out before it becomes the industry's leading high-end amplification topology. According to Stereophile published measurements, Icepower modules do distort heavily in the high frequency region. 

Bear in mind it's not simply a case of 'whack in the B&O amps and you've got a fantastic amplifier', the power supply makes a big difference to the sound.

As the BeoLab 5s have shown, ICE Power amps are perfectly capable of producing the most stunningly good sound...

I haven't had a chance to hear Hypex modules yet, but many class-d aficionados consider them to be ahead of the ICEpower amps.  If I was in the market for a set of class D amps, I'd probably go with them over Ice.

....so I doubt it has nothing to do with blasting a single 1000w (max!) into 4 ohms into a 3-way design.

....so I doubt that it has nothing to do with the crossover arrangement.

....so I doubt it has nothing to do with anything other than the amp itself?

....or is it that it just has B&O pressed onto the side?

....or is it the cables????

 

10% 

Mr10Percent:

....so I doubt it has nothing to do with blasting a single 1000w (max!) into 4 ohms into a 3-way design.

....so I doubt that it has nothing to do with the crossover arrangement.

....so I doubt it has nothing to do with anything other than the amp itself?

....or is it that it just has B&O pressed onto the side?

....or is it the cables????

 

10% 

 

10% is right.  My previous comment was more of a generalized statement and not in regards to your post.  As 10% said, you can't make direct comparisons until all the other variables have been normalized.  When it comes to audio, the speakers are often more influential in changing the sound than the amplifiers, so you can't do valid comparisons on amplifiers unless you keep the other factors such as speakers and crossovers, and even the room, constant.  Also, just because a brand uses the "ICEpower" trademark, doesn't necessarily mean their "ICEpower" amp will be of the same caliber as the B&O amps.  This is why you can find ICEpower amps in car stereo amps costing less than $200 US.  Although, it'd be interesting to do a direct comparison between the two and see how different they actually sound.
 

I'm not sure I fully understand the logic of your deduction.

1. What was the listening room like? It it was a dedicated room, then that validates an impression. If it was a regular speaker setup on a storeroom floor, then there's reason to wonder.

2. Were you standing or sitting? A lot of people who are in stores evaluating equipment remain standing. That means that the top reaches their ears, but does not get balanced out by the mid drivers, which are pointed at their chest or abdomen.

3. You say you didn't like the music you were listening to. Did you try music you knew? It could be that the music you evaluated was harsh in the top -- I trust you listened to a range of tracks before forming your opinion?

4. DId you check the settings on the amplifier? Were they in neutral?

5. Silly question: Were you in the sweet-spot? I find that the definition of mid and top changes quite a lot, with any speaker setup, if my head is one foot in front or behind a given listening position ...

6. You might want to include a description of the other components in the signal chain when delivering this kind of verdict!

7. If all of the above were controlled for, now we can begin evaluating the use of the ICEpower modules.

As Alex mentions, you have to implement them correctly. But one would assume that Bel Canto has done that, and that the signal out from their component should be fairly agnostic as to what it is connected to.
Still, speakers do have their signatures -- and if a particular speaker is configured to deliver in the top, then ICEpowers ability to deliver definition in that region may give too much icing.

I'm not out to defend ICEpower. If you listen to something, and don't like it, then you shouldn't buy it - whether it's tube or class-D. Peoples' hearing differs to an extreme amount, and actually makes it very unlikely that one person's experiences are valid for another (seriously.)

In the BeoLab 5s the technology has been implemented quite successfully, but I have also heard solutions from other makes, in direct comparison to other technologies, that have been very, very good.

To finish: which Stereophile article are you referring to, considering ICEpower and its delivery in the upper frequencies?

Hello Soundproof. 

I was in a shop floor/listening room. There were very little speakers on display and it was just a pair of the Thiels in front of a damped wall. I didn't really do considerable listening to the installation though I can be certain that it was very disappointing. The electronics were all Bel Cantos with necessary power conditioning etc..Bel Canto doesn't sell amplifiers with tone controls so I can firmly say that no frequency domain manipulations were made. As I said, the overall presentation was very neutral so I don't think it was the tonal balance that caused the apparent hardness in the high frequency region that I experienced. In fact, I often prefer brighter sounding installations because I can hear more into the recording without raising the volume. It was the June edition of Stereophile magazine I read that published the results of the Bel Canto Icepower based integrated amplifier, and mind you Bel Canto is a very well respected and loved by the audiophiles and especially the reviewers from Stereophile. Bel Canto had switched from manufacturing highly priced tube amplifiers to making Icepower based electronics. They were considered creme of the crop. As published in Stereophile, the amplifier started to distort at a few KHz. The distortion measured as compared to the distortion measurements at lower frequencies were magnitudes higher. And in response about the reminder of the implementation of Icepower in BeoLab 5 from the fellow members; the Beolab 5s are designed as an integral system so the internal adjustments to sweeten the treble has probably been done, maybe by adding other forms of distortions. Anyways maybe it is just me, but I find the high frequency delivery highly unacceptable. Oh and please do not forget that I am only 21 so my ears are probably more sensitive to high frequencies. 

*Edit. Maybe hard is the wrong word to describe the high frequency. I would say it was more dead and synthetic, cos hardness may imply brightness. 

 Just to add... In response to Mr10percent. Icepower amplifiers deliver lesser distortions into lower than higher impedances, in fact it is lowest at 2ohms.  

I wish to withdraw my previous comments about blaming Icepower that it gives me a ear ache listening to its high frequency performance. Turns out, it is only when Icepower is paired when a loudspeaker with metal dome tweeters that gives me the problem I mentioned. Which totally makes sense since metal domes break up heavily in the very high frequencies. So to those Icepower fans... Apologies..

There is also another factor.

If the demo music is from a CD, there is a problem with too high source sound levels.

On a CD, or other digital media, the sound setting is seperate to the actual music information. On many digital sources, this sound setting is set too high for the digital media to cope with, the effect can either be distortion or it can be to compress the dynamic range.

You can get programs to look at the sound level, and reset it, actually its called 'normalising'

 It can be irritating when you have lots of digital music files and you have to keep changing volume when a new file is played Normalising is a process of setting an upper limit to all the files, and setting their overall volume relative to each other.

One simple version is MP3 gain.

http://mp3gain.sourceforge.net/download.php

This has certain limitations,as it works on peak normalisation instead of average normalisation, you really need a dynamic form of normalising, however the only programs I have come across that can do this are expensive, I haven't found a free one yet.

One would imagine that those promoting and selling CDs and other digital media would want to present their products to the consumer to the best quality and advantage, but they don't, and when you try demos then stuff can sound flat, distorted, boxy, or depedning upon which frequencies are dominant, it can be hard to listen to - stuff in the mid to upper range seems to make the ears tire, you could almost describe it as the BOSE syndrome - since that's how their products sound to me - flashy in the short term but not something to live with.

So the summary of all that is, you cannot always blame the sound reproduction equipment, the source could be rubbish, and its the media companies that make it that way.

 

http://www.hometracked.com/2008/04/20/10-myths-about-normalization/