Originally Posted by Swolephile
Im interested to hear everyone's thoughts on such factors as damping factor and slew rate as well as others in the makings of a good amp.
No need to get all esoteric, when we know why some class D (and many tube) amps sound different from solid state and class D amps (and a few tubes): source impedance.
A good general rule of thumb in audio in 2014 is that differences in electronics that affect the frequency response of the complete system are audible, and most others are not. For example, the black line on the graph below shows how a simulated 2-way loudspeaker modulates the frequency response on the tube amp most recently published on Stereophile's
By contrast, here's the same graph for the first solid state amp listed on Stereophile's
equipment reviews page, which is what one should expect from pretty much any class AB amp:
If one can't hear a 4dB dip over the 2-4kHz octave in a system with nothing else changed, one probably shouldn't be into audio...
How does class D fit into this? Everywhere.
Due to the required output filters, some class D amps have quite non-flat frequency response when driving real speakers. (I suppose there are some full-range planar models that have nearly resistive impedance, and thus would be relatively unaffected.) Others have circuitry that compensates for the output filter, and can produce flat frequency response when driving real speakers.
Originally Posted by AudioJosh
***I personally like class D (such as Hypex, I really dislike ICE based amps)***
I'm curious, why? I have a few of each (amps based on the Hypex UcD180 module, ICEpower 50ASX2, and ICEpower 125ASX2) and they all sound the same when appropriately output-limited: low noise, flat FR.
OTOH, I once tried a 6-channel Rotel unit with early ICEpower modules with the intention of using it for active biamped LCR, and it was a POS. (As was the replacement unit.) Noisy, and sounded distinctly dull up top on the speakers I was running at the time. But the new ones cleaned up the frequency response under load, and I don't know (or care) if the noise issues were from the ICEpower modules or just a crappy implementation by Rotel.
Originally Posted by Archaea
The guys in KC compared some of the following amps ***We tested the amps on the JTR Noesis 212HT and the Klipsch RF-63 in full range mode.
None of us could reliably tell the difference in any of these amps with blind testing, when level matched and fast switching was an ability.
That surprises me, honestly. I would've expected that the iNuke (based on measurements of FR into different loads that was published in a German magazine a while back) and the Lepai T-amp would sound different, though no difference between the solid state amps makes sense.
What was the signal? As I wrote in the post Sibuna linked to above, I found clear differences between a Crest ProLite 2.0 and a modern solid state amp. With music in sighted listening I thought I heard differences, though it was much harder to ascertain than on pink noise, where the differences were obvious to the point of not needing blind testing (and quickly confirmed on an iPhone-based RTA).Edited by DS-21 - 1/25/14 at 7:36am