Vasr
Major Contributor
- Joined
- Jun 27, 2020
- Messages
- 1,409
- Likes
- 1,926
I asked Amir in one of the threads on whether linearity can be measured for amps and he replied in the affirmative but one would need to standardize the power levels to do such a test.
With recent experience from an amp I acquired, I am wondering if this test would show up some poor amps even if they have a flat FR. And whether we should figure out a way to standardize that. They might even explain the "sound signatures", even if the FR is flat. Has any site done such tests for amps to see if this is a non-issue for most amps to bother?
The amp has been OK with no obvious signs of distortion or noise and sounds clean across the range (at least the range I can hear) but it has been very unsatisfactory aurally compared to the amp it replaced. So, I have been doing a lot of listening tests to see what this is missing and I have arrived at a conjecture.
The amp seems to attenuate the low-level signals (across the spectrum not in any part) way too much relative to the higher-level signals. This has a couple of audible effects. One, many of the harmonics that make an instrument sound rich/warm are at very low levels. If those are attenuated too much because of their relative levels, then the perceived sound is altered. Some instruments sound thin for example. Two, some of the details get lost. There might be a small playful work on a symbol by a percussionist which at normal relative levels contributes a lot to the passage but if attenuated may never show up enough. So this is not a masking effect since the attenuated part may be in a completely different part of the spectrum from the louder sounds at the time.
To test this, I compressed the same song with Audacity as a modern mix might and a lot of detail shows up that can never be heard with this amp uncompressed. These were clearly heard in the previous amp with the same rest of the audio chain. If I increase the volume to hear the low level parts, the louder parts get way too loud. It is almost like the amp is designed for compressed sound!
One would think such critical listening issues would only happen for stereo music listening but I have found this amp not satisfactory for even multi-channel HT. Here is how the same problem manifests. A lot of the surround mix has low level signals in surround speakers (rain, twig breaks, etc). These get lost but more annoying are the transitions from back to front and vice versa. They typically use fade outs on one pair and fade-ins on the other side. But if the low-levels during these fades are too attenuated, they you have a gap in sound when it becomes too low at the starting point and it picks up late at the ending point. So, it never gives a smooth transition only a jump which is annoying.
Is there a measurement I can do with REW that would test this conjecture?
With recent experience from an amp I acquired, I am wondering if this test would show up some poor amps even if they have a flat FR. And whether we should figure out a way to standardize that. They might even explain the "sound signatures", even if the FR is flat. Has any site done such tests for amps to see if this is a non-issue for most amps to bother?
The amp has been OK with no obvious signs of distortion or noise and sounds clean across the range (at least the range I can hear) but it has been very unsatisfactory aurally compared to the amp it replaced. So, I have been doing a lot of listening tests to see what this is missing and I have arrived at a conjecture.
The amp seems to attenuate the low-level signals (across the spectrum not in any part) way too much relative to the higher-level signals. This has a couple of audible effects. One, many of the harmonics that make an instrument sound rich/warm are at very low levels. If those are attenuated too much because of their relative levels, then the perceived sound is altered. Some instruments sound thin for example. Two, some of the details get lost. There might be a small playful work on a symbol by a percussionist which at normal relative levels contributes a lot to the passage but if attenuated may never show up enough. So this is not a masking effect since the attenuated part may be in a completely different part of the spectrum from the louder sounds at the time.
To test this, I compressed the same song with Audacity as a modern mix might and a lot of detail shows up that can never be heard with this amp uncompressed. These were clearly heard in the previous amp with the same rest of the audio chain. If I increase the volume to hear the low level parts, the louder parts get way too loud. It is almost like the amp is designed for compressed sound!
One would think such critical listening issues would only happen for stereo music listening but I have found this amp not satisfactory for even multi-channel HT. Here is how the same problem manifests. A lot of the surround mix has low level signals in surround speakers (rain, twig breaks, etc). These get lost but more annoying are the transitions from back to front and vice versa. They typically use fade outs on one pair and fade-ins on the other side. But if the low-levels during these fades are too attenuated, they you have a gap in sound when it becomes too low at the starting point and it picks up late at the ending point. So, it never gives a smooth transition only a jump which is annoying.
Is there a measurement I can do with REW that would test this conjecture?
Just being facetious but this is sort of similar to what happens when I add a noise gate in my effects pedal chain to get rid of the noise from the low output Strat pickups when played with high gain effects to get that "tone". The sound decays sharply as low levels get increasingly attenuated through the noise gate. Yes, I know hi-fi equipment is different. 
