As file resolution goes up sound level goes down?


I've downloaded files (e.g., HDTracks) in various resolutions and I've noticed something strange. I'm getting sound level differences between files of different resolutions.

For example, my Turntable front end has plenty of gain, but when I switch to CD I have to turn up the volume a little more to get the same level. If I play 96kHz/24-bit files I need to turn it up even more. Finally, I just got Marley's Legend in 192kHz/24-bit, and I have to turn it up even further.

This reminds me of when I play a Chesky record, I have to turn things up. What's up?
kennythekey
Different volume levels are probably due to lack of compression. I've had uncompressed cd's in the car and you can't hear the quieter parts at the usual volume you listen at, while the cd sounds great on my main system. That is one reason why so many cd's are compressed, they are recorded for cars and boomboxes, not good stereos. Compression robs music of dynamics which is what makes music thrilling.
Thanks Al!

Looks like my focus should be on the speakers, and you know that "ain't" a bad thing.
Thanks for the additional info, Kenny. I suspect that the harshness at 5:30 is the linearity degradation and/or soft clipping that I referred to, corresponding to the amp approaching or reaching the limits of its power capability. If so, along the lines of my earlier comment a gain increase would not help. It would result in the same thing happening, at the same perceived volume, but at a lower setting of the control.

Regards,
-- Al
Al, Let me give you a more accurate explanation. I don't like to clip my amp, so I had turned up the volume to 3:00 o'clock out of a max of 5:30. At this point, no perceived distortion but the gain was still mediocre.

From your comments, I decided to go further and at 3:30 it started to get louder but the noise floor went up. I decided to take it to 5:30 and the amp did not clip, it was playing loud at this point but sounded harsh.
Hmm, that sounds like even though you are not clipping the amp to a degree that is clearly perceptible as such, it still may be running out of steam, perhaps as a result of degraded linearity as it approaches its maximum power capability, and/or as it reaches the point of "soft clipping" that tube amps are noted for. So I'm still not certain that a gain increase would be a solution.

Regards,
-- Al
Al, Not enough volume. Actually, not too bad with 96/24 files, but 192/24 sound flat, like they resist giving me their information even though I turn the volume up. I'm not getting distortion. So, diminishiing returns as the volume goes up.

BTW - For my Thiels they are using a custom 3-ohm tap.
Is the symptom that (a)you can't get enough volume with the volume control turned all the way up, or (b)that you encounter clipping and/or distortion on musical peaks with the hi rez material, at volume control settings that are below maximum?

If the latter, increasing the gain will not help, assuming (as is likely) that what is running out of headroom is the amplifier. The exact same problem would occur, at the same perceived volume level, except that you would reach that point at a lower setting of the control.

Regards,
-- Al
Thanks guys, It sure is adding up to new speakers, but the cash isn't adding up as quickly as I like to get there.

Swanney has a good point because my DAC's output is 2.5 volts compared to his 4.5 volts, while my phono stage is high ouput to work with any preamp.

I can try an alternative, as I heard back from the builder of my amp, Sam Kim. He thinks I can raise the gain by adding caps myself. I am pretty good with a soldering iron, so maybe worth a try.
Getting more efficient speakers is moving in the right direction.Kenny you`re making the correct choice if you plan to keep the current amplifier.
I suspect that Swanny's diagnosis is correct. Material that is not as dynamically compressed will have a lower average volume, assuming that peak volume is similar.

You can confirm that by using an audio editing program, such as the free program Audacity, to examine and analyze the sound waveforms of the respective files.

I suspect that what is limiting the dynamic range that can be handled by your system is the combination of a relatively low powered amp and inefficient speakers. If so, increasing the gains, sensitivities, or output levels of upstream components will not help, but changing to more efficient speakers would, of course.

Also, it appears that the 86 db sensitivity of your speakers is specified with respect to 2.83 volts, rather than 1 watt. 2.83 volts corresponds to 1 watt into 8 ohms, but to 2 watts into 4 ohms. Given the speaker's 4 ohm nominal impedance, its efficiency is therefore only 83 db for a 1 watt input, at a 1 meter distance.

Regards,
-- Al
Thanks Swanny, but my next major upgrade is to replace my seventeen year old Thiel speakers with much more sensitive ones. That will also fix the problem.
Maybe get a DAC with adjustable gain or higher output? Or, a pre with more gain? Or, would a step up transformer help increase output between your DAC and pre? My Bel Canto DAC3 puts out 4.5 v via XLR, so I get plenty of gain no matter the source.
This is profound in my system because my speakers are low impedance (2.5 to 4ohms) and insensitive (86db), plus I use a tube amp. I'm just fine with my analog front end, but I'm running out of headroom with digital.
I have noticed this too. I think it is because these files use much greater dynamic range. They are not as compressed as standard redbook files or CDs that you are used to listening too, where the overall level is mastered higher and closer to clipping (0 db) output. Perhaps someone with more technical or professional recording background can fill us in more?