Head-Fi.org › Forums › Equipment Forums › Sound Science › Skeptico Saloon: An Objectivist Joint
New Posts  All Forums:Forum Nav:

Skeptico Saloon: An Objectivist Joint - Page 20  

post #286 of 1671
Quote:
Originally Posted by miceblue View Post
 

Gah, sometimes audiophool jargon confuses the heck out of me.

 

Amp A from company X sounds neutral. Then when company X releases a new amp, amp B, it sounds even more neutral. Wat?

 

Amp C has Y power and amp D has 2Y power. Headphone H sounds much better on amp D and the only conclusion is because headphone H "scales up" and can "benefit" from the 2Y power that amp D offers.

 

Portable media player P has DAC 1 and amp 1. Portable media player Q has DAC 2 and amp 2. Portable media player Q sounds much better because of amp 2. Wat? Didn't these people go to elementary school and learn about the scientific method's 3 fundamental variables (controlled, manipulated, responding)? You can't compare P and Q and draw a conclusion that Q sounds better because of amp 2 when they're both different players all together. Ugh....people.

 

Replace the variables with names of people, and you've got society.

post #287 of 1671
Quote:
Originally Posted by miceblue View Post

 

Amp A from company X sounds neutral. Then when company X releases a new amp, amp B, it sounds even more neutral. Wat?

Hah yeah, that's indeed a problem with transparent amps.

 

Fine companies will make amps that sound pretty much the same within reasonable limits and differentiate their products by size, looks/design, features, max power, gain factor(s), connectors etc.

 

Others will botch the sound because they think they can only sell customers an upgrade if it sounds different. Note: different is supposed to be better...

 

 

Quote:
Amp C has Y power and amp D has 2Y power. Headphone H sounds much better on amp D and the only conclusion is because headphone H "scales up" and can "benefit" from the 2Y power that amp D offers.

Evidence of failure to do a proper level-matched comparison and that bias changes what you hear (moa powahhh!! it has to be better! oh I can hear it, yay!).

 

Quote:

Portable media player P has DAC 1 and amp 1. Portable media player Q has DAC 2 and amp 2. Portable media player Q sounds much better because of amp 2. Wat? Didn't these people go to elementary school and learn about the scientific method's 3 fundamental variables (controlled, manipulated, responding)? You can't compare P and Q and draw a conclusion that Q sounds better because of amp 2 when they're both different players all together. Ugh....people.

Adding to that: assuming that none has audible flaws and very similar output impedance we're back to the previous point I made.

 

Amp 2 may have a bit higher gain, or a bit lower output impedance, which makes it louder = usually perceived as better.


Edited by xnor - 10/22/13 at 7:39am
post #288 of 1671
Quote:

Originally Posted by xnor View Post

 

Evidence of failure to do a proper level-matched comparison and that bias changes what you hear (moa powahhh!! it has to be better! oh I can hear it, yay!).

 

The more powerful amp is likely to also have higher gain, and will be perceived to sound better simply because it is louder without accurate level matching (setting the volume "by ear" is biased towards equalizing the position of the volume controls, rather than the actual loudness). The practice of selling amps with unnecessarily high gain, and DACs with a high output level, can be thought of as a form of the "loudness war" for audio devices.

post #289 of 1671
Quote:
Originally Posted by stv014 View Post
 

The practice of selling amps with unnecessarily high gain, and DACs with a high output level, can be thought of as a form of the "loudness war" for audio devices.

 

on the comparison, on that practice


Edited by xnor - 10/22/13 at 12:32pm
post #290 of 1671

Awesome....

I just now used the Dynamic Range meter on a few files.

 

CD rip track: DR11

HD Tracks track: DR10

 

 

I don't know what the heck the HD Tracks version did to the track, but I can clearly hear a difference.

Foobar2000 ABX Log (Click to show)
foo_abx 1.3.4 report
foobar2000 v1.2.9
2013/10/22 10:27:20

File A: C:\Users\Michael\Desktop\01 Crazy.m4a
File B: C:\Users\Michael\Desktop\03 Crazy.m4a

10:27:20 : Test started.
10:29:31 : 00/01  100.0%
10:30:23 : 01/02  75.0%
10:30:31 : 02/03  50.0%
10:30:37 : 03/04  31.3%
10:31:14 : 04/05  18.8%
10:31:25 : 05/06  10.9%
10:31:30 : 06/07  6.3%
10:31:38 : 07/08  3.5%
10:31:45 : 08/09  2.0%
10:31:49 : 09/10  1.1%
10:31:53 : 10/11  0.6%
10:31:59 : 11/12  0.3%
10:32:03 : 12/13  0.2%
10:32:13 : 13/14  0.1%
10:32:26 : 14/15  0.0%
10:32:44 : 15/16  0.0%
10:32:48 : 16/17  0.0%
10:32:52 : 17/18  0.0%
10:32:59 : 18/19  0.0%
10:33:08 : 19/20  0.0%
10:33:13 : 20/21  0.0%
10:33:18 : 21/22  0.0%
10:33:19 : Test finished.

 ----------
Total: 21/22 (0.0%)

I didn't even use a fancy setup for this test: MacBook Pro -> V-MODA Crossfade M-100.

 

Track A had the vocals more centered, whereas in track B it was slightly to the left. I used the Replay Gain option to make sure they were about the same volume level. Track B has a more stereo L/R kind of sound to it, whereas A sounds distant and it soundstage is long and reverberant as if some sort of DSP effect was added. Maybe this reverberant effect is why people say HD tracks "sound better?"

 

Then again, I'm comparing two different album's shared track, so that might be a factor. Dang it. I thought I was on to something. So much for my last post. XD


Edited by miceblue - 10/22/13 at 10:41am
post #291 of 1671

For future reference, "about the same level" doesn't cut it, you would do well to have them 0.5dB apart at maximum. However if the reverb/DSP is that easily audible there's little to no reason to conduct a blind test. You don't have to blind taste an apple in comparison to manure, only between various different flavours of apple and maybe a really mutated pear.

 

Also, I never found the dynamic range meter to correlate with better sounding mix or better mastering, it's just a silly simplistic figure just like megapickles and gigahurtz, or frequency ranges without tight tolerances. Even though using very limiting compression is not wise, restricting the record to a few MSB from the available 16 ones, I find this elitism surrounding dynamic range silly (which coincidentally has also created a craze over storage formats like DSD and 24/196).

 

As far as automatic room EQ goes, slapping a surround set in an un-treated room and assuming that you can fix it with an automated routine is far fetched, even if you averaged the measurement between a few different spots. Nothing is going to fix the room going modal resulting in nulls, and long reverb times are still going to stay. You can only flatten the frequency response by a certain amount, the largest gains are in placement and acoustic treatment, and it might've been placebo since we didn't blind test it but delay played a huge part in the surround experience, more so than the levels between channels.


Edited by Cat Face - 10/23/13 at 1:22am
post #292 of 1671
Quote:
Originally Posted by Cat Face View Post
 

For future reference, "about the same level" doesn't cut it, you would do well to have them 0.5dB apart at maximum.

 

0.1 dB in fact, a difference of 0.5 dB is audible.

post #293 of 1671
I don't exactly know what ReplayGain does, so that's why I mentioned "about the same;" it was mentioned elsewhere that it allows for volume-matching between two tracks.


As for another audiophool term, what is "drive" supposed to describe? "Oh yes, with the X amp, your Y headphones should be properly "driven;" or "unless your Y headphones are properly "driven," they won't sound very good.

I came across this term for the AKG K 701 and consequently, I was expecting them to sound like garbage when "driven" from my Clip Zip or iPod Touch 2G. I didn't even remotely find that to be the case.
Edited by miceblue - 10/23/13 at 9:29am
post #294 of 1671

ReplayGain (version 2) in foobar2000 is based on EBU R-128 which is a very good loudness algorithm. R128Gain is a nice standalone (cross platform) tool too. Atm it's probably the best way to match loudness of multiple tracks automatically.

 

 

I find "properly driven" curious too. The only way German speaking headphones nuts use "drive" is in, for example, the question: "What amp(s) do you use to drive your headphone(s)?"

 

I guess that there are a couple of unwritten rules for an amp to "properly drive" your headphones:

- the better the headphone the more expensive the amp needs to be (it's not uncommon to see people with $200 headphones using >$200 amps)

- higher gain than necessary

- more output power than necessary

- marketing material needs to be targeted at audiophiles

- objective performance doesn't matter as long as it sounds nice to one's biased perception (and to a real audiophile it does not if one of the above points aren't met :p)

...


Edited by xnor - 10/23/13 at 9:59am
post #295 of 1671

Some non-objectivists are after this concept of inter-gadget "synergy". While digging tube amplifiers is okay if you like the sound, I find it ludicrous how it oftentimes seems that coupling differently flawed/non-transparent gear with each other is seen as "real hi-fi" and objectivists who at least conduct some informal blind tests get debunked as people who "only read theory" and lack "real world experience". Silly anecdotes like "non-owners speak the loudest" are often used to try discredit any reasonable questioning of their methodology or mindset regarding criticism towards masturbatory glorification of inferior gear and cable swapping.

post #296 of 1671

Nobody needs to hear (although some of you may have) a SRH1x40 to understand that the 2% - 3% distortion (dominantly 3rd harmonic) that ranges up into to the midrange sounds terrible.

 

Proud owners are blind to that and reason.


Edited by xnor - 10/23/13 at 11:13am
post #297 of 1671

What do you guys think about this?

 

Versus

post #298 of 1671

The first looks like a certain inexpensive closed headphone, an on ear. Peaking at around 9khz or so, and lacking in the upper-mid to low-treble, pretty common as measured. Surprisingly smooth in the bass and mids. As a disclaimer, Tyll's head should be taken with a few grains of margin past 8khz. Looks nice.

 

The second looks like a certain not-quite-sealed up headphone that happens to cost a pretty penny, closed over-ear. Bassy as all hell, much smoother in the mid to treble, with a much lower rise in the 7-8khz area for "sparkle". Good, not necessarily my flavor.

 

Not much can be said about distortion on either without knowing the level of each harmonic, except that it probably won't be bothersome with the low levels on either. Sensitivity plays nice with portable players, power handling seems good.

post #299 of 1671

The former one lacks a light downwards slope, boost around 3 kHz and has a treble peak. Not enough energy above 12 kHz.

 

The second one probably sounds bassier than it really is due to the dip at 500-600 Hz, seems to have better treble extension but also lacks a bit around 2 kHz. Dip at ~10 kHz seems weird, probably no "sparkle" at all.

post #300 of 1671

Hm, interesting to hear your impressions. The former is $30 USD, the latter is $2000 USD. I remember reading somewhere on HeadRoom's website that a typical headphone graph "should" have a flat frequency response up until 1 kHz, in which it should then slope downwards. I guess the slope for the former graph is too steep. For $30 though, I thought it measured pretty dang well.

 

I just saw this:

 

and it makes me wonder why people design DACs and amps with cases that are much larger than the actual components within the case.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
This thread is locked  
Head-Fi.org › Forums › Equipment Forums › Sound Science › Skeptico Saloon: An Objectivist Joint