Head-Fi.org › Forums › Equipment Forums › Sound Science › Hifiman HM-801 RMAA Tests
New Posts  All Forums:Forum Nav:

Hifiman HM-801 RMAA Tests - Page 43  

post #631 of 795

Hello.
 

 

I am from Japan.
So, Please forgive my poor English. :)

 

For a personal interest,
I roughly measured the setting range of the equalizer of HiFiMAN.
It would be greatly appreciated if it helps your EQ setting.

 

Grazie.

 

http://sonove.angry.jp/HiFiMAN_HM801/ARTA_FR/PCM_D50/HM801_EQ_Range.gif

post #632 of 795

 

Quote:

Hifi man lovers posted alot of nonsensical hear it comments and attacked other users in the process ... etc to prove theirs.

 

If you look at some of the graphs posted you would know that the reason for the roll off is due to the implementation of a filter that reduces ringing. The only reason why I promote the radical idea of listening to it before passing judgment on its performance was that the treble roll off is hardly noticeable, and the theoretical (and audible?) improvement to transient response it brings perhaps makes it worth the -3 dB roll off.

post #633 of 795

Quote:

Originally Posted by FrederikS|TPU View Post

 

If you look at some of the graphs posted you would know that the reason for the roll off is due to the implementation of a filter that reduces ringing. The only reason why I promote the radical idea of listening to it before passing judgment on its performance was that the treble roll off is hardly noticeable, and the theoretical (and audible?) improvement to transient response it brings perhaps makes it worth the -3 dB roll off.

 

It's actually a -4dB roll off.

 

It rolls of 2dB easily before 16K even.  We've covered the transients - the odds of them be audible is low to non-existant.  On the other hand, I can guarantee the roll-off is :|

post #634 of 795
Quote:
Originally Posted by Shike View Post

Quote:

Originally Posted by FrederikS|TPU View Post

 

If you look at some of the graphs posted you would know that the reason for the roll off is due to the implementation of a filter that reduces ringing. The only reason why I promote the radical idea of listening to it before passing judgment on its performance was that the treble roll off is hardly noticeable, and the theoretical (and audible?) improvement to transient response it brings perhaps makes it worth the -3 dB roll off.

 

It's actually a -4dB roll off.

 

It rolls of 2dB easily before 16K even.  We've covered the transients - the odds of them be audible is low to non-existant.  On the other hand, I can guarantee the roll-off is :|



Actually it is by no means certain that many people would be able to hear the roll-off. Experiments back in 1978: "Sampling-Frequency Considerations in Digital Audio" , TERUO MURAOKA, YOSHlHlKO YAMADA, AND MASAMI YAMAZAKI provides evidence that a roll-off can be as low as 15K before it is reliably audible, a hard cut-off at 20K, 18K and 16K was not reliably audible (p < 0.05) by any of the 31 trained subjects and only some of them detected the 14K cut-off , in my own experiments I was unable to detect anything above a 13K roll-off, but I am 51 so that is explicable.

post #635 of 795

^^ you look pretty good for 51 mate. Maybe too good.

post #636 of 795

 

Quote:
I think anything audio has to be heard before you pass judgment on it

 


This statement has a lot of wisdom.  

 

There were/are and will be lots of examples of using scientific facts to prove something but ultimately the proof is futile.  It's not because the facts were/are wrong but there are other considerations neglected or misconceptions.  

 

This is a post in the science forum so I would like to put in more terms that are used in science.

In terms of Maths:

I remember someone said something like "once the device rolls off, you can never get the original data back".  This is not exactly true because in principle one can design a device that have the frequency response as the inverse function of the roll off, then the original can be recovered.  Similarly, someone said that over-sampling adds pseudo-data into the original data and that could not make the output sound better is not always true.  Ultimately we want to recover the original wave pattern which is continuous in the time domain.  The PCM is discrete in the time domain.  If the pseudo-data added is actually the same as the original wave pattern at that time, the output can be better.

 

In terms of engineering:

A headphone that has treble roll off coupled with an amp that over-emphasis the treble may produce pleasing results--in fact should be better than when the same headphone coupled with an amp that has a totally flat frequency response.  This is why the term "synergy" exist in the real world and basically why 1st class device should be matched with 1st class devices.

  

In engineering, basically everything is a compromise due to the limitation of the technology or the environment.  For example, why the cut-off sampling frequency of the red-book CD is 44.1 khz?  I know, I know -- everyone said, "Human can hear only up to 20k, the higher range of the frequency spectrum is useless".  However, in the real world higher frequency does exist, and an aliasing frequency will enter the lower frequency spectrum due to the 44.1k sampling rate.  For example, a 24k signal will become a 2k signal using the red book coding.  The SACD or the DVD-A format are some of the ways to overcome this aliasing problem.  Why they could not replace the CD is another matter.

 

When CD just came out in the 80's, lot of people were speculating that vinyl would be dying quickly because the SNR(or other parameters) of CDs were much better.  Now the history has proven that the speculation was wrong.  It's not the fact of the SNR was wrong but they have overlooked other considerations.

 

This shows that using only one parameter/test to make a conclusion could be very wrong.

 

In computer science:

We use benchmark programs to guess the performance of different computer systems.  But the ultimate test is to run the intended applications on the systems and pick the system that runs quickest/ best/steadiest whatsoever.  This demonstrates the limitation of the benchmark programs.  

 

 

 

What I'm trying to say is as a thread in the science forum, we should be open-minded and try to solve the contradicting facts rather than just take side.  There were posts from the Hifi man lovers that were never answered by the other side.  For example, someone posted some high end DAC does roll-off in the treble and then conclude that hi-fi is not equivalent to the flat FR response.  Note, the roll off of the other DAC is a fact.  So the haters need to prove the other DAC is not hi-fi in order to establish the conclusion that HiFiman is bad because it rolls off in the treble.  I’ve never seen such proof.  Another post said that the RMAA of the recorded output from a sound card cannot be used solely to determine the SQ of a device--otherwise all equipment can be tested that way and there is no need to have impressions, reviews, meets, etc.  This is a very strong argument because most of the head-fiers believes in honest impressions rather than just numbers/graphs.  Whether one can tell if an impression is honest is another matter.  The existence of the meets, reviews, impressions rather than just downloads of the RMAA testes of different equipment shows the RMAA is also a kind of benchmark programs and cannot supersede real experiences.

 

My own thought:

I accept that the FR response of the HiFiMan 801 is less than perfect and I truly thanks dfkt for his hard work.  Also I accept that the HiFiMan 801 sounds superb in general from the impressions of most of the head-fiers who have tried the unit and truly thanks for their impressions.  Quite a few of these head-fiers have long time track record, have high reputation and might have dispute Head-Direct before and they all provide excellent feedback on the SQ of the HiFiMan,  Certainly, these two facts (bad FR and great impressions) seems contradicting.  Then we should try to find the reasons why such discrepancy can be existed.  This may lead us to design a better DAP in the future and make this thread much more useful for the other head-fiers.

 

I shall end my sharing by the following case:

Initially many people thought that the FR of a loudspeaker should be as flat as possible.  This approach was then applied to the design of headphones and it was by ear found that a flat FR headphone would produce funny sounds.  If the dead-core engineers followed that the flat FR MUST be the holy grail in headphones rather than following their ears, the performance of the headphones nowadays would not be as good.  Now we know that the influence of the ear and a headphone with flat response would not be ideal and the contradiction was resolved.  

 

 

 

post #637 of 795


 

Quote:
Originally Posted by GreenLeo View Post

 

 


This statement has a lot of wisdom.  

 

There were/are and will be lots of examples of using scientific facts to prove something but ultimately the proof is futile.  It's not because the facts were/are wrong but there are other considerations neglected or misconceptions.  

 

This is a post in the science forum so I would like to put in more terms that are used in science.

In terms of Maths:

I remember someone said something like "once the device rolls off, you can never get the original data back".  This is not exactly true because in principle one can design a device that have the frequency response as the inverse function of the roll off, then the original can be recovered.  Similarly, someone said that over-sampling adds pseudo-data into the original data and that could not make the output sound better is not always true.  Ultimately we want to recover the original wave pattern which is continuous in the time domain.  The PCM is discrete in the time domain.  If the pseudo-data added is actually the same as the original wave pattern at that time, the output can be better.

 

In terms of engineering:

A headphone that has treble roll off coupled with an amp that over-emphasis the treble may produce pleasing results--in fact should be better than when the same headphone coupled with an amp that has a totally flat frequency response.  This is why the term "synergy" exist in the real world and basically why 1st class device should be matched with 1st class devices.

  

In engineering, basically everything is a compromise due to the limitation of the technology or the environment.  For example, why the cut-off sampling frequency of the red-book CD is 44.1 khz?  I know, I know -- everyone said, "Human can hear only up to 20k, the higher range of the frequency spectrum is useless".  However, in the real world higher frequency does exist, and an aliasing frequency will enter the lower frequency spectrum due to the 44.1k sampling rate.  For example, a 24k signal will become a 2k signal using the red book coding.  The SACD or the DVD-A format are some of the ways to overcome this aliasing problem.  Why they could not replace the CD is another matter.

 

When CD just came out in the 80's, lot of people were speculating that vinyl would be dying quickly because the SNR(or other parameters) of CDs were much better.  Now the history has proven that the speculation was wrong.  It's not the fact of the SNR was wrong but they have overlooked other considerations.

 

This shows that using only one parameter/test to make a conclusion could be very wrong.

 

In computer science:

We use benchmark programs to guess the performance of different computer systems.  But the ultimate test is to run the intended applications on the systems and pick the system that runs quickest/ best/steadiest whatsoever.  This demonstrates the limitation of the benchmark programs.  

 

 

 

What I'm trying to say is as a thread in the science forum, we should be open-minded and try to solve the contradicting facts rather than just take side.  There were posts from the Hifi man lovers that were never answered by the other side.  For example, someone posted some high end DAC does roll-off in the treble and then conclude that hi-fi is not equivalent to the flat FR response.  Note, the roll off of the other DAC is a fact.  So the haters need to prove the other DAC is not hi-fi in order to establish the conclusion that HiFiman is bad because it rolls off in the treble.  I’ve never seen such proof.  Another post said that the RMAA of the recorded output from a sound card cannot be used solely to determine the SQ of a device--otherwise all equipment can be tested that way and there is no need to have impressions, reviews, meets, etc.  This is a very strong argument because most of the head-fiers believes in honest impressions rather than just numbers/graphs.  Whether one can tell if an impression is honest is another matter.  The existence of the meets, reviews, impressions rather than just downloads of the RMAA testes of different equipment shows the RMAA is also a kind of benchmark programs and cannot supersede real experiences.

 

My own thought:

I accept that the FR response of the HiFiMan 801 is less than perfect and I truly thanks dfkt for his hard work.  Also I accept that the HiFiMan 801 sounds superb in general from the impressions of most of the head-fiers who have tried the unit and truly thanks for their impressions.  Quite a few of these head-fiers have long time track record, have high reputation and might have dispute Head-Direct before and they all provide excellent feedback on the SQ of the HiFiMan,  Certainly, these two facts (bad FR and great impressions) seems contradicting.  Then we should try to find the reasons why such discrepancy can be existed.  This may lead us to design a better DAP in the future and make this thread much more useful for the other head-fiers.

 

I shall end my sharing by the following case:

Initially many people thought that the FR of a loudspeaker should be as flat as possible.  This approach was then applied to the design of headphones and it was by ear found that a flat FR headphone would produce funny sounds.  If the dead-core engineers followed that the flat FR MUST be the holy grail in headphones rather than following their ears, the performance of the headphones nowadays would not be as good.  Now we know that the influence of the ear and a headphone with flat response would not be ideal and the contradiction was resolved.  

 

 

 


Awesome post man.

post #638 of 795

Here's a picture of the HiFiMAN's PCB board that have put together by machines and not hands:

67635

post #639 of 795


 

Quote:
Originally Posted by ajkda View Post

I don't get it.

 

Hifi man haters posted alot of graphs and charts to prove their point.

 

Hifi man lovers posted alot of nonsensical hear it comments and attacked other users in the process ... etc to prove theirs.

 

Who will you believe?

 

 

 

I'll believe the ones who have actually heard the hifiman, be it good or bad I'll always take the opinion of someone with actual listening experience with a piece of equipment as opposed to looking at some graphs and try and determine the SQ from graphs and charts. Graphs, charts and measurements simply will not tell you how something sounds, you have to actually hear it for yourself.
 


Edited by swanlee - 5/30/10 at 6:14am
post #640 of 795

Correct me if I'm wrong but that PCB looks like its been soldered by hand. There's even a pair of bridged pins on the top left.

 

The two channels are virtually crammed up against each other, which, if the layout isn't effective, would explain the cross-talk. The digital side of the PCM1704s have power decoupling ceramics bypassed with smaller yet further away ceramics, which is just a waste of parts for a package having the parasitic inductance of a SOIC20. The bottom channel seems to be missing a (useless as mentioned) bypass cap.

 

That said, there is use of high quality parts around the DACs and analog stage.

post #641 of 795

How about another question, WHAT is particularly innovative or new in the HFM that would justify the $800 cost?

 

The numbers show even if it does sound good, it probably doesn't sound $800 good.

post #642 of 795
Quote:
Originally Posted by swanlee View Post


 

 

 

I'll believe the ones who have actually heard the hifiman, be it good or bad I'll always take the opinion of someone with actual listening experience with a piece of equipment as opposed to looking at some graphs and try and determine the SQ from graphs and charts. Graphs, charts and measurements simply will not tell you how something sounds, you have to actually hear it for yourself.
 

 

Science, mathematics, and physics, simply will not tell you how something flys, you have to actually fly it yourself!

 

I guess?

 

I suppose to determine whether something sounds pleasing or not one must make subjective remarks based on interpretation.  But I think the $800 issue becomes more acute given the numbers above.

post #643 of 795


 

Quote:
Originally Posted by Trogdor View Post



 

Science, mathematics, and physics, simply will not tell you how something flys, you have to actually fly it yourself!

 

I guess?

 

I suppose to determine whether something sounds pleasing or not one must make subjective remarks based on interpretation.  But I think the $800 issue becomes more acute given the numbers above.

 

 

So something has to measure a certain way to be worth 800$? Science math and physics cannot tell me if I will personally enjoy the sound quality of any piece of audio equipment. It's pretty simple if people  don't deam a piece of audio equipment worth the price they won't buy it and it won't be made anymore.
 

post #644 of 795
Quote:
Originally Posted by b0dhi View Post

Correct me if I'm wrong but that PCB looks like its been soldered by hand. There's even a pair of bridged pins on the top left.

 

The two channels are virtually crammed up against each other, which, if the layout isn't effective, would explain the cross-talk. The digital side of the PCM1704s have power decoupling ceramics bypassed with smaller yet further away ceramics, which is just a waste of parts for a package having the parasitic inductance of a SOIC20. The bottom channel seems to be missing a (useless as mentioned) bypass cap.

 

That said, there is use of high quality parts around the DACs and analog stage.


Thanks for taking a closer look. The bridged pins surprised me quite a bit, is this how it's supposed to be or an "epic fail"? 

post #645 of 795
Quote:
Originally Posted by ajkda View Post
I don't get it.

Hifi man haters posted alot of graphs and charts to prove their point.

Hifi man lovers posted alot of nonsensical hear it comments and attacked other users in the process ... etc to prove theirs.

Who will you believe?

 

I am in the Hi-FiMAN lovers group, and I don't care what scientific graphs are showing me, 

The bottom line IMO is the best sound that I hear in my ears, and all the graphs and scientific data that you can muster

doesn't change my opinion on the sound.

In the end, it's all about the sound. And the 801 sounds spectacular.  This is the end justifying the means.

What good is a graph if you don't care for the sound? 

Is it rolled off? Yes, the measurements seem to qualify this. But dollars to donuts, I want to hear something that fulfills my needs.

And I feel that the price is justified by this. Using better op-amps and buffers in a proper configuration to obtain this, is working.

So, you can say all that you want about ringing and rolloff, but bottom line, the 801 sounds better than any other DAP that I have

had the pleasure to listen to. And to me, that's what counts.

It's easy to play armchair quarterback, but in the end, it's either a win or a loss in the standings, and for me, the 801's are a big win.

In sports, a year later, when people look at the standings, no one remembers how the game was one by walking in the last runner

in the bottom of the 9, what they do see, is a "1" in the win column.

So tear it apart all that you want, analyze the graphs 6 ways 'til Sunday. I want the best sound that my hard earned money

can purchase, and I'll never have remorse about buying them, because they fulfill my needs and my needs are simple.

Give me a player that has versatility, the ability to upgrade, and the best sound that I can find.

Will there be a player that does the same options or even more that sounds better?

If there is, I will save my pennies and buy that one.

 

The HM-801 sounds better than any other Dap that I have tried, and I will continue to enjoy it immensely, until something 

better comes down the pike.

 

I was at CES this year, and there were many chinese manufactures who wanted to buy it, reverse engineer the technology,

and try to come out with something competitive . That speaks volumes about the HiFiMAN. It is a compliment.

 

To Me, with all of the options, no other player can touch the 801 with a 10' pole. So enjoy you graphs, and I will enjoy my music.

 


 

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 › Hifiman HM-801 RMAA Tests