Hifiman HM-801 RMAA Tests
May 9, 2010 at 4:54 PM Post #197 of 795

 
Quote:
I guess what I meant to say got interpreted at face value.hahaahah

 


Or perhaps it is a lesson to you not to post silly strawman arguments.
 
May 9, 2010 at 5:51 PM Post #198 of 795
Well Well, by chance I was having a careful listen to my 801 today and then stumbled upon this thread...
 
I suppose as an owner of ‘this sad, expensive joke of an antiquated, unusable player’ I would only naturally feel somewhat defensive and be a little taken aback with the comments and tests done by the OP. He is obviously is not very impressed...
 
Looking at the graphs I was also struck by the thought that he seems to be using an impedance of 16 Ohms across the output of what I presume is the standard headphone driver board. I am also not sure what actually is an ‘RMAA’ test, but there again there is much to learn.
 
A while back I borrowed a friend’s Lindos automatic test system that is a broadcast industry standard over here (UK) an had a look at various audio recording systems I have including my laptop mp3 coder and Cowon D2 DAP. When looking at the output I was very careful to use realistic loads which included my Grado and Sennheiser headphones as ‘real world’ loads for the Cowon. The fact that the OP used a 16 Ohm load suggests that he was using a system more suited to loudspeaker testing than the higher impedances (30 – 300 Ohms) typically to do with headphones, and so would not be giving the ‘antiquated unusable player’ a fair crack of the whip.
 
By coincidence my 3rdbatch 801 into Sen. IE8’s (32 Ohm) does have a slight but noticeable upper treble roll off in comparison to my Cowon D2. My own rough and ready tests consisted of playing some of my own recordings of live acoustic jazz (recorded on a Sony D50) copied to a pair of FLAC encoded SDHC cards, with each machine synced together. The players were held in one hand and the headphone jack swapped between the two machines.
 
However, when changing to the higher impedance HD650’s I could detect very little difference in upper treble quantity. This led me to conclude that the standard headphone driver board in the 801 was more suited to higher impedance phones (are there any 16 Ohm impedance headphones?). Further A/B tests of line output or Cowon verses HiFiman through a 2007 Slee Solo (PSU1) into HD650’s or IE8’s produced undetectable (to my ears) differences in frequency responses (levels were matched as close as possible by ear).  So I conclude that the poor HF response shown in the OP’s graphs was simply a result of driving the 801 into a too low impedance.
 
As regards how the 801 ‘sounds’ I can only remark that today it just sounded better than the Cowon D2 and my Laptop’s Foobar / M Audio Transit / Slee Solo combination, having what might be described a ‘smoother more refined presentation’ with ‘deeper and more defined’ lower bass (it just sounded better).
 
I must borrow my friend’s Lindos system again and do some objective tests on the 801 for myself.
 
May 9, 2010 at 6:02 PM Post #199 of 795
I think the Etymotic HF5 is 16 or 18 ohm and the ER4-P is 25 ohm
 
May 9, 2010 at 6:12 PM Post #200 of 795
Quote:
Originally Posted by Thermionic /img/forum/go_quote.gif
Looking at the graphs I was also struck by the thought that he seems to be using an impedance of 16 Ohms across the output of what I presume is the standard headphone driver board. I am also not sure what actually is an ‘RMAA’ test, but there again there is much to learn.


Other graphs posted by Nankai have clearly supported the graphs posted by the OP. Load is not the issue here.
 
And you really should learn what RMAA is before commenting further on the technical aspects of the test.
 
May 9, 2010 at 6:12 PM Post #201 of 795

According to my 4P box, the P are 27, and the S/B are 100 ohm :) don't know about the HF, but lower impedance would definitely make sense since they're more consumer-friendly.
Quote:
I think the Etymotic HF5 is 16 or 18 ohm and the ER4-P is 25 ohm



 
May 9, 2010 at 6:17 PM Post #202 of 795
The HifiMan has plenty of amplification, so it would be better to go with the ER4-S rather than the ER4-P.  Or just use a P to S adapter.
 
There are a few custom IEM's with very low impadence, the JH16 has an impedance of 18 Ohms, but the input sensitivity is 118dB at 1mW.
 
-Ed
 
May 9, 2010 at 6:33 PM Post #203 of 795


Quote:
 
There are a few custom IEM's with very low impadence, the JH16 has an impedance of 18 Ohms, but the input sensitivity is 118dB at 1mW.
 
-Ed

Mine customs are 119@1kHZ and 19 Ohms. The only player that doesn't hiss with them is s:flo2 :) however uDAC also is dead silent.
Though, as we are speaking of impedance. I also don't think that 16 Ohm is appropriate load for a player. Typical headphone amp is suited for 32 Ohms load. That said when you load player with 16 Ohms the half of power accumulate on headphone amp. 16 Ohm earphones can't be driven properly with such an amp. The result is i.e less bass amount.
 
 
May 9, 2010 at 7:06 PM Post #204 of 795


Quote:
As far as gross misinformation, I’ll let the measurements and opinion stated stand on their own as it’s more than what you have offered. If you have counter points please feel free to share them and if you call stating opinion trolling you should look at some of the outright defenders posts first as they are outright insults …


I was referring to your completely wrong information about the PCM1704 and you know it.  Also, if you had taken the time to read my post, you'll note that my criticism is of the conclusions made, which are misleading.  You both need to read up about how digital audio works.  Google "digital slow roll off filter" for a wealth of links explaining them.  A snipped from the very first link I found:
 
 
Quote:
A slow roll-off anti-imaging filter also has another consequence. Because of its slowroll-off rate, its impulse response exhibits less ringing and time smearing as compared
to a conventional sharp oversampled anti-imaging filter. Fourier analysis shows that thetime resolution and bandwidth of all filters are inversely related. A slow roll-off filter
allows more high frequency energy through compared to a fast roll-off filter. Therefore,the slow roll-off filter has better time resolution, or equivalently, less time smearing.

 
However, a big curve on a magnified graph certainly looks more impressive when you set out to trash someone's gear, as well as every high-end DAC out there.  However, your contempt for the people here and subsequent negative bias is quite clear from this post by dkft on your forum, in reply to a bunch of posts trashing the 801 because of its size:
 
Quote:
Yeh, I almost died laughing when I saw the photos of people using that ridiculous thing with the "balanced" RSA amp, or the nice looking, portable Pico Slim...

 

Exactly and I just wanted to express my support for this test and it has shown that although the Hifiman sounds good to me a generalized word like "hifi" should not be associated with it due to the roll-off, as I believe that the general consus on hifi is the reproduction of the music as closely to the original as possible. And I suppose we see clearly now that this is not the case. So the hifiman is best described as "musical" instead off staying it sounds "hifi", no? BTW: I also have to express my gratitude for choosing this thread title as it does not imply some off the bashing that has been written here. After all it is a product that does receive a huge amount of customers from people reading on this forum, so I'm glad that we can discuss the results of the test in an objective manner as we are. I know some threads where hype/bashing makes it extremely hard to come to a decision and I end up choosing not to buy a product at all because I'm more confused after reading the thread then before and I believe that this thread will clearify things for possible customers.


See what I posted above about filters.  The musicality of PCM1704-based DACs has to do with their lack of a "digital" sound, that is, instruments and vocals sound real, and not like a poor, digital reproduction.

Quote:
Nankai: I will only repeat what I already said: Anyone criticizing my method of testing should simply point out which one's the original file from CD, which one isn't recorded over my sound card. There's no need to bring fancy Wadia players and such into the equation, since this thread is not about them. My premise is very simple: there are four files - which one is which? Is there a difference, compared to the Hifiman, Clip, Cowon? Is the difference huge or small? Is something "better" or "worse" sounding for oneself, subjectively?
 
I only said that the Hifiman has less treble and sounds a bit more veiled than the other players. Statements I made about the Clip are blown way out of proportion. It's an average player with it's share of flaws, just like other players. It's just what I use the most, so I happen to compare it more than any of my other players.


I am criticising your conclusions, such as about the treble.  You said yourself on your own forum, "Sad that people can't accept facts."  You need to learn to accept some facts about how digital audio works, because while I applaud you for doing measurements, your understanding of the results you get is grossly lacking and, worse still, you are spreading your ignorance as facts.
 
If anyone wants to do a simple test to understand why the slow roll-off filter would have essentially zero effect when listening, find that "Measure your ears' frequency response" page and listen for yourself how hard it is to tell the difference in volume of high frequencies.  
 
May 9, 2010 at 7:59 PM Post #205 of 795


Quote:
Here's a little test for everyone interested.
So here are the four candidates, with a little description on what RMAA shows about those players - but of course you better listen for yourself instead of trusting graphs.

dfkt,
I'm sorry but I can't make the differences between the tracks the lower frequency seems to sound the same (was focusing on high frequency), the highest tone in your sample is the triangle and i would venture that it would hover in the 4-5-6khz range, I don't hear anything higher than that in your sample...
just being honest :wink:
***I usually hear 10khz @ -35 db no problem.
cheers
 
May 9, 2010 at 9:34 PM Post #206 of 795


Quote:
 

Or perhaps it is a lesson to you not to post silly strawman arguments.



Why would you consider it to be a lesson to me, when what i posted is a fact and not an argument.I'm not even attempting to argue for the Hifiman! It's not as if my life revolved around the player.
 
I see the CDP as overpriced because I do not think the components inside justify the 5499 USD asking price.
 
Therefore I do not think it is a fair price to me and hence will not purchase it,because i really think it should be priced at a way way lower price-which in turns helps me to save money.
 
oh and btw, weren't there comments on how nobody will care if the hifiman cost way lesser than the asking price of 790 USD for the quality it gave? i.e overpriced POS as some put it eloquently.
 
May 10, 2010 at 4:16 AM Post #207 of 795
EDIT: F-ed that up.
 
Quote:
dfkt said:

... You make the results, not me. And if anything sound better or worse to you is also up to you, not me.

I think this potentially objective approach is exactly what this thread needs.
 
My comment is that if you are having trouble noticing differences between the different tracks provided by DFKT imagine trying to hear differences when listening to them out and about. Also, if you can't easily determine between them, I don't think spending $790 USD on a player is going to do much for your listening experience, when a $50~ player sounds the same/similar to your ears.
 
May 10, 2010 at 5:17 AM Post #209 of 795
Quote:
Originally Posted by Hero Kid /img/forum/go_quote.gif
My comment is that if you are having trouble noticing differences between the different tracks provided by DFKT imagine trying to hear differences when listening to them out and about. Also, if you can't easily determine between them, I don't think spending $790 USD on a player is going to do much for your listening experience, when a $50~ player sounds the same/similar to your ears.
 


That's exactly what i thought while listening to dfkt's tracks... i'm really unable to tell the difference, except maybe for track 1 which sounds worse.
 

Users who are viewing this thread

Back
Top