[upstateguy]

Quote:

I heard a difference with the Musiland 02US + y2 (not sure about the 01). It really depends on the amp and headphones as well. I couldn't hear the differences with a couple of my previous amps. The only ones that were transparent enough to reveal differences were the M^3 and GS-1. I passed a blind test between the HiFace (stock) and USB of the y2. It required a lot of concentration though, since I'm used to forming impressions over long periods of time. I find that's a more effective way to discern subtle differences in audio.

Do you remember which headphones you used ?
 
USG


 

 

[Shahrose]

Quote:

Do you remember which headphones you used ?
 
USG


 

HD800, HE-5 and HD650 at the time. The HD800 and 650 were recabled. (For the blind test, I only used the 800s).

 

[upstateguy]

Quote:

I report no volume diff but "brightness" compared to optical from my MacBook! Just to make it simpler! MWUAHAHA

LOL...................... 
 
OK, but add another vote for the HF brightness factor. 
 
If you're really serious, and just not kidding around, do you find the brightness to be a "treble tilt", where the bass is less than the bass from your macbook's optical out and the treble is more, or is the brightness just a treble boost with the same bass as the macbook optical?
 
USG

 

[xdanny]

USG,
 
l don't know how you came to the conclusion that I'm using a macbook, my gear has always been listed in my sig...  Windows XP with Foobar2000... 
 
I do have the upsampling turned off on my dac, but it's not the hifaces's fault.  So I don't know why it is being criticized as such...  I prefer the upsampling off with other sources/transports also.  The less processing of my audio files the better.  Is the upsampling feature required to properly evaluate a component?  No.  Personally, I've never wanted it, never cared for it and I will never use it
 
Quote:

 
Ah, I see you're using a macbook.  That is why you hear no differences.  There are several people who report no differences in volume or being able to hear the brightness I spoke about, with macbooks.  Aimlink reports hearing no difference at all between his jkeny mod and the optical out of his mac book and Shamu also reports that he doesn't hear the volume or brightness issues with his. Now you're reporting the same thing.   I'm not sure why this is peculiar to macbooks but it seems to be consistently reported with them.
 
.....  It was just an additional illustration of the problem the HF encounters with uspampling  DACs....that the HiFace does not work well with upsampling dacs, is a fact  and even you have upsampling turned off because of it. 
 

 

[upstateguy]

Quote:

USG,
 
l don't know how you came to the conclusion that I'm using a macbook, my gear has always been listed in my sig...  Windows XP with Foobar2000... 
 
I do have the upsampling turned off on my dac, but it's not the hifaces's fault.  So I don't know why it is being criticized as such...  I prefer the upsampling off with other sources/transports also.  The less processing of my audio files the better.  Is the upsampling feature required to properly evaluate a component?  No.  Personally, I've never wanted it, never cared for it and I will never use it
 

Gee willikers.  I must have scrolled up to see what equipment you were using and accidentally landed on the wrong sig.  Sometimes it's hard to see the sigs when you  are posting.  Many apologies.  It clearly says Windows XP.  What kind of computer are you using?
 
We've been going back and forth but you are talking about your battery powered modification like it was a stock unit.  It is not.
 
I am discussing the stock USB powered unit  and you are not.  I'm sure you don't think your modded unit sounds like my stock unit.
 
Regarding upsampling.  I don't have it turned on with my Stello, but it seems to be built into my North Star and the Lavry 11, Benchmark, DacMagic, Bel Canto, Buffalo, Y2, and many others.
 
USG

 

[upstateguy]

Quote:

HD800, HE-5 and HD650 at the time. The HD800 and 650 were recabled. (For the blind test, I only used the 800s).

Well that pretty much settles that.

 
USG

 

[K3cT]

Quote:

I heard a difference with the Musiland 02US + y2 (not sure about the 01). It really depends on the amp and headphones as well. I couldn't hear the differences with a couple of my previous amps. The only ones that were transparent enough to reveal differences were the M^3 and GS-1. I passed a blind test between the HiFace (stock) and USB of the y2. It required a lot of concentration though, since I'm used to forming impressions over long periods of time. I find that's a more effective way to discern subtle differences in audio.

 
Interesting. I really should get different converters to further corroborate my findings. I used to do it your way, forming impressions after an extended period of time, but I realized that you're more subject to placebo effects in that manner so I've been trying to keep it shorter. What I did was I spent 2 days with the Musiland unit before doing an extensive blind test after I deemed myself familiar enough with the gear already. I'm all ears if anyone has a better reviewing method though as I'm still learning. 

 
Still, at the moment the ADUM4160 isolator is more interesting for me as it's way cheaper to experiment with. I could probably build the isolator and a nice simple linear regulated supply, either TREAD or Sigma25, and case it under $100... 
 

 

[sleepy dan]

I am interested .... Couldn't find your link but did find this from CA ....
 
"Simple... The Legato registers as a 16 bit device. Any 24 bit device will be louder as the 16 bit data is shifted left 8 bits when playing 16 bit material through a 24 bit interface.
Thanks
Gordon"

Is this the reasoning what you are referring to?  If so, it's mistaken - I don't think Gordon thought this through before posting.  According to this logic, 16 bit audio will be 2 to the power 8 = 256 times louder than 24 bit audio when played through a 24 bit DAC.  Obviously it's not, or the DAC would be clipping horribly.  I'm going away soon for a few days, so if you are interested I can go into more detail when I get back next week.  i think it's an important issue that needs ironing out.  But you can easily test this for yourself by comparing 16 bit and 24 bit recordings .... there should be no loudness difference, let alone one being 256 times louder than the other.

If you're interested, I posted the link to the entire CA Legato review and Gordon Rankin's response in that review, as well as links to the PS Audio discussion.  I also posted my correspondence with Mr. Rankin. 
 

 

[sleepy dan]

Upsampling with the Sabre can only be turned off by re-writing it's firmware.  But the Hiface has no "problem" with upsampling.
 
I have no problem with listing my system, other than 1) my system is entirely DIY and heavily modded, so it's hard to make a judgement on how transparent it is by comparing it to commercial stuff.  But it is very, very transparent. 2) I don't see how it's relevant to the current discussion and 3) I started building my own clock circuits when after-market clocks came on the market maybe 15 years ago or so.  I could easily hear the difference between circuits on my then modest speaker based system.  "You can't hear it because your system isn't good enough" that is often mentioned, particularly by regal, really should stop.
 
Quote:

 
I'm curious if you can disable upsampling on your Twisted Pear Buffalo DAC?
 
I'm also curious which Koss headphones you have and what speakers you have, as you have no profile for me to reference.... and I'm sorry if you have mentioned this already, but what is your source?
 
The other thing in question is whether the nuances that can be readily detected with headphones can be detected with speakers and at what volume and with what room treatments and at what distance?
 
As I said to xdanny, I'm not looking for a fight, just continuing the discussion.
 
 
USG

 

[leeperry]

 

  I don't think Gordon thought this through before posting.  According to this logic, 16 bit audio will be 2 to the power 8 = 256 times louder than 24 bit audio when played through a 24 bit DAC.  Obviously it's not, or the DAC would be clipping horribly.

 
What is highly sarcastic is that it's the same guy who sells his custom TAS1020B firmwares for stellar prices..

 

[shamu144]

To clarify my position

, I have never been a Hiface detractor. In fact, I love what this little device does in my system and how it opened my eyes to a new era of digital playback. However, I have been extremely disappointed by M2Tech folks attitude and handling of the "clock war" issue. Hence my decision to not deal in the future with any M2Tech product. As you can see, a very rational and expected consumer reaction.
Considering your skills and knowledge, I would be most interested to read your contribution in other threads if you ever decide to

 
Quote:

This follows a consistent pattern from those who have been detractors of the Hiface, of making personal insinuations.  That you do so says a lot about your position.
 
For the record, as my personal credentials have been called into question, I have no connection with M2Tech other than that I own a Hiface.  I have worked in electronics and software all my life, and have qualifications in both.  I am no guru - I have never worked in audio.  I just like making HiFI, especially digital electronics stuff (DACs).  Any other questions about where I am coming from, please ask.

 
 
You are abslotely right, I have no clue about what is going on in a DAC chip receiver. However, maybe some of those guys understand better what they are talking about:
 
On Audio Asylum, here posted by Thorsten: http://www.audioasylum.com/forums/pcaudio/messages/7/78654.html
 
"As things range from devices with so little output as to make several receivers not lock at all (Musiland Monitor series) to devices with so much output that most modern receivers scream for mercy (HiFace) and it heavily depends on the design of your DAC which device will have what undesirable consequences (they all do BTW) with your DAC."
 
Or the founder of Computer Audiophile in his Hiface review: http://www.computeraudiophile.com/content/M2Tech-hiFace-Asynchronous-USB-SPDIF-Converter-Review

"Note: As shown in the measurements below the hiFace's output voltage is 2.328 Vpp. This is higher than the standard 0.5 Vpp. It is possible the D-07 does not handle higher voltages as well as the Alpha DAC or DAC202. The bottom line is readers should look at the specs of their DAC and test components in person before purchasing."
 
Someone here made concrete jitter measurment of the Hiface and a few other devices, and also mention the issue: http://www.head-fi.org/forum/thread/446375/usb-24-192khz-m2tech-hiface/45#post_6063226
 
"The problem with some of the coupling is almost certainly due to the DAC output level that should applied concurrently to 0.5V pp and instead is equal to 2.74V pp. The designers are working to put in "template" level output signal while ensuring full compatibility with all DAC Marketing "
 
Quote:

Honestly, if you knew what happened at the input of a DAC, you would not be saying this.  2V S/PDIF will not upset a DAC.  Plenty of DACs actually boost the S/PDIF level up to the logic level maximum (3.3V or 5V) because the interface receiver likes a higher level.
 

 

[Sebhelyesfarku]

 
Quote:

Is this the reasoning what you are referring to?  If so, it's mistaken - I don't think Gordon thought this through before posting.  According to this logic, 16 bit audio will be 2 to the power 8 = 256 times louder than 24 bit audio when played through a 24 bit DAC.  Obviously it's not, or the DAC would be clipping horribly.  I'm going away soon for a few days, so if you are interested I can go into more detail when I get back next week.  i think it's an important issue that needs ironing out.  But you can easily test this for yourself by comparing 16 bit and 24 bit recordings .... there should be no loudness difference, let alone one being 256 times louder than the other.

A possible explanation is if the 16 bits are shifted up in the 24bit frame, so the lower 8 bits are filled with 0s, the 16bit max value (0dBFS) will become max value - 256 in 24bit. Or not.

 

[sleepy dan]

Does anyone find 16 bit data to be 256 times louder than 24 bit?  Or vice versa for that matter?  I'd be happy to read Gordon's explanation in full if I can find it in case there's something I'm missing, but i really think he's had a brain-fart on this one. 

  No disrespect intended - he's a very inventive designer.
 
Quote:

 

A possible explanation is if the 16 bits are shifted up in the 24bit frame, so the lower 8 bits are filled with 0s, the 16bit max value (0dBFS) will become max value - 256 in 24bit. Or not.

 

[upstateguy]

Quote:

 
Interesting. I really should get different converters to further corroborate my findings. I used to do it your way, forming impressions after an extended period of time, but I realized that you're more subject to placebo effects in that manner so I've been trying to keep it shorter. What I did was I spent 2 days with the Musiland unit before doing an extensive blind test after I deemed myself familiar enough with the gear already. I'm all ears if anyone has a better reviewing method though as I'm still learning. 

 
Still, at the moment the ADUM4160 isolator is more interesting for me as it's way cheaper to experiment with. I could probably build the isolator and a nice simple linear regulated supply, either TREAD or Sigma25, and case it under $100... 
 

 
I have to agree with you.  I prefer short sessions.
 
USG


 
Quote:

I am interested .... Couldn't find your link but did find this from CA ....
 
"Simple... The Legato registers as a 16 bit device. Any 24 bit device will be louder as the 16 bit data is shifted left 8 bits when playing 16 bit material through a 24 bit interface.
Thanks
Gordon"

Is this the reasoning what you are referring to?  If so, it's mistaken - I don't think Gordon thought this through before posting.  According to this logic, 16 bit audio will be 2 to the power 8 = 256 times louder than 24 bit audio when played through a 24 bit DAC.  Obviously it's not, or the DAC would be clipping horribly.  I'm going away soon for a few days, so if you are interested I can go into more detail when I get back next week.  i think it's an important issue that needs ironing out.  But you can easily test this for yourself by comparing 16 bit and 24 bit recordings .... there should be no loudness difference, let alone one being 256 times louder than the other.

I wrote to him about that comment and he responded with the essentially the same answer, so he's had a chance to think about what he was saying. 
 
(His response is posted a few pages back in this thread or the other HF thread...  you might search the threads for gordon rankin)
 
He seems to be a very nice guy and I'm sure he would write back to you.  Why don't you share your computations him at Wavelength Audio.com. 
 
I'm as curious as you are to know why the HF plays louder if that's not the reason. 
 
 
Quote:

Upsampling with the Sabre can only be turned off by re-writing it's firmware.  But the Hiface has no "problem" with upsampling.
 
I have no problem with listing my system, other than 1) my system is entirely DIY and heavily modded, so it's hard to make a judgement on how transparent it is by comparing it to commercial stuff.  But it is very, very transparent. 2) I don't see how it's relevant to the current discussion and 3) I started building my own clock circuits when after-market clocks came on the market maybe 15 years ago or so.  I could easily hear the difference between circuits on my then modest speaker based system.  "You can't hear it because your system isn't good enough" that is often mentioned, particularly by regal, really should stop.
 

According to what jkeny posted and PS Audio,  upsampling neutralizes the effects of using a low jitter source like the HF, and could possibly add jitter.  
 
That you don't hear the increased volume of the HF seems to indicate that somewhere in your system that effect is being neutralized or that you don't have another transport like the BCT to compare it to.

In any event, I'm glad the HF is working well in your system, I have removed it from mine.
 
What were you using before the HF and what have you had a chance to compare it to?
 
Btw, would it be so hard for you to fill out your profile so I can reference your equipment while I'm responding to your posts?
 
Thanks
 
Quote:

Does anyone find 16 bit data to be 256 times louder than 24 bit?  Or vice versa for that matter?  I'd be happy to read Gordon's explanation in full if I can find it in case there's something I'm missing, but i really think he's had a brain-fart on this one. 

  No disrespect intended - he's a very inventive designer.
 

 

Write to him about it.  I want to hear about the brain-fart too.
 
 

 

[Sebhelyesfarku]

 
Quote:

Does anyone find 16 bit data to be 256 times louder than 24 bit?  Or vice versa for that matter?  I'd be happy to read Gordon's explanation in full if I can find it in case there's something I'm missing, but i really think he's had a brain-fart on this one. 

  No disrespect intended - he's a very inventive designer.
 

That could be the point, it's not 256x, it's just 16,777,216 – 256, the low 8bit is 0. Why would that be 3dB is another question. Gordon is a digital audio guru floating over an asynch interface in lotus position, maybe mortal Head-Fi-ers simply couldn't understand the divine logic.