O2 AMP + ODAC - Page 66

Quote:

Originally Posted by HamilcarBarca 

 

Not many people can hear frequencies as high as 22,050 Hz which is the theoretical limit -- vaguely speaking -- of a 44.1kHz signal. Increasing the sampling rate to 88.2kHz means, again in theory, the DAC can reproduce frequences up to 44.1kHz. However, humans can't hear those frequencies. There's essentially no benefit to the listener.

 

Why might another manufacturer support 88.1kHz or even 192kHz (which has zero benefit to humans)? Ask them why. Don't forget to mention Nyquist rates. I predict the response, if any, will be "science doesn't apply to our products."

Yes, I was about to add to my last post that I might not notice the difference between 44.1kHz and 88.2kHz (or 96 kHz) sampling rates.

This topic has  been all over this site and the Internet.

But, I still do not understand why do  HDTracks.com and others sell 176.4 KHz and 196 kHz tracks? The only reason seems to be  that they want to make more money,

because albums in very high resolution (176.4 KHz and 196 kHz) cost more,

usually $30 vs. $20 (96 kHz, 88.2 kHz or 48 Khz).

So, that's a 50% rip-off!

Claims both published and anecdotal are regularly made for audibly superior sound quality for two-channel audio encoded with longer word lengths and/or at higher sampling rates than the 16-bit/44.1-kHz CD standard. The authors report on a series of double-blind tests comparing the analog output of high-resolution players playing high-resolution recordings with the same signal passed through a 16-bit/44.1-kHz “bottleneck.” The tests were conducted for over a year using different systems and a variety of subjects. The systems included expensive professional monitors and one high-end system with electrostatic loudspeakers and expensive components and cables. The subjects included professional recording engineers, students in a university recording program, and dedicated audiophiles. The test results show that the CD-quality A/D/A loop was undetectable at normal-to-loud listening levels, by any of the subjects, on any of the playback systems. The noise of the CD-quality loop was audible only at very elevated levels.

From: http://www.aes.org/e-lib/browse.cfm?elib=14195

Alex

The ODAC doesn’t support 24/88, it does support the audibly identical 24/44. It’s trivial to re-sample 24/88 audio to 24/44 with no artifacts as it’s a simple divide-by-two operation (and one the operating system will perform for you automatically). I know many audiophiles probably think they’re losing something, but nobody has proven they are. Meyer & Moran demonstrated in a very in-depth study that even 16/44 audio sounded identical to SACD. Another good read is 24/192 Music Downloads. You can try resampling some 24/88 audio to 24/44 and compare them yourself with Foobar and the ABX add-on. It’s been done at HydrogenAudio and elsewhere always with the same result: Unless you mess up the resampling somehow, or change the levels, you can’t tell them apart.

But there will always be this debate over normal 44.1 khz 16 bit stuff  vs higher sampling rates and bitdepths...

I have done many ab's here and cant tell the difference...

Alex

I have a possible power supply issue, hoping someone can shed some light on it.

I bought and built a DIY 02 amp, I hadn't done all the tests listed on the blog but it did work ok on battery power.

Battery power ran down and I plugged in the power supply only to find it was dead.

Told the supplier who was very quick to refund the cost of the power supply.

Bought a new power supply which also then failed on me at some point, not sure exactly when.

I don't think either power supply ever charged the batteries.

I now have a 3rd power supply 16vac 1.25A and I have now bought a new amp from JDS, run all the tests suggested on the blog, as a practice run. The new power supply and pre built 02 amp are working fine on battery and power.

I want to run all the tests on the DIY amp, but a little reluctant to lose another power supply as they are expensive here ($25).

Any thoughts on what the problem might be so i can check before powering up the DIY amp for testing.

Quote:

Originally Posted by adydula 

The ODAC doesn’t support 24/88, it does support the audibly identical 24/44. It’s trivial to re-sample 24/88 audio to 24/44 with no artifacts as it’s a simple divide-by-two operation (and one the operating system will perform for you automatically). I know many audiophiles probably think they’re losing something, but nobody has proven they are. Meyer & Moran demonstrated in a very in-depth study that even 16/44 audio sounded identical to SACD. Another good read is 24/192 Music Downloads. You can try resampling some 24/88 audio to 24/44 and compare them yourself with Foobar and the ABX add-on. It’s been done at HydrogenAudio and elsewhere always with the same result: Unless you mess up the resampling somehow, or change the levels, you can’t tell them apart.

 

But there will always be this debate over normal 44.1 khz 16 bit stuff  vs higher sampling rates and bitdepths...

 

I have done many ab's here and cant tell the difference...

 

Alex

I've been able to hear difference only if I purchased hdtracks' version of some favourite album which I previously had only as a cd-rip... I think that the lack of bad cd mastering and overcompression does the trick here.

However, I cannot hear any difference among two different versions of the same hdtracks' album - like 24/96 and 24/192 Green day - American Idiot.

Quote:

Originally Posted by RustA 

Quote:

Originally Posted by adydula 

The ODAC doesn’t support 24/88, it does support the audibly identical 24/44. It’s trivial to re-sample 24/88 audio to 24/44 with no artifacts as it’s a simple divide-by-two operation (and one the operating system will perform for you automatically). I know many audiophiles probably think they’re losing something, but nobody has proven they are. Meyer & Moran demonstrated in a very in-depth study that even 16/44 audio sounded identical to SACD. Another good read is 24/192 Music Downloads. You can try resampling some 24/88 audio to 24/44 and compare them yourself with Foobar and the ABX add-on. It’s been done at HydrogenAudio and elsewhere always with the same result: Unless you mess up the resampling somehow, or change the levels, you can’t tell them apart.

 

But there will always be this debate over normal 44.1 khz 16 bit stuff  vs higher sampling rates and bitdepths...

 

I have done many ab's here and cant tell the difference...

 

Alex

 

I've been able to hear difference only if I purchased hdtracks' version of some favourite album which I previously had only as a cd-rip... I think that the lack of bad cd mastering and overcompression does the trick here.

 

However, I cannot hear any difference among two different versions of the same hdtracks' album - like 24/96 and 24/192 Green day - American Idiot.

I've made a thread about this, and apparently HD tracks go through a different mastering process so by nature they sound different from their CD counterparts. With Foobar I scored Total: 26/27 (0.0%) between an HD Tracks song and the same one ripped from a CD.

I have yet to hear a difference between downsampled music though (from HD to CD quality), so the lack of 88 kHz sampling rate really doesn't matter to me.

Quote:

Originally Posted by adydula 

The ODAC doesn’t support 24/88, it does support the audibly identical 24/44. It’s trivial to re-sample 24/88 audio to 24/44 with no artifacts as it’s a simple divide-by-two operation (and one the operating system will perform for you automatically). I know many audiophiles probably think they’re losing something, but nobody has proven they are. Meyer & Moran demonstrated in a very in-depth study that even 16/44 audio sounded identical to SACD. Another good read is 24/192 Music Downloads. You can try resampling some 24/88 audio to 24/44 and compare them yourself with Foobar and the ABX add-on. It’s been done at HydrogenAudio and elsewhere always with the same result: Unless you mess up the resampling somehow, or change the levels, you can’t tell them apart.

 

But there will always be this debate over normal 44.1 khz 16 bit stuff  vs higher sampling rates and bitdepths...

 

I have done many ab's here and cant tell the difference...

 

Alex

Thanx to Alex and others who discussed the high-res audio tracks here.

I think that it's good to have this mini-discussion here in one place, when we discuss the ODAC & O2.

Well,  the competition, e.g.  the Dragonfly,  can process (without any resampling, etc.) 88.2 kHz files.

But as it was pointed out in this thread (and on this site and on the Internet) this is of no practical importance at all.

(The same applies to 96 Khz/24-bit files which are handled via the USB interface by the ODAC.)

I guess that  96 Khz files could be cleanly downsapled to 48 kHz files,

just as 88.2 kHz  files are cleanly downsapled to 44.1 kHz files.

I think that in both cases, every second sample is processed,

or alternatively, every second sample is skipped.

(I don't think that there is any division-by-two done  here.)

Finally, how about the lowest hig-resolution handled by the ODAC, namely 48 kHz/24-bit.

(I guess, that the lowest high-res is 48 kHz/20-bit, or perhaps even 48 kHz/16-bit, which is rarely, if at all used.)

Has anyone compared  48 kHz/24-bit vs. 44.1 kHz/16-bit? Have you heard any audible difference?

By the way, the difference in the sampling rates is relatively little, namely 48 kHz - 44.1 kHz = 3.9 kHz.

But the difference in bit depths is huge: 24 bit - 16 bit = 8 bit. That's 50% over the 16-bit depth.

Any comments?

Quote:

Originally Posted by dfferent 

I have a possible power supply issue, hoping someone can shed some light on it.

 

I bought and built a DIY 02 amp, I hadn't done all the tests listed on the blog but it did work ok on battery power.

Battery power ran down and I plugged in the power supply only to find it was dead.

Told the supplier who was very quick to refund the cost of the power supply.

 

Bought a new power supply which also then failed on me at some point, not sure exactly when.

I don't think either power supply ever charged the batteries.

 

I now have a 3rd power supply 16vac 1.25A and I have now bought a new amp from JDS, run all the tests suggested on the blog, as a practice run. The new power supply and pre built 02 amp are working fine on battery and power.

 

I want to run all the tests on the DIY amp, but a little reluctant to lose another power supply as they are expensive here ($25).

 

Any thoughts on what the problem might be so i can check before powering up the DIY amp for testing.

 

I'm not sure, sounds weird to me.

 

What were the specs of the first two adapters?  What's your wall voltage?  There were some comments about charging batteries while using the O2 with an underspec adapter, but I don't think that would actually kill the transformer.

 

Check the schematic (click me).  There really isn't much between the battery and the AC power input, so if everything is okay after the battery... what could it be?  I mean, if the regulators somehow failed as short circuits between input and ground, that could be bad, but they're pretty robust.  I don't know if you could really test that without taking them out and using a different test circuit, but that's really not too likely anyway.

 

Maybe ask at diyaudio or ask the diy subforum if you can't get a good response here.

 

Quote:

Originally Posted by JakeJack_2008 

Has anyone compared  48 kHz/24-bit vs. 44.1 kHz/16-bit? Have you heard any audible difference?

By the way, the difference in the sampling rates is relatively little, namely 48 kHz - 44.1 kHz = 3.9 kHz.

But the difference in bit depths is huge: 24 bit - 16 bit = 8 bit. That's 50% over the 16-bit depth.

 

Any comments?

50% more bit depth is kind of a weird way of putting it, because properly you could say it's 12800% more resolution (smallest possible difference between distinct samples... though I could be butchering the terminology as I'm not a DSP guy), though less in practice if you look at effective number of bits of actual hardware.  50% more bits per sample.

Sound science subforum here ran a recent blind test that included 24/96 vs. a file that was converted  24/96 -> 16/44.1 -> 24/96, and people couldn't reliably (and blinded) distinguish the two.  I've tried these kinds of things and not heard the differences, but some people are better at catching small things than I am.

By the way, some technical minutiae:

Quote:

Originally Posted by JakeJack_2008 

I guess that  96 Khz files could be cleanly downsapled to 48 kHz files,

just as 88.2 kHz  files are cleanly downsapled to 44.1 kHz files.

I think that in both cases, every second sample is processed,

or alternatively, every second sample is skipped.

(I don't think that there is any division-by-two done  here.)

That is not correct.  To do sample rate conversion down by a factor of two (properly), you need to lowpass filter the original.  Then you throw away every other sample (technically this throwing away every other sample is known as downsampling; abuse of terminology is rampant, and I'm an offender too).

If you don't lowpass filter first, then you can end up content in the original 88.2 kHz file above 22.05 kHz getting aliased down into the 0-22.05 kHz range of the 44.1 kHz file, because the 44.1 kHz file cannot properly represent info above the Nyquist 1/2 frequency of 22.05 kHz.

I guess you can do sample rate conversion in many different ways, including skipping the filtering (or say, doing a linear interpolation or whatever else), but these are not by-the-books correct and may introduce aliasing or whatnot.

dfferent...

ok so your diy amp you built has issues. the one you bought assembled works ok.

I would take a very slow and careful look at your DIY amp, check all the  parts you placed on the board for correct insertion and value.

Especially the diodes and capacitors.

Check all the solder joints, top and bottom.

I would take the board out of the case and see if it works ok outside of the case to make sure there are no shorts to the case.

Also check to make sure all the correct value components are in the correct place on the board.

Also on the designers website there is a troubleshooting section if you havent done this  run thru these checks, if you have do it again.

The power supply is not the thing you plug into the wall thats a AC to AC adpater that provides AC power to the half wave rectifier in the O2.

The amp should not be able to "hurt" the AC to AC adapter which is nothing more than a step down transformer...

Good Luck

Alex

Quote:

Originally Posted by mikeaj 

  O2 gets 353 mW @ 15 ohms, 534 mW @ 33 ohms, (interpolated by me) 272 mW or so @ 100 ohms, 94 mW or so @ 300 ohms, on a mid-high charge on battery

 

So the O2 is better for some planar magnetics, maybe AKG Q701 if you listen really loud.  Otherwise, no lower-impedance sets actually need that kind of power.  O2 is also a little louder for high-impedance sets.

Sorry to bring this up again, I'm trying to scale the O2 against some other amps since I might get an HE-500 in the distant future. You mention the O2 will work with some planar magnetic headphones. The LCD-2 is reported to sound quite good with the O2, and its 0.66 mW to reach 90 db SPL seems logical. It looks that an O2 might even work for an HE-6 (19.69 mW to reach 90 db SPL), maybe?

I often read that the Schiit Audio Lyr is a good "pair" with the HE-500 and according to their website, the output voltage is:

What is P-P? If it's the Vrms value, then by the previous discussion's calculations, it outputs a whopping 50 W.

40V P-P means peak to peak.  It would be 14 Vrms, and I believe the Lyr is rated at 6W into 32Ω

Edit: I'm an idiot and really have no idea why I wrote 20 Vrms. Im going to blame a long day at work.

Quote:

Originally Posted by shadow419 

40V P-P means peak to peak.  It would be 20Vrms, and I believe the Lyr is rated at 6W into 32Ω

Oh I see.

P = (Vrms^2)/R

P = 12.5 W from the calculation. Hm...

Vrms ~ (1/3) V P-P = ~ 13.33 V

P = 5.56 W, there we go

MalVeauX seems to like the Lyr with the HE-500's too. I keep reading that the minimum power for the HE-500 is 1 W, in which the O2 can't deliver enough power.

Quote:

Originally Posted by MrEleventy 

Quote:

Originally Posted by miceblue 

Hm I see. Thanks for the clarification.

 

I still don't really like that that happens though. With my Retina MacBook Pro on max volume, the O2 set to as low of volume as possible at 1.0x gain, they are super loud for me with the K 701....which basically renders the O2 useless for the M-100 if the O2's potentiometer is at the 7:30 position (yes there's channel imbalance) for the K 701 even at 1.0x gain. With this kind of listening, I can't physically lower the volume with the amp any more or else I will get a really noticeable channel imbalance. All of this seems unreasonable to me.

 

The creator of the O2 did point out "guilty free" source volume control with the ODAC's 24-bit USB interface and gave a brief sentence or two about the loss of resolution and whatnot. The question is, can you guys actually hear it? I notice no significant increase in sound quality, if any, with the source on max volume, super-limited volume control with the amp versus low volume on source, very diverse volume control with the amp.

 

My current setup is E17 LO > O2 > M80s, looking at the specs, it's not too far off from the M-100s (28.5 ohms vs 32 ohms, 105db vs 103db), what I do, since all 99% of my files are 16 bit anyways, I set my E17 to run at 24bit and lower the volume by about 25% and that gives me more wiggle room on the O2. I don't know the exact math but doing my own fuzzy math, I figure since 16 to 24 is about 66.7%, I can throw away 33% before losing "quality". I actually don't notice a difference at 50% but my OCD goes off and I crank it back up. 

 

P.S. My O2 is still on the defaulted 2.5x/6.5x, FWIW.

I just did some subjective tests using Daft Punk's "Voyager" song (16/44.1 EAC CD-ripped) with the M-100's.

1.0x gain:

• source volume ~ 50% of max
• O2 potentiometer ~ 8:45 position
• volume is louder than I would normally listen to music at
• no background hiss

2.5x gain:

• O2 potentiometer left at ~ 8:45 position
• source volume left at ~ 50% of max
• no background hiss (strange, I could have sworn I heard the background hiss earlier.....stopping the music and turning the potentiometer, I can start hearing the background hiss at around the 12:15 position [I did this after doing finishing the testing session])
• volume is much louder than I would normally listen to music at, so I adjusted the source volume until the volume at 2.5x gain is about the same as 1.0x gain (rough estimate): now source is ~31.25% of max (yeah yeah this isn't objective having the by-ear volume guessing)
• I don't notice any degradation of the signal with the lower source volume

So with the 2.5x gain at a lower source volume, I might be losing bits, but I can't hear it, at least not with the M-100. Having the potentiometer at the 8:45 position is pretty much the bare minimal "angle" before the channel imbalance kicks in for me. So even with the ~50% source volume at 1.0x gain, I have practically no wiggle room for volume adjustment using the O2.

For a sine wave, it's a factor of 2 * sqrt(2) = about 2.83 between peak-to-peak and rms voltages.  The 2 comes from there being a positive and negative component, and the sqrt(2) comes from the shape of the sine wave.  

Power required for any headphones depends on how loud you want to listen.  Power delivered corresponds to the volume setting.  Most people aren't using anywhere close to the maximum capabilities of their amps.  Some amps will distort badly as you try to crank up the power, with lower-impedance headphones; that depends on the design.  1W into HE-500 would be really really loud:  about 117 dB SPL according to InnerFidelity.  1W is just some kind of nice round number people can quote.  It's also a way for HiFiMAN to set expectations and pass the buck.

At an equivalent setting, LCD-2 should only be a few dB louder than HE-500, so it's not in a totally different class.  

Plugged into the wall, extrapolating between known values, O2 on AC power can deliver probably something above 700 mW into 50 ohms.  Anyhow, even being really conservative, you're losing out on less than 2 dB compared to something that can output 1W.  You know how much different 2 dB sounds, right?

Quote:

Originally Posted by mikeaj 

For a sine wave, it's a factor of 2 * sqrt(2) = about 2.83 between peak-to-peak and rms voltages.  The 2 comes from there being a positive and negative component, and the sqrt(2) comes from the shape of the sine wave.  

 

Power required for any headphones depends on how loud you want to listen.  Power delivered corresponds to the volume setting.  Most people aren't using anywhere close to the maximum capabilities of their amps.  Some amps will distort badly as you try to crank up the power, with lower-impedance headphones; that depends on the design.  1W into HE-500 would be really really loud:  about 117 dB SPL according to InnerFidelity.  1W is just some kind of nice round number people can quote.  It's also a way for HiFiMAN to set expectations and pass the buck.

 

At an equivalent setting, LCD-2 should only be a few dB louder than HE-500, so it's not in a totally different class.  

 

Plugged into the wall, extrapolating between known values, O2 on AC power can deliver probably something above 700 mW into 50 ohms.  Anyhow, even being really conservative, you're losing out on less than 2 dB compared to something that can output 1W.  You know how much different 2 dB sounds, right?

I don't know too much about how loud something sounds in correspondence to the decibel system. Isn't a +3 dB difference doubling the perceived volume?