[bLue_oNioN]

This thread is a continuation of the thread found here, which was unfortunately locked.

The purpose of this thread is not to debate anything resembling the absolute statement "iPod Shuffle reigns King," nor is it to discuss the author's sample pool of DAPs. The author had limited access to test models; hopefully we can all accept that fact and leave it be.

The link to the PC Magazine article in question can be found here and the link to afforementioned quantifiable data/graphs can be here. The graphs yield some interesting results, some expected, and some unexpected.

If you scroll down on the last page of the previously locked thread here you will find that breez actually ran tests of his own to supplement the originals.

I personally found interesting the author's comment on the iPod Shuffle:
"The reason for this sterling performance is that the left and right channels each have two transistors, one pushing, one pulling, and no capacitor that gets discharged over time."

Could someone perhaps elaborate on that point?

 

[breez]

Quote:

Originally Posted by bLue_oNioN I personally found interesting the author's comment on the iPod Shuffle: "The reason for this sterling performance is that the left and right channels each have two transistors, one pushing, one pulling, and no capacitor that gets discharged over time." Could someone perhaps elaborate on that point?

I have no idea about the part about transistors, but maybe the DAC/headphone driver chip in Shuffle is a design that allows the usage without output capacitors? The DAC in iRiver H120 works like that and it would make a lot sense for Shuffle too, because capacitors of sufficent capacitance take space.

 

[devwild]

I would like to add a simple note to this conversation, I'm not really getting in on the debate, but I did find the topic interesting. Out of curiosity I sent an e-mail to the author, and he mentioned that he will indeed be adding more players to his comparisons as he gets his hands on them.

Personally, I look forward to seeing the results.

 

[Daroid]

Quote:

Originally Posted by bLue_oNioN ...I personally found interesting the author's comment on the iPod Shuffle: "The reason for this sterling performance is that the left and right channels each have two transistors, one pushing, one pulling, and no capacitor that gets discharged over time." Could someone perhaps elaborate on that point?

For a push-pull configuration for a single channel you will need one NPN and one PNP transistor.

This is what the basic circuit looks like: http://www.ecircuitcenter.com/Circui...l/image002.gif

The circuit shows one NPN, one PNP transistors which are connected by their base to ground. When you connect an audio signal to Vpos and Vneg, the Vpos will only emit the positive swings and the Vneg NPN transistor will only emit the negative swings. RL1 could be a speaker for example. Vpos pushes, Vneg pulls.

EDIT: About the last part about the capacitor. If you add one you'll need to charge and discharge it over and over leaving a square wave a pretty obscure figure. That will explain too why capacitors "blocks" DC - it's square waves.

 

[baijxz]

As far as I can understand the whole issue is about the amps of the players and not the DAC's.

The reasoning beheind the testing seems sound, and it's the only real testing I've seen of any mp3 player.

It seems to me that the author is speculating whether or not the amp in the shuffle is singleended or not.
But if it is singleended then wouldn't it have to be class A and therefore pretty power hungry?
I'm not proficient enough in these matters to come up with a decent theory but maybe somebody else could give us the lowdown on portable amplifiers.

 

[NeoteriX]

Inspired by breez's results, I decided to do some of my own testing -- thanks to breez with confirming w/ me the testing procedure

I wanted to see if I could duplicate Bill Machrone's results with the huge dip in the square wave signal and to see if the results match my subjective opinion of my Pocket PC, that is has significantly better bass response. I've thrown in my roommate's Archos as well. Comments welcome~

Notes: The DAPs were tested with no equalizer boost, through their headphone jacks at a relatively normal listening volume. I'm using regular (not audiophile) quality interconnects and line splitters from Radio Shack. The sound file used was generated using Audacity, a 30 second long 40 Hz square wave, just like Breez and Bill Machrone.

Computer: Dell Dimension 8300 w/ a Turtle Beach Santa Cruz soundcard, line in (Not as nice as Bill Machrone's external USB unit, but it'll do)

DAPs tested:
- 4G 20GB iPod
- Archos Jukebox Recorder 10
- HP iPaq 2210 PocketPC

Headphones used (w/ impedance in Ohms):
- iPod earbuds (32)
- Koss HP-CN5 Noise Cancelling headphones*** (32)
- Sony MDR-G72 Street Style (32)
- Sony MDR-V6 (63)
- Shure e3c (26)

Conclusions: I have not directly compared my roommate's Archos subjectively, but comparing between my iPod and iPaq, the iPaq definitely has punchier bass, and it seems the results confirm that observation. Regardless of what headphone was plugged in (headphones or with no load) the wave did not significantly change for the Pocket PC. With the iPod, the signal dropped like a rock with any load. The Archos faired a little bit better.

One very interesting phenomenon is the Koss Noise Cancelling headphones. They are of the active noise cancelling variety, and they have a single AA battery to power the noise cancelling circuit. Interestingly enough, when the headphones were unpowered, with the Archos, there was no significant change in the signal integrity, and with the iPod, the signal did not drop nearly as much. When the noise cancelling circuit was turned on, the same signal drop phenomina was noticed (Archos/iPod)

Dariod, why would a engineer choose one audio design solution over the other? It seems that if the transitor design was inherently better in terms of sound quality, everyone would adopt it -- what are the downsides?


Results:
All screenshots and extra results can be found here:
http://www.pbase.com/neoterix/dap_testing
Some samples:

- 4G iPod with iPod earbuds:

- HP iPaq 2210 Pocket PC with iPod earbuds

- Archos Jukebox Recorder 10 with iPod earbuds

 

[st5150]

Quote:

Originally Posted by bLue_oNioN I personally found interesting the author's comment on the iPod Shuffle: "The reason for this sterling performance is that the left and right channels each have two transistors, one pushing, one pulling, and no capacitor that gets discharged over time." Could someone perhaps elaborate on that point?

I feel that I'm qualified to talk to this. Here is my dumbed down explanation of this butchery:


Sounds like the output stage of the shuffle is a class B push-pull amplifier. This isn't a big deal, its nothing magical. Its been around for ages in the electronics world. Infact, its nothing to brag about. Surely using a class B output stage does not imply superiority! Infact, the first thing I think of when I hear class-b push-pull is cross over distortion.

As for the "capacitor that gets discharged over time." comment.... this proves the author has absolutely ZERO electronics experience or knowledge and is simply talking out of his @ss for lack of a better term. I don't even know where to start with this comment so I won't even address it.



In short, that statement is a bunch of hypo-mumbo-jumbo psuedo-technical BS. The whole artical is an infomercial for ipod if you're an engineer.

 

[NeoteriX]

Quote:

Originally Posted by st5150 Sounds like the output stage of the shuffle is a class B push-pull amplifier. This isn't a big deal, its nothing magical. Its been around for ages in the electronics world. Infact, its nothing to brag about. Surely using a class B output stage does not imply superiority! Infact, the first thing I think of when I hear class-b push-pull is cross over distortion.

I see -- we can test for that though right? How can we go about testing the Shuffle to see if its implementation does suffer from cross-over distortion?

Quote:

Originally Posted by st5150 As for the "capacitor that gets discharged over time." comment.... this proves the author has absolutely ZERO electronics experience or knowledge and is simply talking out of his @ss for lack of a better term. I don't even know where to start with this comment so I won't even address it.

For the sake of us who don't have a technical background in electrical engineering, please do explain it. It sort of made sense to me, but admittedly, I am a layman. My only experience with the understanding of how amps work is from reading through the design notes for cMoy and other small amps (I hope to build one soon

) From reading the design notes for the Tangent cMoy headphone amp, I found this quote that I thought might be relevant:

Quote:

From "Using Different Caps" The caps' values will change the way the amp performs. For the power caps 220 µF is adequate, but bigger ones will provide a bigger current reserve, which can be useful in handling high instantaneous loads, like big drum hits. I've tried 470 µF caps in my amps before, and it does improve the bass handling significantly and they're not all that much bigger than the 220s I was using previously. [size=xx-small]Source: http://tangentsoft.net/audio/cmoy-tu...eaks.html#caps[/size]

This is where I got my impression that a capacitor acts as a "reserve" tank -- if the signal consumes more power than the capacitor's reserves can handle, it seems like it would cause the signal to drop off, precisely what is happening with the square wave. It seems like the 40Hz square wave tested here is a worst case scenario "high instantaneous load", and the quote does seem related with this study in that we are referring to bass performance. Any insight you can provide in this would be very appreciated.

Quote:

Originally Posted by st5150 In short, that statement is a bunch of hypo-mumbo-jumbo psuedo-technical BS. The whole artical is an infomercial for ipod if you're an engineer.

As a last note, please, please, please don't let this degrade into an iPod vs. everything else debate again. I'm not as interested in promoting one DAP over the other as much as I'm trying to get to the bottom of the sound quality contraversy and to also understand how the DAPs work. Also, it seems like this isn't just a pro iPod article -- by the author's analysis, the 3G and Mini iPods have the worst bass handling/signal integrity of all the DAPs tested, and it seems that way too between my 4G, Archos, and PocketPC test as well. Cheers,

 

[st5150]

"I see -- we can test for that though right? How can we go about testing the Shuffle to see if its implementation does suffer from cross-over distortion?"

With out going too much into the details, picture a sine wave. Class B uses one transistor to amplify the wave above the X axis, and another for the wave below the X axis. Due to the turn on voltage of a PN junction, there will be a "flat spot" when ever the wave crosses the axis. There are tricks around this, and I'm sure its not an issue or even worth measuring with the ipod or most modern amplifiers.
I made that comment to point out that class-b doesn't pop " superiority" into the minds of those-in-the-know, as it is portrayed in the infomercial by PC Magazine.


Regarding capacitors, imagine them as tiny tiny rechargable batteries. They can VERY quickly be charged, or discharged. They are built for this very purpose. Just like rechargable batteries, if you leave them alone, they will discharge bythemselves. With with that in mind, the authors " and no capacitor that gets discharged over time." comment is rediculous. It implies that capacitors discharging over time is "bad" or not desired. This is just silly. If you had a capacitor that never discharged over time, well hell you'd have an infinite energy source..... sell me the plans so I can take over the world please

The link you gave/quoted is just the observance of the filter effect capacitors have when put on or across a signal source. A capactitor in series with a signal acts like a high pass filter. In laymans terms, the smaller the capacitor, the less bass gets through. This is why the person you quoted observed better bass response with a larger capacitor. There are trade offs I won't get into though.


In closing, I wasn't s degrading this thread into an iPod vs. everything else debate again. You're right, that story is not a pro ipod artical, its an Apple Ipod infomercial that insults 50 years of electrical engineering advancement and a hundreds of years worth of deductive scientific testing proceedures. It just bothered me how hard the author was trying to confuse people and pass off witch-craft as science.


I'd like to end this post by stating that I raced my honda against a nissan and a toyota and scientifically concluded [and now we take a break from this post to tell you that our sponsors tell us civics are now having a $2000 factory rebate sale] that hondas are the fastest cars on the road because of their great fuel economy and resale value

 

[guyverl]

Sorry, just want to say I think that's one of the best posts I've read on this forum

Quote:

Originally Posted by st5150 "I see -- we can test for that though right? How can we go about testing the Shuffle to see if its implementation does suffer from cross-over distortion?" With out going too much into the details, picture a sine wave. Class B uses one transistor to amplify the wave above the X axis, and another for the wave below the X axis. Due to the turn on voltage of a PN junction, there will be a "flat spot" when ever the wave crosses the axis. There are tricks around this, and I'm sure its not an issue or even worth measuring with the ipod or most modern amplifiers. I made that comment to point out that class-b doesn't pop " superiority" into the minds of those-in-the-know, as it is portrayed in the infomercial by PC Magazine. Regarding capacitors, imagine them as tiny tiny rechargable batteries. They can VERY quickly be charged, or discharged. They are built for this very purpose. Just like rechargable batteries, if you leave them alone, they will discharge bythemselves. With with that in mind, the authors " and no capacitor that gets discharged over time." comment is rediculous. It implies that capacitors discharging over time is "bad" or not desired. This is just silly. If you had a capacitor that never discharged over time, well hell you'd have an infinite energy source..... sell me the plans so I can take over the world please The link you gave/quoted is just the observance of the filter effect capacitors have when put on or across a signal source. A capactitor in series with a signal acts like a high pass filter. In laymans terms, the smaller the capacitor, the less bass gets through. This is why the person you quoted observed better bass response with a larger capacitor. There are trade offs I won't get into though. In closing, I wasn't s degrading this thread into an iPod vs. everything else debate again. You're right, that story is not a pro ipod artical, its an Apple Ipod infomercial that insults 50 years of electrical engineering advancement and a hundreds of years worth of deductive scientific testing proceedures. It just bothered me how hard the author was trying to confuse people and pass off witch-craft as science. I'd like to end this post by stating that I raced my honda against a nissan and a toyota and scientifically concluded [and now we take a break from this post to tell you that our sponsors tell us civics are now having a $2000 factory rebate sale] that hondas are the fastest cars on the road because of their great fuel economy and resale value

 

[BigFil]

An iPod informercial from PC Mag where Apple refused to comment. What's the world coming to? BTW good electrical explanation in your post but for the sake of the thread you could have left the last two paragraphs out. Remember the whole reason the the previous thread got closed.

 

[IpaqMan]

Quote:

Originally Posted by NeoteriX Inspired by breez's results, I decided to do some of my own testing -- thanks to breez with confirming w/ me the testing procedure I wanted to see if I could duplicate Bill Machrone's results with the huge dip in the square wave signal and to see if the results match my subjective opinion of my Pocket PC, that is has significantly better bass response. I've thrown in my roommate's Archos as well. Comments welcome~ Conclusions: I have not directly compared my roommate's Archos subjectively, but comparing between my iPod and iPaq, the iPaq definitely has punchier bass, and it seems the results confirm that observation. Regardless of what headphone was plugged in

Thanks for the informal testing. I always thought that my Ipaq 2210 had punchy powerful audio.

 

[PeterR]

Quote:

The speed with which the square wave sinks back toward zero indicates how long the player can sustain bass notes and their harmonics under load.

Uhm, no, it simply shows the low frequency extension.

Quote:

All but one of these players use single-ended

Certainly they don't
Quote:

, capacitively coupled output stages.

That's right.

Quote:

but the size of the coupling capacitor and the impedance of the headphones have a significant effect on the player's ability to sustain a complex bass tone.

It will affect the frequency response. If you filter out the lower frequency components of a square wave they will look like what he's measured. No magic involved, and 'sustaining' a tone has nothing to do with it whatsoever.

Quote:

Sounds like the output stage of the shuffle is a class B push-pull amplifier

Of course it's push-pull, like every other of those amps. Push pull doesn't necessarily mean class B, BTW.

Quote:

Due to the turn on voltage of a PN junction, there will be a "flat spot" when ever the wave crosses the axis.

No, there won't if biased correctly.

Quote:

The link you gave/quoted is just the observance of the filter effect capacitors have when put on or across a signal source. A capactitor in series with a signal acts like a high pass filter. In laymans terms, the smaller the capacitor, the less bass gets through. This is why the person you quoted observed better bass response with a larger capacitor. There are trade offs I won't get into though.

Actually that quote is talking about reservoir caps in the power supply and has nothing to do with this.

That PC Mag Author may not have the slightest idea what he's talking about, it's still worth noting however that the Shuffle doesn't suffer from smallish output caps like its bigger brothers.

 

[st5150]

Well we now know who has the bigger shlong here

Fun thread

 

[Tempus]

Speaking as someone who's planning the purchase of his first DAP in the near future, I don't suppose some clever person out there with access to one of the Cowon iAudio players, perhaps the 5 or the G3, would be able to post similar loaded/unloaded square wave graphs for it? I'm curious how it compares to the Shuffle's graphs. I'm not at all sure, but I believe that I heard somewhere that they both use the same Sigmatel chip. This may have nothing to do with the graph, as that may be all about the output stage, which may not be on that chip, but I'm still curious about how an iAudio graph would look, particularly under load.