READ THIS: Serious flaws in ipod classic

[Hifivoice]

Hi Jook,

Quote:

Originally Posted by jook /img/forum/go_quote.gif Yes. There is no mention on your page (maybe I missed it?) of the very evident bass attenuation problem in previous generations (most significantly in the 3rd gen and 4th gen, but also evident - although less so, in 5th gen measurements).

True, it's not on this page as I already published about that one a long time ago:
http://www.hifivoice.com/audio/ipod/...gineering.html

There you can see differences in capacitor size in the output circuitry.

Marc

 

[Buckster]

well sorry to say ... I bought a 5.5G IPOD today to replace my 6G one ...

I have both at the moment -before the 6G one goes back to Apple (and I pay restocking fee) ..

Gave the 6G the best I could - but some tracks which I love and have listened to many many different players - I just couldn't listen to .. female voices sound amazing about 85% of the time, but sometimes sounds really "spitty" and "tizzy" with spoken s's etc .. like if you partnered say a NAIM CD player, with a bright amp, and bright speakers ...

did A to B comparisons with my Sony 705F .. this is same track same place in track, and switching headphones between the two ... and the Sony sounded as good but without the "tizzy" and "spitted out s's" - the Sony only cost 49 pounds ..

got a colleague to listen too and he thought the same, tried with 3 different pairs of phones too.

now I will report back if I've made a mistake (can send the 6G one back - but 5.G is non returnable) - and sound quality is worse on the 5.5G

every other function of the 6G I love, the GUI, the battery life, the design and size- are all superb.

 

[Hales]

Buckster - I am hearing the same things - certain tracks that are ripped in lossless and sound fine on cd players, my wife's 4g color, and my computer will sound a little metallic with sibilance and a "not right" sound in the treble.

I have been trying not to post anything else on the boards about it any more. People keep saying that x.xx db of difference is inaudible, the classic measures perfect, blah blah blah.

The reality is that some people got units that sound bright to them. Is it a bad headphone / classic combo? Is it a issue with with the dac or codec? Is is a pissing contest where people don't know the difference between db and dbfs? (hint - all of the measurements so far are dbfs) (hint 2 - unlike db which can be calibrated, dbfs is just how many db's an audio signal is from clipping digital dac)

 

[Hifivoice]

Steve,

A few posts backwards you claim that I have no clue about acoustics, you make all kind of assertions about the validity of my measurements and observations, and now you are explaining us that you're just about to understand where phase response and group delay are about...

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif I'm just starting to delve into this, but from what I've found out so far, there are two types of phase issues... group delay (time is constant with frequency) and phase shift (time varies with frequency). Group delay is OK, because it all comes out equal in the end. Phase shift is not, because frequencies are being delivered in different times.

Group delay is the (negative) derivate of the phase response.

If you have the same group delay for all frequencies, it essentially means the sound is just delayed. A group delay of 0 means no delay, reflected by a flat phase curve. A positive group delay of a few 0.1ms is reflected by a tilted phase curve.

If the group delay is not constant for all frequencies, the phase curve is not a straight line anymore, but a curved line. For the iPod 6G measurments, one can see that the phase of the treble is about flat (so group delay about 0, so no delay), whereas the bass is delayed.

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif If that's the case, the 6G actually has marginally BETTER phase response than the 5.5G.

As explained on the website, the measurement grid is not sufficient to subtract small group delays, and to make the 6G curve flat. Furthermore, there is a comparison with the minimum phase response, which purpose is to remove any group delay in front of the measurement, which is calculated from the frequency response. There is some math beyond that why this is possible, but with one very important restriction. Namely, that the system is a minimum phase system. Obviously, measurement show this is not the case, and hence I cannot use this facility.

The 5G shows a consistency of phase over more octaves than the 6G. Which one is "better" is not the discussion, treble of the 6G is ahead of bass and mid, and the curve deviates more from a linear reponse than the5G. Those are the only conclusions that can be drawn from the curves, and is exactly what I do.

 

[mirumu]

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif Mirimu, I would bet that the problems you are hearing with the piano notes have more to do with the headphone output than anything else. There is definitely a difference between the 4G and 5.5G headphone out. From line out, I bet the differences are much less pronounced, if you can even hear any at all.

I can certainly do line out tests to verify but I would point out that both the 4G and CD player yielded the same result in the area I was focusing on despite being very different sounding devices.

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif Lossy files probably wouldn't even have the frequencies we are talking about. The supposed shift takes place above 15kHz. I'm pretty sure that most lossy formats would filter that off.

No, that's a different issue, there's the 0.1-0.2db treble boost in that 10Khz+ region. The phase problems would seem to exist across the frequency range in varying degrees.

 

[mirumu]

Quote:

Originally Posted by Hales /img/forum/go_quote.gif Buckster - I am hearing the same things - certain tracks that are ripped in lossless and sound fine on cd players, my wife's 4g color, and my computer will sound a little metallic with sibilance and a "not right" sound in the treble. I have been trying not to post anything else on the boards about it any more. People keep saying that x.xx db of difference is inaudible, the classic measures perfect, blah blah blah. The reality is that some people got units that sound bright to them. Is it a bad headphone / classic combo? Is it a issue with with the dac or codec? Is is a pissing contest where people don't know the difference between db and dbfs? (hint - all of the measurements so far are dbfs) (hint 2 - unlike db which can be calibrated, dbfs is just how many db's an audio signal is from clipping digital dac)

I don't think anyone is disputing that it sounds brighter to certain people, it's too widely reported for that. The question is why and what is the reference player to which the 6G is comparatively bright and is that player itself providing a flat response. All I've been arguing is that if this reference player is in itself some way dark then perhaps the 6G is not necessarily bright, just brighter. Real music can be both harsh and fatiguing so these aspects are not necessarily a good indicator of a player having boosted treble. I do take your point about dbfs.

 

[Hales]

My Reference Player was a Marantz SA-8260 CD/SACD player. It was Marantz's 1kilobuck CD player that got a Class A rating in Stereophile's recommended components list.

Compared to the Marantz, the 4g color had almost exactly the same dbfs freq response - except for the bass rolloff.

Compared to the Marantz, the 6g had a steady rise in treble energy starting around 3.5khz and rising to peak around 19khz.

Next steps would be to measure the player under load. My Shure e3c's have a rising ohms value as the treble goes up - I need to see if under load the slight excess of treble is amplified.

 

[Buckster]

......

well I prefer the sound of the 5.5G on quick listening ...

but what has REALLY surprised me is the video quality difference - on my particular samples ... the 5.5G video is a LOT better than the 6G one .. same clip ... much better white balance - sharper, and more detailed ! I was expecting the other way ... Also the 6G one you can in some things with a lot of grey- see the 3 LED backlights at the bottom - you can't on the 5.5G ...

I may post a photo - the difference is quite stark !

Mark.

 

[Night Surfer]

Quote:

Originally Posted by Buckster /img/forum/go_quote.gif ...... well I prefer the sound of the 5.5G on quick listening ... but what has REALLY surprised me is the video quality difference - on my particular samples ... the 5.5G video is a LOT better than the 6G one .. same clip ... much better white balance - sharper, and more detailed ! I was expecting the other way ... Also the 6G one you can in some things with a lot of grey- see the 3 LED backlights at the bottom - you can't on the 5.5G ... I may post a photo - the difference is quite stark ! Mark.

Maybe your screen isn't entirely burnt in

 

[japc]

Quote:

Originally Posted by Night Surfer /img/forum/go_quote.gif Maybe your screen isn't entirely burnt in

Ahahahaa. Nice one

 

[mirumu]

Quote:

Originally Posted by Hales /img/forum/go_quote.gif Compared to the Marantz, the 4g color had almost exactly the same dbfs freq response - except for the bass rolloff.

I measured the 4G iPod Photo headphone output last night and the treble was significantly rolled off on the dbfs frequency response. I'm at work currently so can't see or post the graph now but the roll off was into double digits at 20Khz. Performance on the line out could well be another matter of course.

I'd certainly expect a player at the level of the Marantz to be very good in this regard though and I'd hope it would look a lot better than what I was seeing from the 4G iPod at both ends of the spectrum.

Quote:

Originally Posted by Hales /img/forum/go_quote.gif Compared to the Marantz, the 6g had a steady rise in treble energy starting around 3.5khz and rising to peak around 19khz.

How were you determining this?

 

[Buckster]

means the 5.5G will get better than as the 6G display is already past its burn in period ....

my wife tolerates a lot when it comes to hifi etc - as I have a tendancy of swapping and changing, and 99% of the time she can't tell a blind bit of difference and I've just forked out lots of money !

but the difference in video between the 2 even she could spot ! completely different - the problem with trying to get a photo is that when you pause quite often you get a bit of motion blur ...

how can reviewers who must have both types at hand not spot these types of difference ? or is it variation in samples ??

sound quality I MUCH prefer on the 5.5G - not to say its technically better or worse - I don't really care about charts - but the 5.5G sounds just as bright but doesn't have the artefacts than the 6G does - again perhaps I have a duff one - and YMMV as always !

from a video point of view the older version seems slower than the one one - particularly when skipping chapters etc ...

also cover-art when jumping from track to track comes up slower on the 5.5G vs the 6G .. but I can live with that.

Mark.

 

[markopolo]

Well boys, I'm done with this. I've put in all kinds of time comparing the 5G/6G using several phones and amps, both headphone and line out, and no amount of squabbling over graphs or measurements or what different audio terms mean is going to change what I hear. I think Apple really screwed the pooch here. This thing's going back and I'm ordering a 5.5G from a 3rd party (I killed the battery on my 5G). The $35 restocking fee bites, though. Anyone know a way around that?

 

[RickB]

Quote:

Originally Posted by Buckster /img/forum/go_quote.gif ...... but what has REALLY surprised me is the video quality difference - on my particular samples ... the 5.5G video is a LOT better than the 6G one .. same clip ... much better white balance - sharper, and more detailed ! I was expecting the other way ... Also the 6G one you can in some things with a lot of grey- see the 3 LED backlights at the bottom - you can't on the 5.5G ... I may post a photo - the difference is quite stark ! Mark.

Ha. On my Classic the LED's are clearly visible on the Now Playing screen. I was thinking maybe it was a defect of my particular unit, but looking at photographs on Flickr it's apparent they're all like that.

Rick

 

[Hifivoice]

Quote:

Originally Posted by Hales /img/forum/go_quote.gif people don't know the difference between db and dbfs? (hint - all of the measurements so far are dbfs) (hint 2 - unlike db which can be calibrated, dbfs is just how many db's an audio signal is from clipping digital dac)

dBfs is a measurement where the output is compared to a maximum level. My measurements haven't been calibrated with the max level being 0dB, because that is irrelevant for the properties that need to be measured. I measure relative response, with 1kHz taken as a reference value. From that perspective, dB is a perfectly fine expression for measurement, as it defines the relative ration between two signals. Hence, the measurements can be perfectly (and should be) expressed in dB, and should not in dBfs, because that would be faulty.

Audio measurements performed by computer equipment always imply a AD-convertor inbetween. Unless the measurement is pushed beyond the clipping level of the audio equipment, or below the threshold of the measurement equipment, the term dBfs has not really a usefull meaning for those measurements.