Pitch bug on Clip+?

I think using ABX for differences in pitch is somewhat misleading.  With a reference distinguishing tiny differences in frequency is actually pretty easy.  Thats why people use pitch forks when tuning instruments and certainly manage to tune much better then 1%.  But being able to do it with a pitch fork, digital tuner, etc doesn’t mean you can tune it by ear . . .

What really matters is if you can notice a difference during actual music without a reference tone which is much harder to do since its equivalent to tuning by ear.  I certainly can’t do that.  

Edit:  Not that you shouldn’t ABX, I just mean the results are difficult to compare to normal listening in this particular case because the reference actually changes your perception.   

Message Edited by saratoga on 10-31-2009 01:56 PM

That’s pretty funny considering that a lot of people who will announce that the effects of switching to oxygen-free cable, magic green markers for the edges of CD’s, “breaking in” cables, etc are obvious will start hemming and hawing when asked about ABX or double blind testing and come up with reasons why it’s invalid from that end.  Either that it doesn’t work, or is not worth the trouble when the difference is so clear without it!

I agree in that IMO ABX testing is more sensitive to differences than I care to worry about for most things, but I was responding specifically to a post that most people probably couldn’t ABX the difference in pitch.

BTW, I did run it again, listening to the A (1000 Hz) and B (1003 Hz) and X samples only in the first trial and listening  to ONLY the X sample after that and still ended up with 8/11 right… with practice I think I could do better.

Tuning with a tuning fork is easier as you are hearing both tones at once and can key on the beat frequency to tell if they’re the same.   

" agree in that IMO ABX testing is more sensitive to differences than I care to worry about for most things, but I was responding specifically to a post that most people probably couldn’t ABX the difference in pitch. BTW, I did run it again, listening to the A (1000 Hz) and B (1003 Hz) and X samples only in the first trial and listening to ONLY the X sample after that and still ended up with 8/11 right… with practice I think I could do better."

I was referring to ABX testing using a few different music tracks, not test tones. While some small percentage of people might be able to hear a 0.25% difference in pitch, the vast majority of people won’t be able to hear it.

To my ears, the pitch variation of the original Clip was never a major issue, a fair trade for running the processor in its lowest power consumption mode.  The Clip gets a crazy amount of play time out of that wee LiPo battery.

I would like to see the timing issue resolved for the original device, as it has been effectively done for the new Clip+.  Time will tell on that issue (pardon the shameless pun).  The Clip+ does consume those electrons a bit more than its elder sister, but I do like the improved functionalities of the new toy.

Hey, the podcast order is oldest-to-newest as well, a welcome improvement.

Currently, I’m trying to kill my Clip+, down to 50% on the battery .  I’ve been working quite a few trials on it these past weeks, with various codecs, so it’s been plugged in for a large percentage of its time.

Bob  :smileyvery-happy:

There is one thing I am very curious about: 
I never heard that any other manufacturer had this particular problem with their players (correct me if I am wrong). 

It would be really interesting to know what are the real  technical reasons (in the first place) that such a problem creeps in such a long time and why is Sandisk so secretive about it.
I mean why they just don’t release concrete numbers to support the claims that using correct clock frequencies would have such a major influence on battery life. And if so, what kind of brain-damaged SOC is that (presumably the core is AMS AS3525).

I can’ t really enjoy my Fugue (which consist of a single constant 1kHz Tone) but I must admit that thanks to Sandisk it got more interesting :slight_smile:

Message Edited by m9zf3n5w on 11-06-2009 03:07 PM

@m9zf3n5w wrote:

 

It would be really interesting to know what are the real  technical reasons (in the first place) that such a problem creeps in such a long time and why is Sandisk so secretive about it.

They started with a PLL clock that couldn’t do 44.1kHz and didn’t want to change it once they realized their mistake because its a lot of work.  Not sure what you mean by secretive.  

That what you just said is similar to that I suspected as well (and already posted here before). 
Where did you get that information from ?
The only official Sandisk statement I know of is something like, it had to be that way, otherwise your player will quickly run out of battery and then explode :slight_smile: , no mistake on our side :slight_smile: That’s what I meant by “secretive”.

Message Edited by m9zf3n5w on 11-06-2009 03:26 PM

Ok, I originally started this thread.

I finally bought a 4GB Clip+ on Friday.

If there’s a pitch problem with it, I cannot hear it, unlike with the Clip v2. That’s all that matters to me.

Talk about full circle!

:slight_smile:

@miikerman wrote:

Talk about full circle!

 

:slight_smile:

How is it full circle? He started the thread asking if the pitch issue was present in the Clip+, and now that he has one, he can’t hear any pitch variance. :wink:

Looks like a circle reasoning to me- from 8/31 to 11/8, it’s nice to see that the original poster is happy with the new device.

All too often, I see a post, then by searching the user for history, they either do not log in again or do not reply whether the issue has been solved.  I’m glad that you’re happy with the new Clip+ !  I sure like mine, and its interface compared against its bigger brothers, the e200 and Fuze.

The Clip+ does consume more electrons than the basic Clip, trading capabilities with µSDHC for the familiar ring light, plus “snappier” response during communication.  So far, I welcome the changes, though the cool blue illumination of the controls is nice indeed.

Bob  :smileyvery-happy:

I was a little hesitant in buying a Clip+ because of the reported pitch problem, but I am giving mine some initial listens.  I must say, that to my ears, the slight pitch error on the Clip+ is not noticable to me.  I have a pretty good natural sense of relative pitch, played instruments when I was younger, in general am pretty obsessive compulsive when it comes to music, and it doesn’t bother me so far. 

I played back some tracks from an album on my desktop, then played back the same tracks on my Clip+, and the pitch increase of a quarter of a percent on my Clip+ was not immediately noticable to me.  Maybe, if I gave it some full attention, I might be able to detect it in a blind experiment, but maybe not.

Anyway, I thought I’d put my two cents in if anyone out there thinks their set of ears might be similar to mine and is curious as to another user’s opinion of the pitch issue.

Message Edited by richter on 11-17-2009 06:35 PM

Newbie here,  So let me jump right in to the DEEEEEP end of the pool…  I am trying to take a workaround approach to this and see where it goes.

   

I have read all 27 pages of this thread with interest.  I have a Clip,  Fuze and a new Clip+.  I recall that the pitch accuracy at 48khz sample rate on both the Clip and Clip+ is on par with other players.  I have all my music archived in Flac right now,  and can convert a working copy with dBpoweramp to either OGG or LAME at 48khz sample rates overnight.  I currently have a 44.1khz mp3 copy of everything  at q=0 vbr so windows media player has a clue what to do with my music.  I would of course keep my original flac files.

  

1.  If I do this to avoid the pitch issue,  will the resulting files have any more loss or artifacts than I would normally have converting to the same bitrate mp3 or ogg at 44.1khz?  I have usually been setting things to max quality ogg q=1 or lame q=0 vbr.  I am certain it wouldn’t be “lossless” to convert to 48khz flac,  so I wouldn’t bother with that for my archival files.

  1. I see that 48khz is the DAT standard.  Do most other mp3 players natively support this by default?  What about Windows Media player (using windows 7) or Winamp?  Other than Sansa I have an original Ipod with Video 80 Gb media player.

  2. Will I take a hit on battery life compared to a file encoded at the same mp3 or ogg bitrate vs 44.1 khz?  I know that “ogg is a hog” when it comes to battery life.

  3. I am not an expert on media, but am I correct in assuming that this would have no relevant effect on IDtag data?

Thanks… 

Gliderguy wrote:

I have read all 27 pages of this thread with interest.

 

You , sir, are a glutton for punishment! :wink:

As far as your workaround thoughts, first thing is to consider this…do you hear the pitch difference? If you had never read about the issue, would you be aware of it?

I read all about it here ever since it was first brought up, and I was never able to hear it for myself. I was able to prove to myself that it existed, by playing a Clip and Fuze in sync with a CD player and comparing the times they finished playing the track. But it never bothered me when I was listening to tunes…apparently I don’t have perfect pitch or golden ears. :smiley:

So if you can hear it, then explore your workaround (which seems pretty well thought out) but if not, just enjoy them the way they are.:smileyvery-happy:

I at least want the forum regulars to know that I will TRY to do my research before asking a question that might have been posted 648 times before.  :smileyvery-happy: 

This is also partly for the people who are so up in arms about the whole thing,  just to remind them that there is a middle ground out there somewhere between a million-man march to the Sandisk headquarters and “Oblivious Contentment”.

 If one is a serious musician (which would NOT be me) it might not be a great hardship to have a set of MP3’s that were encoded at 48khz sampling rate, at least for the critical files that you might be trying to play in tune with. 

It did really have me wondering if 48khz sampling rate for MP3’s might have any SQ benefits for ANY or ALL devices when converted directly from a lossless format,  so this seemed to be the right time and place to bring that up.

Also the power thing… if it takes a tweak to get the speed (nearly) right at 44.1 are we already paying a battery cost vs what SEEMS to be a native speed for the Clip’s D/A converter  of 48khz.  I suspect higher sampling rates  will automatically be more battery intensive,  unless there is a designed native speed where there is an optimum efficiency.

By the way,  I like the irony and alliteration of using Klipsch headphones on my Clip’s.  (Image X10s for critical listening and a set of S4’s for those times I am worried about damage or loss)

Message Edited by Gliderguy on 12-20-2009 06:00 PM

If you had advertised the defect; that would have been honest.  I just bought this for my viola-playing daughter. I’m busy I did not know Sansa made lo quality players.  Ruining her sense of pitch is a great idea.

 

Cut 10% out of the battery life if you must, fix the problem. I’ll do the workaround. Oh boy! I get to spend my time researching defects in a name brand technology and how to solve them. 48 kHz sample rates does emusic sell those files? Oh boy!

  

This is not some impulse purchase from Microcenter for $40. (yes you can get a 4 g mp3 player with a sizeable video screen for that price there.) 

 

IF you want to have a brand, either advertise like apple, (in which case you can sell ipods with batteries that fail or laptops like the 5300 and get away with it cause you’re hip (irony here)) or do not sell junk.  Fix the problem. Any of you remember when the Feds came to regulate analog audio specs (particularly power).  Maybe we need a standard spec sheet made available so I can spend my time listening to music rather than shopping and learning work-arounds. 

@bigbear9 wrote:

Cut 10% out of the battery life if you must, fix the problem. I’ll do the workaround. Oh boy! I get to spend my time researching defects in a name brand technology and how to solve them. 48 kHz sample rates does emusic sell those files? Oh boy!

 

10%???  That’s 1 1/2 hours!  The battery’s already limited to 15 hours!  Get real!  I want more battery power (and no, I don’t notice any ‘pitch’ issues so far).

@fla_hotrod wrote:


10%???  That’s 1 1/2 hours!  The battery’s already limited to 15 hours!  Get real!  I want more battery power (and no, I don’t notice any ‘pitch’ issues so far).

I bet there’s some potential to pick up more battery life if they reduce the S/N ration to 50 dB and the frequency response to 100 Hz to 10 KHz.  And you know what? Most people wouldn’t notice. 

I have read quite a lot of this thread and found it mostly interesting and certainly entertaining. I was intriqued with the idea that the Clip+ might be more efficient with 48khz files and made some test files and started by comparing 44.1 khz files and discovered that my unit played a 322 second file to within one half second of my computer - allowing for lag time with me pressing buttons and/or devices starting, pretty much identical… Given that I am not blessed (or cursed) with perfect hearing and/or pitch (I struggled in high school band), I can muddle through with my limitations and enjoy my $50 music player.

I just bought this for my viola-playing daughter. I’m busy I did not know Sansa made lo quality players.  Ruining her sense of pitch is a great idea.

She plays viola?  Don’t worry, she doesn’t need a sense of pitch anyway (ducks).

Seriously, though, don’t worry about it.  If she has absolute pitch, it is already developed and nothing you can do short of a lobotomy will change her ability.  If she is very strong here then she might find it uncomfortable, but that’s down to her.  Relative pitch perception is unaffected by the actual pitches she listens to. I have very keen pitch perception in certain domains and listening to a reference track I know well, I can feel the difference but it is certainly not significant. 

If I do this to avoid the pitch issue,  will the resulting files have any more loss or artifacts than I would normally have converting to the same bitrate mp3 or ogg at 44.1khz?

Not unduly.  Provided you are above the Nyquist frequency, sample rate conversion does an order of magnitude less damage to the signal than psycho-acoustic compression techniques. For completeness, I would carry out the conversion in FLAC and then compress to MP3 or Vorbis, but in practice I don’t think it would make much difference which you did first.

I see that 48khz is the DAT standard.  Do most other mp3 players natively support this by default?  What about Windows Media player (using windows 7) or Winamp?  Other than Sansa I have an original Ipod with Video 80 Gb media player.

Sampling rate is not really something you “support” as such.  Think of it as tape speed - provided your player accepts that kind of tape, you can play it.  The question is, does it have a speed control so you can match the playback speed to the record speed?  I would suspect WMP and iPod dont and Winamp does, but that is a guess.  

Will I take a hit on battery life compared to a file encoded at the same mp3 or ogg bitrate vs 44.1 khz?  I know that “ogg is a hog” when it comes to battery life.

 No.  The whole point of resampling is that the Clip+ won’t even realise there is a difference and so will feed it through at the correct speed (though that said, 44->48kHz does not sound right).

Vorbis is only battery hungry because nobody makes hardware decoders for it - it is actually a more efficient protocol than MP3 in computational terms.

I am not an expert on media, but am I correct in assuming that this would have no relevant effect on IDtag data?

Can’t see why.