Hello,

there is strange characteristic in my PCDP Operating instructions:
Is it normal, or is it just a typo? Should it be '+0.5 to -0.7'? (just to compare: all recent cd/mp3 walkmans have frequency response 20-20000Hz, +1 to -2 dB)

The player is panasonic sx428 and everyone who has recent panasonic cd/mp3 players can see exactly the same specification for frequency response in their operating instructions (e.g., sx430, ct510, ct710, ct810).

I tried to measure up frequency response with RightMark AA and my soundcard (cmi8738-based, not that good for measurements, but anyway) and got value of '+1.05 to -0.47'. I see that ripple of 1.5 dB is not as bad as 7.5 dB, so, maybe there was just a misprint in specifications?

edit: grammar

That certainly doesn't sound right to me. Ripple (or just LF and HF rolloff) of 7dB across the audible range would certainly not count as HiFi. Maybe you should contact Panasonic and ask them the clear up the specification with you.

It's very probably that was a error in the specifications manual...

7.5 dB of variation ripple is very bad...

Excellent is for example:
20Hz to 20Khz [+0.5dB to -0.5dB]

Very good:
20Hz to 20Khz [+1dB to -1dB]

I believe seriously (but I'm not sure) that is a error in your manual.

Is that maybe the total system response, including the earpods that came with the player? -7dB at 20 hz for cheapy earphones wouldn't be unusual.

Response could be -7dB at 20KHz. Even the "almighty" Sennheiser HD580/600/650 series has fairly notable rolloff at 20KHz. But I highly, highly doubt any modern electronic product made by a major player (Panasonic) would have a -7dB ripple anywhere in the audible spectrum, i.e. most people can't hear 20KHz tones within music anyway.

I believe that is only the frequency response of the player (in the specifications manual).

Which isn't necessarily a bad thing. CD is arguably fundamentally limited to about -6db at 20kHz for one. For other, smoother output filters would possibly actually sound better than harsh no-matter-what goal of having 20kHZ at 0db.

Umm. Any decent can MUST have a fairly notable roll-off beyond 12kHz or else its a dog killer. Its to do with natural HF roll-off due to head transfer and normal air losses, that headphones MUST simulate, being in direct proximity to ear canal.
So - its not something that Sennheiser is unable to achieve - they make them roll-off on purpose, and its something that is decidedly difficult to make right. Whenever you see headphone specs with perfectly flat frequency response - laugh.

If there is a good reason behind it and its a design decision, then actually I'd be pleased to see them showing honest specs despite all the marketroids.
But better ask their support to get the answer.

I spent some time and collected various Panasonic's PCDP specifications.
It seems they measured ripple (deviation) using entire sound reproduction system (with earphones) or they did something wrong. Besides, I did not find any 'measured according to JEITA' remarks about frequency response value. Models of 2005 seem to have more realistic specs

2005 models
sl-ct720: 20Hz~20kHz(+0.5dB~-1.5dB)
sl-ct820: 20Hz~20kHz(+0.0dB~-1.5dB)

2004
sl-sx428/430: 20Hz~20kHz(+0.5dB~-7dB)
sl-ct510: 20Hz~20kHz(+0.5dB~-7dB)
sl-ct710: 20Hz~20kHz(+0.5dB~-8dB)
sl-ct810: 20Hz~20kHz(+0.5dB~-8dB)

older
sl-ct790: 20Hz~20kHz(+0.5dB~-10dB)

Such limitation doesn't exist in CD format, not in theory nor in practice. In today's good CD players and soundcards at 44.1 KHz it's common to see actual rolloffs of just around 0.5 dB at 20 KHz. Also, there's nothing wrong with this, there's no reason why it should sound harsh, given that the reconstruction filter is well designed, as most are today in players and soundcards. Flat response from 20 Hz to 20 KHz is what CD is supposed an intended to be, and was designed with that goal in mind. With today's hardware it's cheap and easy to achieve.

-7 dB at 20 KHz could or couln't be audible, depending on the implicit rolloff at lower, more audible frequencies, and also on the listener's ears.

About the original poster question, if you have measured the PCDP output with your CMI8738 card and found it much better than the -7 dB in specs, then the -7 figure is probably some kind of typo, in my opinion.

Edit: elaboration.