Two IDENTICAL units, different sound!?!?

[bear]

Well I thought I would post it...

"...submitted for your approval..." Rod Serling

Anyhow, I acquired a 1 year old lightly used DVD player at a hamfest. And, after bringing it home thought, because my venerable Esoteric bit the dust, why not try it as a transport... into my custom PCM63 based DAC... which I did.

Quite astounding, my jaw dropped.
Illusions shattered - almost analog purity of sound, relaxed, free of digititis and electronic aspect. The ******* "holy grail" of digital audio!!

Amazing.

Ok, it did seem to want just the right digital interconnect, which still to this day bothers the heck out of me, but it was good from the start... please keep in mind that my system is just very good and I don't mind saying it, even though I'd say it is only at about 75-80% of what the inherent setup is capable of if optimized in several respects (not important for now). Point is that you can come here, bring your <whatever it is> plug it in and really hear it clearly and immediately - good or bad. No ambiguity. Ok, enough telling everyone how good it is, the only point is that you can hear what is going on quickly and clearly. For better or worse.

So, I figure I have stumbled on something BIG here... a cheap DVD that does magic! Time to get a backup for myself, before I tell the world... after all got to have a backup in case it croaks (as they seem to like to do). In fact that is why I bought the darn thing, because the last two CHEAP DVD players I bought CROAKED quickly for no apparent reason.

I got another on eBay.
With great anticipation, plugged it in, same outlet, same interconnect to the DA, etc...

BAH.

NOT THE SAME.

Not awful, but not magic.
Much more digitalized, thinner "solid statish" bass, harder highs, a bit more grain...

Must be impossible. But there it is. Two seemingly identical units, different sound. Crud.

This weekend I will pull the tops and look at the chipset and the parts to see if there is any variation in the runs or PCBs...

But, two otherwise identical units, different sound.

Bits is bits??

_-_-bear

[compressit]

I bet 1 adult beverage;
Same processing boards,drive, chipsets in both.
Different power supplys.

[Bill Fuss]

It's most probably laser alignment, second unit is having a much tougher time reconstructing the stream

Bill.

[analog_sa]

Originally posted by bear

Must be impossible.

It would be strange if they sounded the same. There will definitely be different components somewhere.

Similar break-in time?

Cheaply made equipment is commonly not consistent from sample to sample. At the time of the NAD 3020 several of my friends owned a unit and they all sounded differerent. From very impressive to completely mundane. After examining the boards in detail i found 6 different types of components between the samples.

[bear]

Ok...

- no adult beverages present, different auditioners on different days
- the "stream" has only one reconstruction because it is Reed-Solomon code
- break-in time is unknown, as is IF it could possibly be a factor...
- power supply? dunno, will open up over the weekend and look.
- components? same, will look...

They really do need to sound the same unless there is some real difference in the physical realm...

_-_-bear

[pilli]

Originally posted by bear

...unless there is some real difference in the physical realm...

Could the difference come from "different aging and stress" on otherwise identical components"?

Like maybe one has a dirty/dusty laser,
or in the supply, which is maybe SMPS like in cheap DVDs, one unit is affected with "bulging capacitors", and the other isn't (go figure, maybe different lots in the two players, one luckier than the other)

Or something like that.
But it'd have to be something extreme, where the "stressed" component is quite far from its nominal value/behaviour.

(sorry for this obvious troubleshooting hint...
I'm kinda surprised that someone with your experience is so surprised... )

[bear]

Not so much surprised as disappointed...

I will open them up shortly and see what there is to see. Maybe look at things like PS rails, if there is time... Doubt there will be anything glaring or obvious.

The real question is why does the "good one" sound so good!
Not why the other one is merely ordinary.

This is not a troubleshooting issue.
Neither is "broken".

Again, the laser being weak, bad, dirty or whatever plays nil role in the reconstruction of the signal, IF it reads at all. It can't because of the way CDs are encoded to avoid issues of this sort. At least that is what the book says.

My experience? Who says I have any that is worthwhile?? Ask anyone who really knows what they are talking about?

_-_-bear

[richie00boy]

Originally posted by bear
Again, the laser being weak, bad, dirty or whatever plays nil role in the reconstruction of the signal, IF it reads at all. It can't because of the way CDs are encoded to avoid issues of this sort. At least that is what the book says.

I would disagree with that. There is only so much the error correction can do, it might end up replacing a few dodgy segments with random. CDs aren't all or nothing "digital", the read process is analogue.

Music CDs are different to data CDs, the latter will let you know the data is corrupt, the former will struggle on as best it can until the errors are so frequent that it skips.

[analog_sa]

Originally posted by bear
Neither is "broken".

There is another way if you want to get to the bottom of this: hook-up the outputs to the digital-in on a sound card and compare if both players produce bit-perfect outputs. If they do you can point your attention to the spdif circuit and PS. Or the crystal oscillator.

[bear]

Originally posted by richie00boy


I would disagree with that. There is only so much the error correction can do, it might end up replacing a few dodgy segments with random. CDs aren't all or nothing "digital", the read process is analogue.

Music CDs are different to data CDs, the latter will let you know the data is corrupt, the former will struggle on as best it can until the errors are so frequent that it skips.

You are incorrect, afaik.

Read about the encoding method of Reed-Solomon.
The read process itself is analog, but then the digital reconstruction filter is applied. When it skips, the errors have grown to the point that the original data can no longer be reconstructed using the inherent error correction of the code.

_-_-bear