Now I measured lots of chips with both 7220A and B version, and both channels (!), 7 silvercrowns, 1 S1, 1 A, 3 plain 1541s, and the winner is the S1.
All harmonics <= -60dB with the -60dB sine.
But:
Still I'm not sure if this one sounds best because it is very warm sounding, maybe another chip that has higher low order distortion, but same high order distortion might sound better because the not so bad low orders help to mask the bad high orders.
The thing is that with SAA7220B chip, K2 is always below -67dB noise floor and K2 also with 50% of tested chips, so higher orders might become more noticeable.
With SAA7220A, all 3 plain 1541s have low orders around -58 and high orders around -63, but have 2 ( out of 10 ) "bad" values around -53.
Compared to that, with SAA7220B, the silvercrowns and the A go as bad as around -55 for K 5 on both channels.
Also funny is that with SAA7220A, the S1 performs worst of all.
All harmonics <= -60dB with the -60dB sine.
But:
Still I'm not sure if this one sounds best because it is very warm sounding, maybe another chip that has higher low order distortion, but same high order distortion might sound better because the not so bad low orders help to mask the bad high orders.
The thing is that with SAA7220B chip, K2 is always below -67dB noise floor and K2 also with 50% of tested chips, so higher orders might become more noticeable.
With SAA7220A, all 3 plain 1541s have low orders around -58 and high orders around -63, but have 2 ( out of 10 ) "bad" values around -53.
Compared to that, with SAA7220B, the silvercrowns and the A go as bad as around -55 for K 5 on both channels.
Also funny is that with SAA7220A, the S1 performs worst of all.
The analyzer destroys my dreams
Measured the Technics SL-P770 which seemed to sound much inferior to Philips players just some weeks ago.
Most harmonics are about 3 dB better, compared to best CD880 setup with S1 and 7220B.
Also noise floor gets better 5dB.
Seems the human ear prefers some extra distortion like tube amps and even el cheapo TDA1543
Measured the Technics SL-P770 which seemed to sound much inferior to Philips players just some weeks ago.
Most harmonics are about 3 dB better, compared to best CD880 setup with S1 and 7220B.
Also noise floor gets better 5dB.
Seems the human ear prefers some extra distortion like tube amps and even el cheapo TDA1543
till said:
I prefer the sound of TDA1541 with higher distortion - for the moment - to that of the better S1, but I did not like the 1543.
Also, as mentioned before, left and right channel harmonics are always different, this will affect soundstage.
If you have that el cheapo DAC chip, tolerances between left and right will be even higher.
Did you compare 1541 vs 1543 with spectrum analyzer ?
till said:
3dB difference is only for Technics vs. Philips.
The SAA7220B improves only the low order harmonics by 12dB ( a little more tan +/-2 ), high order harmonics remain the same.
I feel the 1543 has also worse, ugly sounding high order harmonics.
Parallel chips could improve the left /right balance of harmonics --> better soundstage, and lower distortions.
The problem with comparing chips is that they are so different.
One might be good, another one bad.
Anybody tested spectrum single vs. multiple parallel chips ?
And os vs. nonos ?
Now this is non oversampling in CD650
Sound is not bad from first impression.
But higher distortions.
Again the -60dB 1 kHz sine.
The plain 1541 got stronger odd distortions while the S1 got stronger even distortions.
What is believed to sound better ?
I feel this is a very complex problem because every single chip sounds different even if it is same type plain/A/S1.
Also every chip sounds (and measures) different with A & B versions of SAA7220.
Same for non oversampling, except SAA7220A/B doesn't matter..
Plain 1541:
Sound is not bad from first impression.
But higher distortions.
Again the -60dB 1 kHz sine.
The plain 1541 got stronger odd distortions while the S1 got stronger even distortions.
What is believed to sound better ?
I feel this is a very complex problem because every single chip sounds different even if it is same type plain/A/S1.
Also every chip sounds (and measures) different with A & B versions of SAA7220.
Same for non oversampling, except SAA7220A/B doesn't matter..
Plain 1541:
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.