Andy,
.....and perhaps read what you want to read 🙂. I didn't mention the TDA1541 as I haven't tried it, I hope to use one in my next DAC though.
Tim.
.....and perhaps read what you want to read 🙂. I didn't mention the TDA1541 as I haven't tried it, I hope to use one in my next DAC though.
Tim.
The TDA1541A with 1kHz -60dB has all even order harmonics in non os
and
K5 in os.
That sounds different.
and
K5 in os.
That sounds different.
TimA said:
😀 er yes, I had a data transmission error...
Bernhard said:
Different yes, what is K5 on OS - odd order I guess? Is that a feature of the DAC or of the Filter used? (ok, so its almost always the SAA7220)
I'm not sure this test is entirely "accurate". My understanding is when you defeat the filter you have to redo the I/V resistors to new values ?
hi Regal,
Nah, pretty sure its as accurate as you can get. part of the aim of the experiment was the 'all other things being equal' aspect
the DAC still sees the same components as before - ne5532 / op2134 op-amp in this case.
Please don't think I was trying to suggest an *optimised* setup of one topology was better than the other.
In theory you need to completely alter the filter characteristics post dac for the non-os business. But even that becomes a hot topic, causing any roll off at 20Khz hurts the OS 'argument' (if you are a bat 🙂 )
As I understand it the point of oversampling is you can have a pretty lazy filter that preserves phase in the analogue domain at the expense of a bit of digital jiggery pokery. Its a trade off, you're either going to be happy with one method or another and, If I may be so bold, its probably going to rest entirely upon the merits of your system and your perception of what music should sound like. Hence my have a go attitude.
Incidentally I've only tried this with bipolar power amps, has any one had a similar experiment with tube amps or fet's? - I have this vague idea that a SE tube amp and NON-OS might work well. lots of even order harmonics? 🙂
Andy
Nah, pretty sure its as accurate as you can get. part of the aim of the experiment was the 'all other things being equal' aspect
the DAC still sees the same components as before - ne5532 / op2134 op-amp in this case.
Please don't think I was trying to suggest an *optimised* setup of one topology was better than the other.
In theory you need to completely alter the filter characteristics post dac for the non-os business. But even that becomes a hot topic, causing any roll off at 20Khz hurts the OS 'argument' (if you are a bat 🙂 )
As I understand it the point of oversampling is you can have a pretty lazy filter that preserves phase in the analogue domain at the expense of a bit of digital jiggery pokery. Its a trade off, you're either going to be happy with one method or another and, If I may be so bold, its probably going to rest entirely upon the merits of your system and your perception of what music should sound like. Hence my have a go attitude.
Incidentally I've only tried this with bipolar power amps, has any one had a similar experiment with tube amps or fet's? - I have this vague idea that a SE tube amp and NON-OS might work well. lots of even order harmonics? 🙂
Andy
- Status
- Not open for further replies.