Maybe the extra mini transients weren't handled accurately enough by a limited bandwidth reproduction chain resulting in "muddiness"
Perhaps what you perceive as "muddiness" is more accurate to the original recording than the BW limited version
Perhaps what you perceive as "muddiness" is more accurate to the original recording than the BW limited version
Last edited:
...and the 'packing' of 44kHz in an 88kHz file was accomplished how, zero stuffing, duplicated samples (sorta- zero order hold), other?
EDIT: Not going to complain about however it was done, just curious. The results are not surprising to me. I have said before a lot of detail exists in 16/44. Also, it occurs to me that making all the files play at the same sample rate in the dacs takes any effect of dacs reproducing audio differently at different sample rates out of the equation.
Last edited:
I did post another option in an edit 😉 The most interesting thing is that you heard quite a dramatic difference (so you say 🙂)
Last edited:
Yes it’s not something I’m even slightly doubting. Like I said B and C was almost a coin flip,
but A was by far the worst of the bunch......odd indeed.
but A was by far the worst of the bunch......odd indeed.
Bob, If the files were scrambled and relabeled do you think you could reliably pick out A by ear?
It could be that A is a bit muddy, but I downloaded it without any modification straight from 2L.NO.
And all I did to create B was to brick wall filter A while keeping Fs the same, just carefully doing one single step at the time nothing else.
And then C was also just one step from B by saving the 24 bit file into 16bit.
The beauty to my opininion of this sequence is that everything else remained unaffected like Level, Fs and the Dac reconstruction filter.
I could start all over again with the DXD version, downsample it to 96/24 or 192/24 and do the whole exercise again. Would this satisfy you ?
Hans
Thank you.
Not the same process at all used by DAW software to produce CD audio from higher sample rate source material. Probably explains part of why your files sound so much better than is commonly the case.
Just a straight up mixed/unlabeled comparison of these three downloaded to Dropbox......I have no way to identify which ones which.
Attachments
Last edited:
When Brick Wall filtering the 88.2/24 file, this was done with a fir filter working in 88.2/32 float for optimal accuracy, the result stored again in 88.2/24.
No zero stuffing or anything like.
Hans
Matt might be right, maybe your on to something Hans..... I just did another comparison between A and B , again preferring B
Could also be I’m just so used to sub par I actually prefer it!
Could also be I’m just so used to sub par I actually prefer it!
Ha, ha, the only thing I was on to was to find out whether perceived differences have anything to do with other factors as level, upsampling filters, reconstruction filters or whatever that can manipulate the signal in the reproduction chain.
All these variables have been excluded in this test.
Hans
Might try it again on a better CD player. I also bought the top OPPO thinking it would be my last CD player. But it sounded so bad and I could do nothing that improved it at all. (jitter?) I didnt even try to sell it... was new .. just gave it away, as i was so disgusted with it.
THx-RNMarsh
yep. Sounded that way to me also.
'A' was way bad. Obvious with my computer dac and old ears and ear buds.
So, it had to be really bad to sound so much worse than B and C.
Might recheck what was done on that A file?
THx-RNMarsh