Well, I stay corrected.
Forgive-me to had put yourself in the group of the dangerous illuminated.
It's just that you bring plausible technical explanations (consistent with mine, more intuitive) to the phenomena that RNM and I never stopped observing and reporting.
Nyquist's theories don't interest me that much, out of an intellectual point of view. What interests me is that the surface of half a period of any audible musical signal is reproduced as accurately as possible, if you know what I mean. The more points we have to design-it, the more I enjoy ;-)
There is a time more points increases the cost and complexity without making large profits. Everyone has to fix his own limit.
My personal limit is 44.1 is good. 24/96 slightly better, 24/192: I don't hear the difference... till now.
Last edited:
For me, the question of cost vs advantage would be a real concern if I were to own a studio and I were coming to the point where I had to decide on new electronics and what format.
Now, I could actually try a high bandwidth mic with various rates on the difficult instruments to see if the difference is worth the cost. A more informed decision as it were.
Also, as memory continues downward, I could decide to go high rate at the front end, and dumb down (as it were) for the format currently popular. Should the popular format rate go up, the already recorded material is there.
Jn
OK. Are they? Is that what we actually get when we buy and use? In practice? is that what we hear?
before that article was pointed out, I also asked what is the fastest signal Tr to be recorded and then go 5-10 times that. That got danced around also.
If it can be done, why doesnt TEK do it long ago in scopes?
Thx-RNMarsh
Last edited:
Double responsibility. Produce the best possible musical objects for today's market, limit as little as possible future progress in this artwork reproduction.
What we have always done, trying to make the best possible compromises between the studio big monitors listenings, and "Radio" listenings.
Didn't 5x turn out to be a bit higher sample rate than DXD? Or, was that about something else?
2MHz/18bit should be enough for anything happening in music. However, Richard, how would you transmit it through speakers? And let's not forget how heavily limited were the master tapes. 8us best?
The trouble with inverted rise time is it raises the channel bandwidth to roughly half a megahertz. I truly don't see that need for audio.
Edit: actually on inspection, you didn't say that, sorry.
I could see using a 40k mic and testing humans for discernment at various rates as a start. Using a 20k mic serves no purpose in that kind of test.
Jn
Edit: actually on inspection, you didn't say that, sorry.
I could see using a 40k mic and testing humans for discernment at various rates as a start. Using a 20k mic serves no purpose in that kind of test.
Jn
Last edited:
The 26uS only applies when you get to +-1 bit sine stream, if it is =+3 bits you get about 7uS or so, that means 1 -90db sine is difficult to locate in space, well if it is burried down in the backround noise I think you can expect it to be difficult to locate.
You can enjoy super cars, even with a cruise control ;-)
Mics are not hidden behind brick walls. And we use response curves correction everywhere and differently on each instruments in the mixing desk to sculpt the result. Lot of effects in action: slew rate, response curve, phase curve.
Don’t worry, my mind is always open without any prejudice.
Seen in that light I don’t understand the word naive at the end of your posting. Nowhere in the document for which I have provided the link is mentioned that 44/16 is not enough, so anybody who reads this is guilty of wishful thinking.
At this very moment, I blame much of perceived difference on mediocre recordings and or equipment.
I would be the first to admit in case I could hear improvements from higher sampling rates.
But this can only be done when starting with a proper HD master comparing it to a downsampled 44/16 version.
Several times friends have tried to show me the superiority of DSD against 44/16 on discs with two sides, one for cd and one for SACD. At first it seemed like a better sounding format, but the recordings where different and so did the equipment.
After having downloaded the DSD master and having compared it to a downscaled 44/16 with JRiver, all the differences where gone.
Same experience with 96/24 and 192/24 master files.
My DAC is a PCM1792A reproducing all input to analogue at 192Khz, so all files are played over the same chain, only file that is offered differs in sampling rate and number of bits. To my feeling this DAC is one of the top ones available with excellent digital filters. So untill now I have heard no reason why 44/16 is not enough, but because of having an open mind I asked Richard to send a file to convince me, but so far without result.
But all this is just personal experience and I will not stick to it when convinced with proper material.
Better from a technical point of view does not automatically mean an improvement in perceived sound.
This is almost in contradiction with your earlier statement, but this time I can fully agree.
What I want to prevent is jumping to conclusions based on false facts or on someone else’s hear say or even plain BS.
When you want to compare things, exclude all possible differences.
When someone can sent a decent file that demonstrates a difference in whatever form, I’ll be the first to admit.
Hans
P.s. a few pages ago a file was made available for which I hadn’t yet the time to listen to, but to my surprise nobody else has reported his findings so far, neither from you.
In photography, using automatic white balance do not always provide the best subjective believable results (and atmosphere).
Unfortunately, the quality of downsampling processing can make the downsampled version sound worse. It might be best to start with something like 24/192, then use the same software to downsample it to two different lower sample rates which are then the files to be used for comparison purposes.
i agree on everything you wrote here, while I don't see any contradiction where you saw one ;-)
Sometimes we can invest a little bit more than we need just for our peace of mind. I'm less tired after one full day driving with my big car than I was with my little previous one. Hum, with the price of the gaz, here, my pleasure is more and more limited. Yes, i know, contradiction ;-)
Last edited:
8usec is better than 26. When I made my own master recordings of musical instruments, I used cap. mic's. When you play a non copied, original master, it sounds so much better than CD. So, something funny is going on.
Now with 24/192+ It is more like that master tape. So, whats going on? As pointed out, it may not be frequency response to focus on. Or Tr ? Sampling of Tr? Cymbals are a blur with CD. Higher speed ADC/DAC are a lot better. Sounding.
Looking for answers and not excuses to stay with CD (or LP). Some progress has been made here into new areas to explore. But, it's been like pulling teeth. 😱
THx-RNMarsh
Last edited:
My point was, to sample at 44.1 required bricking the analog prior to sampling. Comparison of a 20k mic and a 40k mic both fed into a 44.1 system is not very useful. Comparing both fed into a higher rate system is useful.
Jn
Richard, you are wrong.
Pulling teeth is far more pleasant. You owe apologies to the dentists out there.
Jn
Since the upper side band falls outside the limits of human perception anyways, how could it be discernable?
Oh. OK. I am so sorry to all dentists out there. You do a wonderful job of pulling the truth, I mean tooth, out of peoples heads with no pain what so ever. Its truely an amazing thing to watch such masters of the tooth profession dong their good work. And, yes, i do believe in the tooth fairy. It just seems so real to me. Thank you for teaching it to me.
🙄
-Richard
Last edited:
- Status
- Not open for further replies.