In practice, yes. Filters only attenuate out-of-band signals, not eliminate them. The question is always how good it good enough.
Last edited:
And im just providing information also. He tries to prove that 44.1k is insuficient by showing us a mic with 40k BW and implying its should be a standard for live/recorded sound, and im saying hes wrong about the mic.
Yes, you won’t find many integrated digital filters that are fully into the stopband by 22.05 kHz because they would not meet the marketing requirement of being flat to exactly 20 kHz given the number of taps required to narrow the transition band. I’m not sure it’s that big of an issue in reality. Their paper did not do much to justify the argument, more just put it forth. I don’t disagree, though.
Last edited:
Plural of anecdote is still not data. And you now join RNM and JR in quoting your bulging mailbag as somehow showing your are right? It's not really a good debating point 🙂
Indeed. No interest in spending my time collecting data for now. More than enough other things to do. If some people want to visit, take a look/listen for themselves, and measure whatever they want, fine. After we have some of that maybe I will decide collecting data is something I want to spend my limited hobby time doing.
15ips vs 30ips. Bw 20k vs 22k, so what. Thats right a beloved Studder analog 2 track, max BW 22k. Why would you need a mic thats 40k?
STUDER A810 Professional Tape Recorder
Paul McGowan gets a lot of mail too, strangely in blank envelopes, all the same size with a single sheet of typed A4, what's that all about?
Interesting you asked us to measure 20kHz distortion on a system that is brickwall limited at ... 20kHz.
And more like 15ips.
30ips Vs 15ips! - Gearslutz
Almost all gear used for recording is an effects box, does it sound better thru it, not does it sound the same. Analog tape isnt more acurate, if used correctly it sounds better. One of the exceptions is the final mix, but not always, and one of the first digital recorders in studios was the 2 track DAT that was used to send mixes for mastering.
Question, as my google is failing me. How many cycles at say 4kHz does it take for our reptilian brains to recognise a tone, then how many more to say 'aha it's a picolo'?
I can see why a system buried deep underground might need to get everything right for a single cycle, but darned if I can work out what that has to do with Bruckner.
I can see why a system buried deep underground might need to get everything right for a single cycle, but darned if I can work out what that has to do with Bruckner.
In reality, 30ips recording can easily extend to 40KHz. I proved this with the Supermaster for Mobile Fidelity, and the Ultramaster for Wilson Audio. More commercial products from both Studer and Ampex can approximate this IF they want to extend their bandwidth by NOT resonating the reproduce head at 20KHz or so.
Here are some examples:
Only 24-96K, or DSD can approximate this. CD's can't, obviously. That is why Richard Marsh and many others, including me, tend to concentrate on the faster sample rates. Of course, we HAVE to listen to 44KHz day in and day out, but that is not the best hi fi reproduction that is practical or possible.
Here are some examples:
Only 24-96K, or DSD can approximate this. CD's can't, obviously. That is why Richard Marsh and many others, including me, tend to concentrate on the faster sample rates. Of course, we HAVE to listen to 44KHz day in and day out, but that is not the best hi fi reproduction that is practical or possible.
Attachments
I still have one on my shelves. Howful, so many drops ! There were better alternatives (ADAT etc)
We did with the best we had and that authorized exchanges.
To get back to the analog 15 VS 30, the problem of 30 was the bass frequencies non linearities rejected in a more audible frequency range(the resonances due to the width of the head). But the trebles were better.
I don't understand why you advocate 16/(44/48) is enough. 24/96 is objectively better and no harm.
I bet my shirt that the 44.1 will not survive much longer than the CD, already endangered. We lose our time.
Now, all this is not essential, compared to the musical quality of a record.
Last edited:
Here's another set of files for those interested.
WeTransfer
Not a plain cymbals, but nevertheless. This time I started with DSD file (1 bit, 2.8 MHz) and decimated it down to:
1. 24 bit 88.2 kHz using TPDF dither
2. 16 bit 44.1 kHz using TPDF dither
and then upsampled 16/44.1 file to 24/88.2 using the same industry standard software that was used do produce DSD->PCM conversion.
This should be a valid 88.2 vs. 44.1 test, I think.
It's a commercial release, so it's only a small excerpt.
Download link will expire in 5 days.
Have a listen and have fun.
WeTransfer
Not a plain cymbals, but nevertheless. This time I started with DSD file (1 bit, 2.8 MHz) and decimated it down to:
1. 24 bit 88.2 kHz using TPDF dither
2. 16 bit 44.1 kHz using TPDF dither
and then upsampled 16/44.1 file to 24/88.2 using the same industry standard software that was used do produce DSD->PCM conversion.
This should be a valid 88.2 vs. 44.1 test, I think.
It's a commercial release, so it's only a small excerpt.
Download link will expire in 5 days.
Have a listen and have fun.
Last edited:
Right and most music now is delivered over wires and compressed ( data rate) ( rotten Apple) and the public buys it up. So CD might be dead but the alternative might be worse.
So why dont they? Because thousands of recording engineers say it dosnt matter. Why do you think you know better?
The argument would be transients, but I doubt any music has large transients at the speed of 20kh.
So that nobody, without further analyzing the files, will be able to tell which one is which. Think A/B/X testing.
P.S. If anyone is interested I can upload the 16 bit version as well.
Last edited:
- Status
- Not open for further replies.