You can use Foobar2000 (freeware) to ABX the files yourself.
In any case, you don't know what files are from what ADCs, so at least you are blind to that.
The wav file names I have are quite obfuscated from the ADC models, and I don't remember which one is which. There are a couple of scrambled lookup tables to sort through. But I will be happy decode the file names and to provide all the information after people try to see if they can hear any difference. I don't want to bias things unnecessarily at this point.
They are some very interesting threads, with pretty good S/N on the discussion. Has given me much to digest and research further, which is always good.
It wasn't available 10 years ago, so far as I know. So I am already spoiled. Don't let the same thing happen to you.
By the way, do you believe people can hear undithered 16/44?
A couple quick points:
1. Use the same EQ/preamp you liked to listen too. Never mind over analyzing why you like it. You may not like the answers.
2. A/D conversion of audio REQUIRES noise to dither the LSB. If your pre-amp doesn't have enough noise, then you have to decide the shape (Gaussian, triangular, rectangular,..) that you will add/use in the A/D process. A possible exception is the delta-sigma A/D conversion method which is based on dithering.
1. Use the same EQ/preamp you liked to listen too. Never mind over analyzing why you like it. You may not like the answers.
2. A/D conversion of audio REQUIRES noise to dither the LSB. If your pre-amp doesn't have enough noise, then you have to decide the shape (Gaussian, triangular, rectangular,..) that you will add/use in the A/D process. A possible exception is the delta-sigma A/D conversion method which is based on dithering.
Yes. I have tested this formally in DBLTs.
Not everyone can .. eg most (all?) of da Audiophools (who claim to hear 'chalk & cheese' between cables/capacitors bla bla) can't. ie they are deaf in the context of this test.
Last edited:
Do you mean you didn't believe prior research to that effect?
And, yes of course I will agree with you there are a lot of people who think they can hear things they can't.
Not sure what you mean.
Is that NO, it won't allow you to define a FIR?
If not, how many bi-quads can it do?
George, I take it you can use either loads bi-quads or a 'large' FIR
My first set of format tests was circa 1980 so probably some of the earliest research into the subject .. at least for HiFi. At that time, there were certainly many big $$$ 'professional' A/Ds that weren't properly dithered.
The DBLTs was part of some stuff to show some big wig Marketing Audiophools that 'Digital' was OK. Another part was showing a sine wave or piano music was easily heard 'distortion free' 20dB below the noise floor of a properly dithered 16b channel. Fairly easy then to find big $$$ crap codecs where this wasn't possible.
My next set of formal tests on the subject was circa 1990 in the context of 'self-dithering' filters.
Knowing if someone's opinion on this (and some other related) subject(s) is (or not) from a properly conducted DBLT tells us its importance & relevance.
But I'm tired of wanking. Do you have anything useful to add to the subject of Digitising Vinyl?
I'm still interested in which ADCs you used and how the files were made. You don't have to tell us which is which.
Last edited:
Hah! Guru Wurcer's 2 x 1st order bi-quads use 4 feedback and 4 feedforward coeffs.
So I'm going to claim the SAI version is only 0.2dB out for the same computational load 😀
_________________
Have we come to any concensus about Trustworthy, Readily Available Test Records yet?
FIR or IIR or zillion bi-quads?
I think we are very close to getting a good practical solution. I hope 0.2dB is good enough for George & Bill 🙂
No.
For now, this: Old 2x4MiniDSP (48KHz)
No FIRs for the old 2x4MiniDSP (48KHz). Only IIRs (biquads):
“Input Parametric eq” section: 6 biquads /channel
plus
“Cross-over” section: 16 biquads /channel
plus
“Output Parametric eq” section: 6 biquads /output
https://www.minidsp.com/images/documents/Product%20Brief-%202way%20Xover%2021%20ADV%20plug-in.pdf
-------------------------------------------------------
Maybe in the future, this: New 2x4HD MiniDSP (96KHz)
Both IIRs (biquads) and FIRs for the new 2x4HD MiniDSP (96KHz):
“Input Parametric eq” section: 10 biquads /channel
plus
“Cross-over” section: 8 biquads /output (or 16biquads/channel)
plus
“Output Parametric eq” section:
Either
10 biquads /output
or
6-2048 tap FIR/output
https://www.minidsp.com/images/documents/miniDSP%202x4%20HD%20User%20Manual.pdf
George
Ah the 2x4HD is more advanced than my 4x10. I only have 5 PEQs on input, xover and 5PEQs on each output (using 96Khz). And I lose the compressor. This is way more toys than I have ever had before so I don't feel I am missing anything.
Sorry to keep popping in links but found this in my travels.
Pro Audio Design Forum • View topic - Flat Phono Preamp Based on John's P10 and 2SK389
Not something I currently need but useful if you do ripping as it allows you to monitor the live feed whilst sending flat to your PC and switch to what is coming out the PC. I hadn't thought about having this remote from the turntable, but makes a lot of sense as the computer (and user) are generally remote from the turntable, at least in a domestic setting, so being able to run significant cabling with no problems is good.
Pro Audio Design Forum • View topic - Flat Phono Preamp Based on John's P10 and 2SK389
Not something I currently need but useful if you do ripping as it allows you to monitor the live feed whilst sending flat to your PC and switch to what is coming out the PC. I hadn't thought about having this remote from the turntable, but makes a lot of sense as the computer (and user) are generally remote from the turntable, at least in a domestic setting, so being able to run significant cabling with no problems is good.
Interesting can they be configured as 3 - 4096 FIR's? If not it might be an interesting deconvolution exercise to decompose a 4096 FIR into 2, 2048 ones.
Scott, may be I confused you.
They write (page 33 section 6.5.1 of the owner's manual) you can import FIR with minimum 6 taps and maximum 2046 taps per output.
George
They write (page 33 section 6.5.1 of the owner's manual) you can import FIR with minimum 6 taps and maximum 2046 taps per output.
George
😡 Looks like a quick solution would have be 5 or less bi-quads.billshurv said:
2 bi-quads for Guru Wurcer's RIAA. We can probably fudge these to do Aurak type EQ ... where some of RIAA is done with cartridge inductance & resistance.
3 left to twiddle cartridge stuff ... mechanical resonances & to simulate 'capacitive loading' if Aurak virtual earth is used.
Looks like its Ramos & Lopez http://www.diyaudio.com/forums/analogue-source/298896-digitizing-vinyl-34.html#post5053444 😡
_____________________
2048 pt FIR for each channel would be the simplest for the unwashed masses .. I mean the aficionado who wants to digitally EQ his record playing system without mind bending maths ... using probably GNU Octave. 🙂
No need for beach bums.
Last edited:
VST RIAA plugins
I thought Wayne's recommendations for RIAA plugins (and da bad ones) were on his supa dupa preamp threads (which Bill has Shanghaied by claiming he found them 😡)
They are actually here
Pro Audio Design Forum • View topic - Phono Transfer System
The free Diamond Cut RIAA VST plugin that he likes seems to have gone from their website.
And IIRC, Wayne had a list and more comments about bad ones too on another thread.
I thought Wayne's recommendations for RIAA plugins (and da bad ones) were on his supa dupa preamp threads (which Bill has Shanghaied by claiming he found them 😡)
They are actually here
Pro Audio Design Forum • View topic - Phono Transfer System
The free Diamond Cut RIAA VST plugin that he likes seems to have gone from their website.
And IIRC, Wayne had a list and more comments about bad ones too on another thread.
- Status
- Not open for further replies.