Bootstrapping does nothing for noise. One might ask why 200Meg resistors certainly special order non-standard values. AB only made carbon comp 1/8W up to 56 or 62 Meg in the standard sequence IIRC (try and find them now). Four of these in series would work fine. Before you say it, no DC current no excess noise, just plain physics please.
Okay, on first dissection, the 192kHz version has some genuine ultrasonic in it but it falls below 40k frequency. All the material above this frequency is residual, borderline noise, nothing above -60dB down. It is barely possible to hear normal audio at this low level, so I think we can safely ignore it.
Which means, though I haven't tried it yet, that there should be no meaningful differences between 192 and 96 versions, of that track ...
Will continue ...
Frank
BTW, has it registered with anyone, especially with those wetting themselves with excitement about DSD recordings, that on the 2L website the ultra high resolution files are also available as DSD, and these are 1/4 the size of the DXD version, both zip'd ? Now, people into computing and knowing how compression of data works are aware that one can pack data down to an irreducible level, where if you go any further you start to discard real information.
So, at a simplistic level DSD has thrown away 3/4 of the real audio data when encoded that way, not a good sign to my way of thinking ...
Frank
So, at a simplistic level DSD has thrown away 3/4 of the real audio data when encoded that way, not a good sign to my way of thinking ...
Frank
A very big hmmmmmmmmmm ...
There's somethings very dodgy going on here -- yes, the two downloaded tracks, 96 and 192 sample rate, are different, but they shouldn't be, it's in the wrong place !! Meaning what? That, they differ most resoundingly at frequencies way below 20kHz, someone's pulling a swifty, big time ... !
I took the right channel of that track, and got a difference between the two after time aligning -- anyone else who wants to try this needs to be aware that these 2 versions, at least, are not in sync. And got a rather glorious sound file, which peaks at between 200 and 1,000Hz after plotting the spectrum!! There is something very, very wrong here ...
No wonder Esperado heard them as being different, the difference peaks at only -30dB down ...
Frank
Last edited:
Frank, you mean there's obvious noise modulation? 😀
As for DSD throwing away most of the data, to my way of thinking random noise doesn't compress very well. However .zip is poor at compressing music, I believe .rar is better.
As for DSD throwing away most of the data, to my way of thinking random noise doesn't compress very well. However .zip is poor at compressing music, I believe .rar is better.
Last edited:
Well everyone, I hope I have given you enough to make my point. However, I am still missing (they are here somewhere) a complete set of noise measurements taken by B&K where they changed the microphone's R's to higher and higher values. It will come around sooner or later. I'll give you a sample, with 100G resistors.
Attachments
I think the need to modulate people's thinking to the idea that hi res formats do make an audible difference is the "noise" here ... 😛
Resampling is purely a software process, no excuses for it to not work precisely, I've done this exercise many times in Audacity, up and down -- there are no "funnies" that climb on board ...
As a general rule, music files can compress by about 50% at best. This is FLAC and all the other lossless formats; once in this form I would expect every compression algorithm to be struggling to get any more 'squeezee'. Of course, if you want 51% compression rather than a mere 50 just be prepared for the PC to grind for a number of hours ...
Frank
Oh but hi res formats do indeed make a difference. They're worse for noise modulation with multibit DACs. Through an S-D DAC though, since the final output symbol rate doesn't change they might sound better.
Back then there wasn't much of a standard on I/O signal polarity wiring. The European and American mics and equipment were often different in their wiring for polarity. -RNMarsh
RNMarsh,
I was told that you need a very rare adapter to attach the 1/4" capsule to the 1/2" body, not sure how easy those are to find these days. I asked my friend who still had an old catalog from the 70"s who has a complete B&K lab setup including turntable, strip recorders, mics, gates, and everything that most of us just dream of.
Steven
I was told that you need a very rare adapter to attach the 1/4" capsule to the 1/2" body, not sure how easy those are to find these days. I asked my friend who still had an old catalog from the 70"s who has a complete B&K lab setup including turntable, strip recorders, mics, gates, and everything that most of us just dream of.
Steven
Measurement only -- yes. The 2615 is for a 1/4 inch capsule.... where's the 1/2" adapter issue come from? -RNM
Last edited:
RNMarsh,
I was under the impression that the 2615 was a 1/2 body. If 1/4" then I am more than interested. I will have to get the capsule and the power supply but I see those in Ebay many times. Do you have any use for some 6 1/2" bass mid drivers, I could give you a pair of my own devices.
I was under the impression that the 2615 was a 1/2 body. If 1/4" then I am more than interested. I will have to get the capsule and the power supply but I see those in Ebay many times. Do you have any use for some 6 1/2" bass mid drivers, I could give you a pair of my own devices.
Because they probably learned physics in school, so loaded capacitive capsules on higher dynamic resistance than they could get statically using available resistors, to get better frequency response on low end.
Actually, it was a routine even before that, because tubes had significant grid currents that limited values of resistors that won't allow to use bigger values of resistors.
Last edited:
- Status
- Not open for further replies.