jerryct

Hi Barry, i know - i already said this some posts ago (Post #15): Of course not very well. I provided an example not 24Bit vs. 16Bit but 16Bit vs. 8Bit. With 16 Bit vs. 8 Bit it is more obvious what happens with the sound and neither 8Bit undithered (wrong - you hear distortion and noise) nor 8Bit dithered (better - but you hear noise) is great for listening. And i also agree: why bother with 16 Bit if you can provide 24 Bit. (but it may be the case that we have different opinions whether 16 Bit is sufficient or not because i still think it is only about signal to noise and thus dynamic range ) This is the example "down conversion to 8Bit" Dynamic Range, Dithering and Noise Shaping jerry

I think it is common sense that you need dither if you go for a reduced word length. Then with this statement you also need dither (infinite -> 24 Bit). Whether it is adding noise (i.e dither) in the digital domain (a digital filter) or analogue domain (noise of electronics) shouldn't matter. The only problem with this statement is thinking of analog is infinite whereas digital is not: This comes from the point that the bit depth or word length means something like resolution. But you can represent any signal with any bit depth + noise. http://productionadvice.co.uk/no-stair-steps-in-digital-audio/ Another interesting point - the Metric Halo documentation says: This is important if you want to compare 24 Bit vs. 16 Bit. In this case you will hear of course a difference but one reason is you did no dithering and not because 16 Bit is per se bad. jerry

Hi, as dithering is in a simple case only adding white noise to the least significant bits and as most ADCs do not exploit the complete dynamic range of 144dB with 24 Bit, the noise of the electronics already dithers you the signal (i.e. adds noise to the least significant bits). jerry

Sorry, but the explanation is not really correct. An example: Your loudest sound gives you 2 Volt For the sake of simplicity we use full scale for 2 Volt in the digital domain. I.e. 2V are represented as 65535 with 16 Bits. A quieter part gives you about -32dB. I.e. 0,05V (calculation: 20*log(2V/0,05V)=32,041dB) An even more quieter sound gives you about -72 dB. I.e. 0,0005V Then 0,05V is represented as 65535/2V*0,05V = 1638,375. As you can only represented natural numbers (no float or fix point) this can be represented as 1638 or 1639. The rounding error (or quantization error) is 0,375 or 0,625. 0,0005V is represented as 65535/2V*0,0005 = 16,38375. So it is represented as 16 or 17. Rounding error: 0,38375 or 0,61625 The rounding error for the quieter part is almost the same. You don't use 4Bits for -72dB (i.e. 0,0005V) or 11Bits for -32dB (i.e. 0,05V). You only have rounding errors which are not random. So you hear artifacts. Dither adds random noise. So you don't have artifacts. If you use 24Bit you have smaller rounding errors. jerry

This is the explanation i found so far. Bit depth is only about dynamic range if properly dithered. 24bit vs 16bit, the myth exploded! Here also some nice test files. You can here the 8 Bit audio file without dither which has artefacts and the other with dither where you only here more noise as the voice level decreases. So it is only about signal to noise ratio. Dynamic Range, Dithering and Noise Shaping Often there is a very misleading analogy to computer vision where more bit depth means better image quality. But this is not how audio bit depth works: Color depth - Wikipedia, the free encyclopedia jerry

Is the coax connector RCA or BNC? jerry

But this not a problem of PCM or 192kHz audio. You have to design your filters according to your oversampling rate. If you design your filter with an oversampling rate for example 16 for 48kHz in mind, it is obvious that an oversampling rate of for example 2 for 192 kHz does not work. jerry

Hello, i want to bring in the Naim DAC which is in my point of view a true multibit PCM DAC. It uses two PCM1704K. Naim DAC is an oversampling DAC. Thus it oversamples (by zero-stuffing) each input signal to 768kHz (or 705,6 kHz). IIR filter to remove the unwanted frequencies from oversampling. Feed the output to two PCM1704 (one for each two channels) Current to voltage conversion Analogue filter Step 1 and 2 are in the digital domain Step 4 and 5 are in the analogue domain. The DAC uses a custom made, discrete (no integrated chips) design. This is what chris mentions: A DAC is more than the single step of actually doing the digital to analogue conversion (step 3). You can buy integrated chips which do all the steps above in one chip. And this could sound different to another approach. Here is a diagram of the 1704. It includes only step 3 (it has two DACs because it has a complementary design) And here is a diagram of the PCM1791 where everything is integrated in one chip. (The DAC also provides an usb input. It is only for usb sticks. But with the usb port you can play audio files with 768kHz sampling frequency directly.) jerry

why do think it is upsampled? Are you referring to this site where in the comments someone asks/claims this? Miles Davis: Kind of Blue (Again) | AudioStream the difference between the two pictures on audiostream is that the authors diagram goes down to -144dB. Whereas the diagram of ronalde only goes down to -90dB. You can change it in the preferences of audacity. Than the diagram looks the same as the authors one. Thus no upsampling.

Hello, regarding measuring dynamic range of vinyl i found this video very interesting: jerry

Sometimes there is a claim that one can distinguish a 50 Ohm BNC connector and a 75 Ohm BNC connector based on the insulator. File:BNC 50 75 Ohm.jpg - Wikipedia, the free encyclopedia Is this correct or wrong? Wikipedia also says: "The 75 ohm types can sometimes be recognized by the reduced or absent dielectric in the mating ends but this is by no means reliable." jerry

Maybe i was not very clear. I mean AD converters with mic preamps which have SPDIF outputs. Like this Lavry Engineering "The AD11 includes built-in microphone preamplifiers..." "The USB output is available in parallel with both XLR and RCA AES/SPDIF digital audio outputs..." or this ULN-2 Forget about mic preamps. But pro gears which can output digital only have RCA digital audio outputs. Why? jerry

Can anyone explain why the so called "pro audio" hardware all have rca spdif connectors. I am looking for mic preamps and have yet not found any with bnc connectors. The only bnc connectors are for word clock in and out. Is it because this is only the consumer version and the standard dictates a rca connector? I have only seen BNC SPDIF in the home/consumer audio world. jerry

Dan, do you think it would be possible to not only copy the flacs onto the ios device but also to access a afp or samba share with flac player and play them? jerry

From a technical point i understand it now. From a personal point of view it is still frustrating! I think the problems lies in the fact that there is no real definition of "hires files". I am buying hires files not solely because they have a higher sampling frequency than 44.1kHz. I am buying hires files because i like the idea to get files which are exactly the same as the mastering engineer hears it - without any alteration. For a distribution by cd or sacd the files must be converted to the target format. But with hires files i had the hope to skip this transformation. In this special case: It was false advertised with 96kHz. So i thought i get the "original" pcm files. It was not clear that i "only" get dsd-derived files. I do not like the idea, as the the sacd is not a straight analogue-to-DSD transfer, to have a analog master - digitalize it to pcm 96kHz- convert this to dsd - convert this to back pcm 88.2kHz.