Patience, grasshopper.
- ...
- 8 posts total
Most recordings made in the last 30 - 40 years have been recorded digitally, that means no tape used. It was recorded to a hard drive or some other type of digital storage. I’d guess that early on, when digital storage was more expensive, recordings were mostly made at a sampling rate of 44.1khz or 48khz and a bit depth of 16. CD quality, in other words Now that digital storage is cheaper and more compact, and streaming bandwidth is wider, many recordings are made at higher sampling rates and bit depths, 96/24 192/24. Sometimes DSD recordings are made. If tape was used to record a new album, that will usually be noted on the sleeve, the booklet (for digital discs) or in the advertising because recording to tape is not the norm anymore. It is very hard to find out at exactly what sample rate and bit depth a digital recording was made. For some reason the record labels like to keep that information secret. Download sellers like HDtracks sell downloads of the same recording at different sample rates and bit depths, but, in the past, when they have been asked at what sample rate/bit depth the recording was made, they have said they don’t know, they just get the files from the record label without that information. |
@seventies I've been a recording engineer for nearly 3 decades. I can tell you what I know based on having worked in some of the biggest name studios in the industry. I've never seen anyone use DXD. Though I do know that Channel Classics is one studio that intended to but actually work in DSD https://www.channelclassics.com/general-info/about-channel-classics/ They are in any case an outlier. The standard in the industry these days is quickly becoming 24Bit/96K where as it's been 44.1K for music production and 48K for movie / TV audio production for the last 20+ years. It's common practice for mastering engineers to upsample to 96K and work there before converting to whatever res and format their customer desires. If the intention is to release a 'high res' remastered recording the most common practice is to upsample material recorded at 44.1K or 48K to 96K and mix / remaster there and output at 96K. The delivery container for that could be PCM, DSD, MQA or converted back to analog for vinyl. Nyquist’s Theorem states that we only need 44.1K to capture the highest frequencies for the human ear. While that's true it hasn't prevented exploration into the higher sampling rates. The question is whether the differences that can be heard there are natural or byproducts, namely distortion. I'm not sure I've answered your question but I hope this helps. |
Glissando and others, Thanks for your responses. Per part four of Rushton Paul's newly released article in 'Positive Feedback', several studios now record in DXD or DSD128. Whereas arguments for DXD vs PCM devolve on 'what you hear', I believe the resolution obtained with DXC and DSD256 is audibly superior to that obtained with 'lesser' iterations of the PCM and DSD formats. I believe also that the differences between PCM and DSD diminish to near-inaudibility at the highest resolution available in those formats. Which brings me back to file size, ie bit depth times sampling rate. Though one might argue that frequency and dynamic range, not digital sample size are what's important, I believe it unarguable that greater bandwidth (translate file size per duration of a 'song') offers greater potential resolution, and I emphasize the word 'potential'. To my elderly ears, particularly listening to violins or soprano voices, that potential is realizable. Further comments would of course be much appreciated. |
- 8 posts total