henris94 Posted January 16, 2005 Report Share Posted January 16, 2005 Some recorders available can record up to 48KHZ, but only 16 bit. Others can record 24 bit, but only at 44.1 KHZ. Which would provide a better sound?Also, would using an XLR to miniplug adapter be very detrimental to sound (Of course, while somehow providing phantom power if necessary). Quote Link to comment Share on other sites More sharing options...
bobt Posted January 16, 2005 Report Share Posted January 16, 2005 Some recorders available can record up to 48KHZ, but only 16 bit. Others can record 24 bit, but only at 44.1 KHZ. Which would provide a better sound?Also, would using an XLR to miniplug adapter be very detrimental to sound (Of course, while somehow providing phantom power if necessary).←Since most people can't hear much above 18 khz, the sampling rate shouln't be as critical as the bit rate.The easiest answer is to get a good piece of music with lots of transients and lots of dynamic range, (LP copy of Loren Maazels 1812 Overture comes to mind) and make recordings on a multitude of settings, and see if you can truly hear the difference. With analogue you were able to be much more creative with recording as it was a smoother process, and more forgiving of levels etc. I still have many reel to reel tapes that still sound great, as well as cassettes done on a quality machine.Method does not matter as much, as the care you give in recordinf Quote Link to comment Share on other sites More sharing options...
dex Otaku Posted January 16, 2005 Report Share Posted January 16, 2005 The difference between 48kHz and 44.1kHz sampling rate is minimal at best. Also, if you're recording anything for later distribution, chances are -very- high that you'll end up resampling to 44.1kHz anyway.16 vs. 24-bit recording is another matter. If you're going to be making recordings that need to be post-processed in any great way, it makes sense to use higher than 16-bit resolution when you can. Converting it down to 16-bit later isn't much of a problem. Properly [carefully] processed 16-bit audio is more than adequate for most people's needs, but [especially] if you're going to be messing around with amplitude-related [volume] processing after, the higher the bit resolution is, the better things will work out.Caveat: Crappy A/D converters are crappy A/D converters, regardless of their resolution. Bad 24-bit recording is worse than good 16-bit recording. You tend to get what you pay for in this department, though it's nice to note that relatively high-quality equipment is available for not that much money these days. Quote Link to comment Share on other sites More sharing options...
722Forever Posted January 17, 2005 Report Share Posted January 17, 2005 I second what dex otaku says - Use 24Bit if you really need it, otherwise if you're just copying tracks then 16Bit will do.bobt - No offence, I think you've got a little confused: You're mixing up sampling rate with frequency response. 44.1kHz doesn't mean that all the audio is constantly at 44.1kHz - that would be pointless and unhearable. The samplerate is the amount of samples per second (rather like moving images have frames per second). So a second of sound will be made up of 44100 snapshots. Inside a a digital recording you can have any frequency you like. Just because human hearing tapers off at around 20kHz doesn't mean that we should be listening to things at or below 20kHz. They are two completely different things. They just happen to use Hz in their measurements! Quote Link to comment Share on other sites More sharing options...
corien Posted January 17, 2005 Report Share Posted January 17, 2005 Inside a a digital recording you can have any frequency you like.←Hmmm... no offence, but I have some doubts regarding your statement... I'm not an expert about digital audio, but I am a physicist. Mathematically speaking, the max frequency content (found in the Fourier transform) of any sampled signal *is* actually limited by the signal sampling rate. Moreover, Hz is *properly* used just because we are speaking about samples per seconds: their physical unit is 1/sec, i.e. Hz.Sampling TheoremNyquist FrequencyPlease let me know if I have misunderstood you. Quote Link to comment Share on other sites More sharing options...
dex Otaku Posted January 17, 2005 Report Share Posted January 17, 2005 To further corien's bit, there:According to the Nyquist theorem, the maximum recordable frequency of a given analogue to digital converter [and the maximum playable frequency of a D/A converter] is fs/2, or the sampling frequency divided by half in order to ensure a full alternation is made at that frequency. For a sampling rate of 44.1kHz the highest reproducable frequency is 22.05kHz.Further, as the recorded frequency approaches the Nyquist freq., distortion [harmonic and otherwise] increases as the sampling resolution is not actually enough to reproduce actual sine waves. This is part of why oversampling and now delta-sigma converters are used, as well as various and sundry methods of filtering and noise-shaping. Quote Link to comment Share on other sites More sharing options...
bobt Posted January 17, 2005 Report Share Posted January 17, 2005 To further corien's bit, there:According to the Nyquist theorem, the maximum recordable frequency of a given analogue to digital converter [and the maximum playable frequency of a D/A converter] is fs/2, or the sampling frequency divided by half in order to ensure a full alternation is made at that frequency. For a sampling rate of 44.1kHz the highest reproducable frequency is 22.05kHz.Further, as the recorded frequency approaches the Nyquist freq., distortion [harmonic and otherwise] increases as the sampling resolution is not actually enough to reproduce actual sine waves. This is part of why oversampling and now delta-sigma converters are used, as well as various and sundry methods of filtering and noise-shaping.←Hi Dex Otaku,Thanks for the correction, I have been recording for over 40 years in all medias. Again it is the quality you take. It's uderstood that frequency responce is 1/2 sampling rate, anyone who has taken an interest in the medium knows that. Also digital is not as forgiving on higher recording levels.And again, most people cannot hear much above 18 khz, sor a proper 44.1 recording should not miss much content Quote Link to comment Share on other sites More sharing options...
722Forever Posted January 18, 2005 Report Share Posted January 18, 2005 Ha ha! None taken. What an interesting thread.I'll check out those links Corien. I knew I was probably tempting fate by saying 'any frequency'...! I was just pointing out the differences between tonal and sampling frequency. Surely a higher sampling rate would theoretically improve the sound reproduction as well as frequency range - simply by having more samples per second and therefore a more detailed sound? bobt - I bow to anyone who has come through the age of 1/4 inch reel-to-reel. You are a true soundie. Quote Link to comment Share on other sites More sharing options...
streaml1ne Posted January 18, 2005 Report Share Posted January 18, 2005 Surely a higher sampling rate would theoretically improve the sound reproduction as well as frequency range - simply by having more samples per second and therefore a more detailed sound? ←This is simple calculus. The more samples under the curve the better (disregarding crappy hardware) you can represent that curve. Now if you can actually hear a difference is another story. Solve both problems, 96kHz/24-bit Quote Link to comment Share on other sites More sharing options...
henris94 Posted January 19, 2005 Author Report Share Posted January 19, 2005 Thanks for all the info, people. This is a great forum. Basically, people I have talked to all said the same thing: The more info you have to work with in the beginning, the better. Ideally 24 bits at 192 would be perfect. Wish my wallet would agree! Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.