record vinyl to laptop

There has been an ABX test study conducted by the Audio Engineering Society into whether “CD quality” damages high quality audio. AES E-Library » Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback

Steve, first of all I am well aware of the scientific view on the better than 16/44.1 CD standard you bring up.

For that reason I am prepared for the ridicule and scoffing I will undoubtedly be exposed to when I say there is no doubt whatsoever in my mind that on my system I obtain audibly significantly better transfers when conducted at 24/96.

Separately I am also fairly certain that the declick software I now use provides materially better results when fed with 24/96 source material.

In each case I have no clue whether it is the increased sample rate, bit depth or some combination thereof that provides the improvement I am certain exists. And frankly with the benefit of a fairly new 8tb RAID array I don’t face the space constraints that make further investigation imperative; I just need to get on with the task at hand of getting a good transfer on another 1,500 or so 12" singles :open_mouth:

That is quite possible. A/D converters are a complex combination of analogue and digital technology. In my own experience I’ve found that particular sound cards tend to have a “sweet spot” - certain settings at which they perform best. The “sweet spot” is not necessarily their maximum settings (and for budget equipment it is often not the maximum settings). For recording in formats greater than 16 bit 44.1 kHz I would always recommend that users test their equipment critically to find what produces the best result, but also caution that the placebo effect is a very real and surprisingly strong effect (clinically proven), so double blind testing would be the ideal.

Again, that is quite possible, depending on the algorithm used to detect clicks.
It is also important to distinguish between “recording / playing” and “processing”. For processing there are definite advantages to using 32 bit float format - which is why it is the default format in Audacity.

As in the second of the articles that I linked to, there is no harm at all in using high bit depths, other than disk space, which as you say, is cheap these days. I’d caution against using sample rates greater than 96 kHz (for audio) for the reasons given in that same article. I’m not aware of any drawbacks to using 96 kHz, other than disk space, provided that the equipment works well at that rate.


Well there you go; my air or respectability is temporarily - if probably undeservedly - restored.

I had been under the firm impression that the Nyqvuist theory implied all meaningful content could be accurately described by 16/44.1, and that excluding DAW processing the only possible rational reason for using 24 bit was to avoid clipping the ADC if input levels were too hot.

Thanks Steve, from what I have seen on the interwebs I doubt this is widely understood.

24 bit audio can be clipped just as 16 bit can be. 16 and 24 bit audio are “integer” formats and both will clip off anything over 0 dB.
“Float” formats are different. They can go way over 0 dB without clipping. (The explanation is rather complicated but is available on Wikipedia if you’re interested).

The main advantage of 24 bit for recording is that it has a much wider dynamic range, so, because the noise floor is so much lower, the recording can be safely made at a much lower level without losing quality for quiet sounds. We generally recommend setting the recording level so that the maximum peaks are around -6 dB (6dB headroom), but for 24 bit recording it is common to allow at least 20 dB headroom, virtually eliminating the chance of an unexpected peak clipping.

According to the AES study, 16 bit, 44.1 kHz can accurately describe “all meaningful content” - that is, that except at very high listening levels, listeners cannot distinguish the difference between this format and higher formats. However, that is only looking at the digital format and is not looking at the “non-ideal” (real world) deal performance of specific A/D, D/A converters which may be optimised to perform better at their higher settings than they do at lower settings.

Well that certainly wasn’t my understanding based largely on this chart here: Headroom (audio signal processing) - Wikipedia

However, on a re-reading the whole blurb now maybe the headroom chart isn’t as unambiguous as I originally took it to be.

In any case I always keep a healthy margin of safety on my transfers, and presumably there is no danger or possible degradation in peaking around -6db using 24bit?

BTW, when you say it is normal to allow for 20db of headroom with 24bit, do you mean peak or RMS?


thanks everyone for the replies, i read where if i can record in stereo then the mic plug will work as a line in, all i had to do was to turn the mic booster down and switch off any effects from the mic menu. i tried this and it worked perfectly. thanks again to all who answered.