48K or 44.1K...?

My new laptop came with its audio set at 48000 Hz. My old desktop rig has the audio set to 44.1 Khz.

Now, I can exchange files between them, and everything seems to play along correctly, though I am of course aware of the notorious ‘semitone drop’ when something doesn’t match up.

Is there any reason to favour one over the other (apart from the mere fact that 48 K is ‘one higher’)…?

Noteperformer recommends 44.1, I see.

I’m not likely to do anything particularly sophisticated with audio mixing, and I can’t hear the difference between mp3 and raw audio most of the time.

Audiophiles may well give a different answer, but in practical terms, the answer is “no”. (I use 48000kHz on my M1 simply because that’s what it was set to by default.)

1 Like

Historically, 44.1 kHz was chosen as a sampling rate for the CD when it came out because it was the sample rate that allowed the Beethoven 9th Symphony SONY’s CEO wanted to have on a CD to fit in that CD (I don’t remember which one it is, but that would be quite easy to find, I guess). Later, more information could be put on a CD…
Or this ?
48kHz is a sampling rate that allows to filter out high frequencies (overlaps) with less damage to the audio signal, but hey, this is history. Pro audio work with 96kHz or 192kHz, and this is not what we’re doing with Dorico… so 44.1 or 48kHz are fine :wink:

I thought the Beethoven 9 recording determined the radius of the CD disk, rather than the sample rate.

I’ll probably set them both to 44.1 then.

FWIW, in the past, I used to work a lot with video, and for video (especially the old DV standard), 48 kHz was the way to go, so all my computers and outboard is set on that, a far as it’s adjustable.
Plus, my mixing desk has additional ADAT channels at 48…


Yes, you’re probably right :wink: Watkinson is a very reliable source, and his explanation with the video recorder is perfect. So the density of the “grooves” must be related to that piece of music.

44.1 kHz comes from the fact that we used to record digital sound on video recorders in the 1970s and this was a convenient frequency. It was taken over by CD because a lot of technology had already been developed around it and because it was at that time the lowest sampling rate that could deliverv a 20kHz audio bandwidth with reasonably inexpensive anit-aliasing filters.

More details are here if you really want them!

48kHz was preferred professionally because it allowed a wider audio bandwidth, and this was adopted for DVDs.

However, a more significant limitation of CDs is the bit depth of 16-bits.

The dimensions of the CD were chosen because Karajan was friendly with Akio Morita, the head of Sony, and encouraged him to be able to accommodate his recording of Beethoven IX without changing the record, as was always necessary with LPs.


Thanks for the precision for that 9th symphony, I didn’t remember it was Karajan’s :wink:

While we’re talking precision, I just want to mention that the interval between these two sample rates is 146.7 cents, so ¾ of a whole step. Thank you, carry on.


44 kHz / 16 bit is the standard for CDs

48 kHz / 24 bit is the standard for film and TV.

When compared side by side, no one can probably tell the difference in terms of “quality”, assuming the source material is the same. But “purists” love to debate this to death (a debate I really like to avoid).

When you are running multiple computer programs at the same time (DAW, Dorico, Pro Tools, Sibelius, etc.) I recommend to have the same bit rate for all programs to avoid possible issues with audio playing a half note too high or low.

When you are running multiple computer programs I can also recommend - highly recommend - using a professional grade audio card, the hardware that communicates between your computer and your speaker / headphones. Steinberg has fine audio cards, but the gold standard imo is RME in terms of user support, quality, low latency, etc.

1 Like

I’ve always found all the Steinberg audio interfaces I could use excellent, as far as audio quality is concerned. The build quality looks usually firm and solid. I still use an old UR824, still top quality for converters, and still very good for preamps. Signal coming from a pro preamp is incredibly open and transparent when entering these converters; and the integrated preamps, while not at the same level for clarity and dynamics, are good performing companions on less sensible channels.

The audio driver is still developed, after all these years. The RMEs of the same age may no longer have working drivers. When they work, the Steinberg drivers are solid workhorses; never had them crash or the like. A repeated issue I’ve had (and still have sometimes) is that the control software, dspMixFx on the Mac, may not recognize the audio interface, unless you restart it. A problem with that particular software, since the system always immediately recognizes the hardware.

dspMixFx is one of the worse software I’ve seen. It also succeeded in porting the Windows 95 user interface on the iPad. It can’t still make a difference between the scrolling arrows and the scrollbar. The window is not resizable, and it becomes a small thing when switching to the panoramic view. An inexplicably bad UI, that damages the first impression for a perspective user, and is probably part of the bad rep these beautiful devices sometimes have, without deserving it. Really a shame, since the general concept is fine, and the ease of use, compared to RME’s TotalMix, is admirable.

As for technical support, I had a problem with a broken power supply, and couldn’t get an answer from the Steinberg support for an incredibly long time; but the power supply being external and standard in the Yamaha catalogue, I could identify it and purchase it myself. This can be of great importance when depending on these devices and needing spare parts.


One thing to bear in mind in all this is CPU load. 44.1 @ 16 bit is a lot less stress than 96 @ 24, so I draft in the former and master in the latter. Also smaller audio files.