Rega DAC-R Filters

[beadlesabout]

Dear current and previous Rega DAC-R owners.

I am having difficulty settling on a filter and it's actually a bit annoying. It's annoying like valve swapping because you think you've settled and then you discover one filter sounds better on one album and another filter on another album. These days I prefer less things to tweak so I'm not tempted to muck about with things.

I'm curious, what filter have you settled with and why? I am using the 48k modes only, so:

Filter settings 1–3
(low sample rates 32/44.1/48K)
1. Linear phase half-band filter
2. Minimum phase half-band filter
3. Minimum phase apodising filter

It doesn't seem to be the case that they are progressively more rolled off in the treble from 1-3 because with filter 3 for example, some tracks sound more rolled off than 1 but others sound brighter.

I'd like to understand what they're actually doing. I am reminded of the Quad 33 manual which has several pages of graphs and technical explanations. No such luck in the Rega manual!

[Lawrence001]

Can't speak for the Rega DAC but I had an MDac and after playing around with the various filters I realised it sounded horrible and synthetic with all of them and switched over to a cheap Chinese dac which sounded much nicer

Sent from my HRY-LX1 using Tapatalk

[Filterlab]

The actual answer is quite confusing unless you’re into music production, and it’s easier to demonstrate with a single transient noise, like a snare hit for example.

Some DACs have variable filters, most don’t. On the M-DAC I preferred ‘Slow’ which doesn’t make it slow sounding, it refers on that DAC to the speed of the upper frequencies roll off. I do agree with the above though, the M-DAC ain’t much good.

The best thing to do is play a very familiar track on repeat at a set volume, and listen to it each time with a different filter. If you can’t hear any difference then worry not, if you can hear a difference (normally VERY subtle), then make a mental or written note of how each one alters the presentation of the track. Switch to your favourite of the three, or refer to your notes and select one that you’d imagine matches best for the genre of music you happen to be playing.

I say that as unplugged jazz is very different to liquid trance (for example) in terms of how the musical signal ‘uses’ the audio equipment upon replay. Your mileage may vary with each style.

If it doesn’t make any difference to your ears then switch the DAC off, unplug it for a minute or two, plug it back in and leave it at the default setting.

Always use your ears as final guidance.

[Barry]

My Sony CDP 930 has a switchable filter. Presumably it is a low-pass filter rolling off above 22kHz, with the optional switchable settings adjusting the rate of roll off. The instruction do describe the audible effect of each setting, but I have not noticed much difference. The problem is you cannot switch between them and do an instantaneous comparison - there is a something like a 10ms silence between settings. In the end I just left it to the Sony-recommended setting.

[Macca]

There will be a difference in the frequency of roll off but unless the filter is very odd or your hearing is superhuman it won't be audible. Can you really hear frequencies up around 22Khz?

Try to tell which filter you have switched in without knowing beforehand (select one blind by using the remote or get someone else to do it), the perceived differences will vanish like the morning dew.

It's just something they include because they can (already built in to the DAC chip) and because a lot of nonsense is talked about the audibility of pre and post ringing.

[Lawrence001]

Ringing and phase effects are I think the main reason the MDac has so many filters.

I was rather dismissive above, but I actively used filters for a long time, first on various Sony CD and SACD players of the 90s (I suspect those filters came in around the advent of SACD and supertweeters when everyone was raving about the impact of inaudible frequency response, a matter still not totally dismissed).

I slagged off the MDac but I was reasonably happy with it in a headphone only system for a while and landed on the slow roll off filter for that use. Similarly when using my Breeze Audio Sabre DAC now I use the slow filter rather than the fast one.

My advice is that quick A/B demos of such subtle changes are not helpful. I find living with the filter for a week or so, listening at various times of day and when in different moods is what you need to decide if you enjoy the overall presentation more or less. What you have to do is relax, enjoy the music and not keep trying to spot subtle differences in tracks you know very well (although if you do happen to spot them it's still useful information). Therein lies the path to audiophilia nervosa.

Sent from my HRY-LX1 using Tapatalk

[Barry]

There will be a difference in the frequency of roll off but unless the filter is very odd or your hearing is superhuman it won't be audible. Can you really hear frequencies up around 22Khz?

Try to tell which filter you have switched in without knowing beforehand (select one blind by using the remote or get someone else to do it), the perceived differences will vanish like the morning dew.

It's just something they include because they can (already built in to the DAC chip) and because a lot of nonsense is talked about the audibility of pre and post ringing.

Depending on the filter characteristic chosen, the steeper the rate of roll-off, the greater the signal group delay for frequencies close to the roll-off frequency. It is just possible that such phase errors might be audible but I doubt it. I believe most CDPs use a Bessel filter on the output, which have a 'maximally flat' group delay response, precisely for that reason

[Barry]

My advice is that quick A/B demos of such subtle changes are not helpful. I find living with the filter for a week or so, listening at various times of day and when in different moods is what you need to decide if you enjoy the overall presentation more or less. What you have to do is relax, enjoy the music and not keep trying to spot subtle differences in tracks you know very well (although if you do happen to spot them it's still useful information). Therein lies the path to audiophilia nervosa.

That is my approach to assessing any changes I make to my system. Listen to things for a week or two, then revert back to the arrangement before the change was made. That way any difference noticed will either be a genuine improvement (or not), or simply only a change with no improvement.

I should also add that if any changes (improvements or otherwise) are important to me, I will hear them - they just 'pop up', regardless of whether I am listening for them or not.

[Lawrence001]

Good point if you don't notice any differences then there probably aren't any

If you listen intently in an A/B test and consciously listen for small changes then your mind will make some up if there aren't any. If you have to keep going back and forth to make sure then that's pretty definite! This is the basis of double blind testing. Since I don't have the time or inclination (or support!) for such things I just do the extended listening comparison.

Sent from my HRY-LX1 using Tapatalk