Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion MQA - Related technology, products and music MQA listening impressions (without a MQA decoder)

There have been calls for a new thread so I have started this one. First let me welcome the appearance, at long last, of some publicly available MQA tracks.

I have listened to all the 2L MQA tracks from their free download site. My impression is that they are good but not outstanding. The Finzi, in particular, did not seem so good through my 8kSEs. I have some CDs that sound better. I did not compare them with other iterations of the same recordings. That can wait until we have the firmware to unpack the files to reveal their full potential.

But many thanks to 2L for being pioneers.

To encourage others to listen to these free, ground-breaking MQA tracks I offer the link to them here: free MQA tracks

I have enjoyed the Kate Rusby tracks very much. They sound 'Natural' and very well recorded and produced without being 'Hot' either.
Roon, MS200, G61R Analog Trifield.

With the 'Allegro' track from 2L I have a 192/24 comparison (I know this is down sampled) and I kept settling on the MQA version. The Violin sounded sweeter perhaps.
This could be expectation bias but the fact that they are in the same league is remarkable.

I don't think MQA will save anything that was badly treated in production but I am keen to hear more.

Just thoughts, Chris

I downloaded the 2 channel versions of MAGNIFICAT 4.

I found that the (un-decoded) MQA file at 44k was less good than the 24bit/96kHz PCM file when played back analogue out from my 808v5/ID41.

This is most probably to be expected as, from what has been said, I believe the un-decoded MQA file is only positioned as better than standard CD quality not better than full 24bit/96kHz PCM and higher.

Clearly I was not able to unfold the MQA to play at a higher resolution and the 2L site does not provide a corresponding CD quality file to compare against the MQA version.

The 24bit/192kHz file sounded identical to the native 96kHz file when down converted by my MC200 to 96kHz before sending on to the 808v5/ID41.

John

Hi Fitcaz,

I tried the 2nd demo from the list, the Violin concerto, and was surprised: I clearly found the MQA version better than the 96/24 version. The imaging was much clearer with pinpoint accurate placing of instruments and most notably the violin itself. Now of course the 44.1 MQA version was upsampled with my filter to 88.2, but I can not imagine that this made the difference.

It sounds like the other versions are from a "normal" master and not taken from the MQA master. I can ask, why not? It is clearly an advantage to use the best possible master when you want to distribute your music. Would licensing prohibit this?

My thoughts on the 2L recordings is that the MQA FLAC file is slightly brighter in tone than the 96 or 192 versions and I found them easier to listen for longer.

According to the 2L website all (but 2) files are from DXD masters so I am not sure what you are referring to "normal" and "MQA" masters.

Happy listening

The MQA files seem to me to sound better with HS Output set to N.

The soundstage opens up. Some of the treble harshness -eg the violins in the Mozart -seem less strident.

Can this be the case or is it me listening too late at night?

.???????????

Hi,

With HS=Y the 861 outputs x2 sample rate to the DSP's.

With HS=N the 861 outputs x1 sample rate to the DSP's which the 5200's the upsample to x2.

If you've got the latest firmware in both I'd not expect much difference.

That said, I prefer to have the 861 upsample (as per M recommendations, to upsample as earlier as possible in the audio chain).

I'll try it myself at the weekend, when I can have a proper listen.

Regards,
Carl

I would describe HS=Y in the pre/pro as putting a spotlight on the musicians. I'm very used to HS=Y sound. Letting the Speakers upsample turns the spotlight off while musicians playing. It's subtle and could warrant more listening.

Different hardware ie 818v3 without MQA could be of different audiocharacter from my 861v8 since they are of different hardware and software versions and from the post-MQA era vs pre-MQA era.

Edit: OMG I think I just had a moment with spirit on the "Kyrie" 2L track @volume75.

Is the HS=y in 861 destroying the MQA?

Infact, as upsampling happens internally, is 861 destroying MQA anyway.

I know that MQA is not being decoded, but how much of the extra MQA goodness is being diluted by 861 processing?

May be worth temporarily bypassibg 861 and going direct to speaker if yiu had another way of playing the source material.

Bob described (un-decoded) MQA files a "pre-apodized", so I'm guessing that they are "up-sampling-proof". It's a question that we need to have answered though.

I cannot perceive any detrimental effects from upsampling - to my ears they sound better upsampled to DXD. To be honest it is quite difficult to tell the difference between undecoded MQA upsampled to DXD and the original DXD master, based on the 2L samples on the test bench.

Carl, I wasn't aware that the speakers upsample if the 861 is set to HS Output N? I assumed N meant N. So the differences I heard is even more counter-intuitive.

After further listening this evening (with usual caveats, including my cloth ears):

There is a difference between where the up sampling takes place with a 'normal' CD file played from Sooloos. It is clearly better done in the 861. A crude blind testing confirms this.

With the MQA files from 2L the reverse is clearly the case. To me it is obvious on the blagutten track (listen to the bass for example which is more defined and crisp). It is less clear on the Finzi track - it shows up most on larger scale music such as the Arnesan and the Mozart.

With HS set to Y the soundstage seems flattened, it is more difficult to follow background instruments. When set to N the soundstage expands and the music becomes more enjoyable.

I don't think I'm imaging this and again I tried to blind test. The difference is not always immediately obvious and I confess to calling the Finzi wrong a few times. But in my experience, the difference is real.