MQN

[nige2000]

...
the gap between sse2 and avx used to be big enough to buy a new avx2 compatible pc
...

i remember there was also a gap between sse2 and sse4.1 (or 4.2 forgot which one).

sse2 gets bullied a lot... i used to dislike it until mqn3.3x

ya its about three times the gap between sse 2 and sse4

when sse4 was the best

[jrling]

3.56 is incredibly good.

But I had a couple of times when in mid-track it stopped outputting music altogether, although the DOS Window was still showing it running.

I appreciate that could be system specific but I have not had that with any previous versions and so wonder if the changes to 3.56 might be causing it?

Just read that for WaveIO users, the latest Thesyscon drivers v2.19 have been reported as having bugs which might have caused my issues and they are being reissued soon with fixes.

[Fujak]

3.56 is incredibly good.

But I had a couple of times when in mid-track it stopped outputting music altogether, although the DOS Window was still showing it running.

I appreciate that could be system specific but I have not had that with any previous versions and so wonder if the changes to 3.56 might be causing it?

Just read that for WaveIO users, the latest Thesyscon drivers v2.19 have been reported as having bugs which might have caused my issues and they are being reissued soon with fixes.

I am using the WaveIO (XMOS) (driver v 2.19) without any issues in all MQn versions.

[Sligolad]

I am using the latest XMOS Thesycon drivers v2.20 from Meitner with no issues, all sounding very good with 3.56sse.

[sbgk]

going back to first principles with MQn, still think page align memory reads reduce noise

have uploaded mqncontrol.exe 3.60 1024 and mqnplay1644.exe 24 bit 3.60 avx 128 1024

which illustrates this, I think, reads 4096 bytes at a time.

sounds very good, punchy, detailed.

[satshanti]

Wow! I've been abroad for about a month without access to my reference systems, so have missed a lot. I've spent some time these last few days catching up, and what amazing improvements have happened! Incredible...

How many times in this epic thread have I read Gordon's assertion that "this is it, this is as far as it can go, this is the final version", and then more came, and more, and more... at each of these "milestones" I thought it could not possibly get any better, and boy, was I wrong! :-)

Before I left, the 100000/100000 avx combo was still my favourite by a small margin, compared to the second best 2.78 avx, so after my return I started from there. At 2.82 avx it was getting really hard to decide which one I liked better. 2.85 and then 2.92 sse were so good, that I actually preferred these over my old reference. Still, sse has a different signature from avx, so I was really curious about the next steps. I also switched control to 1.6, which now sounded better than control 100000.

I tend to be in line with Aleg's experiences, so I tried 3.14 avx next. What a jump! Then on to 3.27 avx, possibly another slight improvement. I then switched to control 2.97 and that again made a huge difference. Next steps were testing the latest versions, 3.56 sse, 3.56 avx 256 and 3.57 avx 128. Again these were a step up in resolution and clarity, particularly 3.57. I then tried control 512 and 1024, and the latter was the winner for me. So the current new favourite is 3.57 avx with control 1024. I also went back to try various combinations with control 2.97 and play avx 3.27, as Aleg uses those, but on my system to my ears, my preference is clear.

But as great as it sounded at this point, that wasn't the end of the story. I upgraded to the latest MQnload (excellent work, taggert!) and included tasker being executed. I had tried that in the past with MQn.bat, but that never worked for me. This made a huge difference again, made the whole sound stage rock solid and holographic, phenomenal!

I also briefly compared it to Aleg's addition of setting the clock to 448. I'll have to revisit that, possibly experiment a bit more with finding my own optimal clock rate.

All in all I am tremendously impressed and really happy with the current versions. Thank you very much, Gordon!

Now I'm eager to try the latest 3.60 avx 1024 play/control versions you uploaded, maybe tonight...

[Karl]

Sorry,

the versions after 3.41 are very detailed but not well-balanced. A little bit too slim or fast.

After 30-60 minutes listening to the music makes me tired.

1.4 and 3.41 are my favourites.

Karl

[Aleg]

going back to first principles with MQn, still think page align memory reads reduce noise

have uploaded mqncontrol.exe 3.60 1024 and mqnplay1644.exe 24 bit 3.60 avx 128 1024

which illustrates this, I think, reads 4096 bytes at a time.

sounds very good, punchy, detailed.

3.60 avx 128 1024 is overly bass heavy.
Bass overpowers all other tonal ranges.
No balance.
Cannot even begin to compare to other avx versions because other tonal ranges are not clearly distinguishable to be able to compare to other avx versions.

Cheers

Aleg

[2channelaudio]

Sorry,

the versions after 3.41 are very detailed but not well-balanced. A little bit too slim or fast.

After 30-60 minutes listening to the music makes me tired.

1.4 and 3.41 are my favourites.

Karl

I tend to agree Karl....
MQN walks a very fine line between musicality, long period listening and detail retrieval.
My experience is more often than not MQN favours a slightly forward & bright presentation.
Sometimes this balance may not suit everyone's system.

In saying this MQN is easily the best rendering software out there. At least that I have heard.
Install jplay onto your system and compare.... It absolutely murders Jplay on a resolving system.

Personally I like 3.34sse2, its noticeably richer than 3.56sse2..
I also like 2.59... not sure which version I have..

I also enjoy playpcmwin...
Which although is not as accomplished as MQN in rendering detail, does provide a relaxing listen when you just wish to enjoy the tunes. In fact I'm using playpcmwin right now

Just my 2 cents. Don't shoot me.

To me it seems that some of the main contributors to feedback on this thread prefer a slightly forward and thinner sound... Just saying.
I think this may have something to do with music genre preferences and MQN system synergy.
I think more resolving solutions do better on less forward MQN versions.

[jesuscheung]

My experience is more often than not MQN favours a slightly forward & bright presentation.
....

Personally I like 3.34sse2, its noticeably richer than 3.56sse2..
I also like 2.59... not sure which version I have..

To me it seems that some of the main contributors to feedback on this thread prefer a slightly forward and thinner sound... ...

yes. 3.34sse2 has better bass than 3.56sse2. better bass=richer sound.

my favorite bass version is actually the ancient mqn.exe2.2

do you find 3.56 bright? if so, something else is causing it.

Sorry,

the versions after 3.41 are very detailed but not well-balanced. A little bit too slim or fast.

mqn.exe2.2 is the "slowest" version. you would like it. i like music slow too.
also why i like mqn.exe2.82. it is also very slow. i can hear exactly how piano key is pressed.

i still like 3.56 sse2 the best. best music flow. no excess. good micro