Tír Na HiFi

Listened sse2 1644 3.56+1.6 for about 3.5 hrs yesterday afternoon. I still think it's extremely good on my system, well-balanced. No listening fatigue.

Sligolad wrote:...
I pulled up 3.14avx on the same control and I was hearing music again.
...
Cheers, Pearse.

Agree, this is a very fine one as well. Personally, can exchange 3.27 and 3.14 without feeling left out by either of them.

These are the best MQn versions around at the mo.

Cheers

Aleg

let see if sbgk can solve everyone's problem hehehe

3.27 see is good. i like it too. its weight > 3.56 sse.
drum weight sounds more close to the real thing. this implies vocal/piano has more power than norm.

3.56 cuts a little weight. vocal power becomes perfect. piano ok. flow is amazing.

is it even physically possible to have full weight in drum, without affecting vocal and other instruments?

this is music presentation i am talking. seems like it is either one or another.

texture/bass/vibration/quantity can always improve.

music presentation is a personal preference, or not?

uploaded 3.72, adds a bit moore weight to the sound compared with 3.71

sbgk wrote:uploaded 3.72, adds a bit moore weight to the sound compared with 3.71

Comparing 3.14 avx to 3.72, there remains a characteristic difference.
3.14 sounds as if it has been recorded with a close miking technology. Direct sound, with all harmonics as if separate, a good shine and a very nicely timed decay.

3.72 sounds as if miked from a greater distance. Sound lost the shine sounding a bit flatter, the harmonics not as separate but more amalgamated into one. Details still OK, reverb also OK I believe (not really listened to that aspect, but not noticeably out of order).

I wonder if the 3.14 avx might have been with the 512 bit commands instead of 256 or the current 128?

Cheers

Aleg

Aleg wrote:

sbgk wrote:uploaded 3.72, adds a bit moore weight to the sound compared with 3.71

Comparing 3.14 avx to 3.72, there remains a characteristic difference.
3.14 sounds as if it has been recorded with a close miking technology. Direct sound, with all harmonics as if separate, a good shine and a very nicely timed decay.

3.72 sounds as if miked from a greater distance. Sound lost the shine sounding a bit flatter, the harmonics not as separate but more amalgamated into one. Details still OK, reverb also OK I believe (not really listened to that aspect, but not noticeably out of order).

I wonder if the 3.14 avx might have been with the 512 bit commands instead of 256 or the current 128?

Cheers

Aleg

I still manage to be without a pc
So I'm bugger all use
I don't think there's 512 instruction capability yet
Think the old good avx's were 256 instruction 10 ms period size and we were fine tuning with the 448 clockrate and using tasker
Think the 448 clock rate and 10 ms period size combo with tasker
Is responsible for most of those sq qualities you've mentioned above
512 period size never had that same magic before
Which is sort of strange as the page alignment theory sort of makes sense

nige2000 wrote:

Aleg wrote:

sbgk wrote:uploaded 3.72, adds a bit moore weight to the sound compared with 3.71

Comparing 3.14 avx to 3.72, there remains a characteristic difference.
3.14 sounds as if it has been recorded with a close miking technology. Direct sound, with all harmonics as if separate, a good shine and a very nicely timed decay.

3.72 sounds as if miked from a greater distance. Sound lost the shine sounding a bit flatter, the harmonics not as separate but more amalgamated into one. Details still OK, reverb also OK I believe (not really listened to that aspect, but not noticeably out of order).

I wonder if the 3.14 avx might have been with the 512 bit commands instead of 256 or the current 128?

Cheers

Aleg

I still manage to be without a pc
So I'm bugger all use
I don't think there's 512 instruction capability yet
Think the old good avx's were 256 instruction 10 ms period size and we were fine tuning with the 448 clockrate and using tasker
Think the 448 clock rate and 10 ms period size combo with tasker
Is responsible for most of those sq qualities you've mentioned above
512 period size never had that same magic before
Which is sort of strange as the page alignment theory sort of makes sense

Yep, you're right AVX-512 has been proposed but is being realised only from 2015.

I just keep wondering why there is such a big difference in the avx versions nowadays compared to the the 'older' ones like 3.14 and before. They all used to have this magic quality about them and the latest avx-s seem to have lost that extra quality.

What's keeping you from getting your/a PC running?

Cheers

Aleg

well ive two problems
one im busier than usual and the other ive broke all my motherboards trying to figure out a three clock replacement, i might have to pull my horns in and go with two
what really sickens me is ive already heard three working its just been unreliable :((
maybe ill park the project for a few wks clear the head

nige2000 wrote: ....
Think the 448 clock rate and 10 ms period size combo with tasker
Is responsible for most of those sq qualities you've mentioned above
512 period size never had that same magic before
Which is sort of strange as the page alignment theory sort of makes sense

ignoring rounding error, i actually like 480 period size as much as 448.
possible to try?

(yes 480 is odd. somehow it works well with my soundcard)

Aleg.

Are you still using the old 2.71 for HiRez?