MQN

Anything to do with computer audio, hardware, software etc.
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: MQN

Post by jesuscheung »

satshanti wrote: 4.39 > 4.40 > 4.41
think i agree...

4.39 > 4.40 > 4.41 > 4.44
---> treble seems more nervous
nige2000
Posts: 4253
Joined: Thu Feb 14, 2013 10:47 am
Location: meath

Re: MQN

Post by nige2000 »

sbgk wrote:
nige2000 wrote:Didn't get any testing done today but will do soon
Is it time for a 448 mqnplay for comparison?
4.44 is pretty sorted

it uses hardware prefetching which can't go over page boundaries, so the buffers need to be page aligned, currently 2048 bytes, 448 is 1792 bytes and would have irregular noise due to hw prefetch issue.
I understand the theory
It does sound logical
It's just that I've tested up to 4.33
And I'm still using 3.14/3.39 avx 2 for listening
Just wonder if its a factor
If we had a new 448 we could compare just to be sure 512 is the best fit
sd card player, modded soekris dac, class a lifepo4 amp or gb class a/b amp, diy open baffle speakers based on project audio mundorf trio 10's
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: MQN

Post by jesuscheung »

sbgk wrote:
4.44 is pretty sorted

it uses hardware prefetching which can't go over page boundaries, so the buffers need to be page aligned, currently 2048 bytes, 448 is 1792 bytes and would have irregular noise due to hw prefetch issue.
hmm... 448 * 4 = 1792

how about
448 * 16 = 7168
7168 = 7 * 1024
sbgk
Posts: 1950
Joined: Mon Oct 07, 2013 9:45 pm

Re: MQN

Post by sbgk »

jesuscheung wrote:
sbgk wrote:
4.44 is pretty sorted

it uses hardware prefetching which can't go over page boundaries, so the buffers need to be page aligned, currently 2048 bytes, 448 is 1792 bytes and would have irregular noise due to hw prefetch issue.
hmm... 448 * 4 = 1792

how about
448 * 16 = 7168
7168 = 7 * 1024
very good

first buffer 1792
second buffer 3584
third buffer 5376
so third buffer goes over page boundary
nige2000
Posts: 4253
Joined: Thu Feb 14, 2013 10:47 am
Location: meath

Re: MQN

Post by nige2000 »

4.42>>>4.44
didnt like 4.44

seems like your trying to remove noise and taking detail, hf, vibration with it
sd card player, modded soekris dac, class a lifepo4 amp or gb class a/b amp, diy open baffle speakers based on project audio mundorf trio 10's
sbgk
Posts: 1950
Joined: Mon Oct 07, 2013 9:45 pm

Re: MQN

Post by sbgk »

nige2000 wrote:4.42>>>4.44
didnt like 4.44

seems like your trying to remove noise and taking detail, hf, vibration with it
just trying things out, this may have been a dead end.

Maybe you'll like 4.47, have done an sse2 version as well.
nige2000
Posts: 4253
Joined: Thu Feb 14, 2013 10:47 am
Location: meath

Re: MQN

Post by nige2000 »

sbgk wrote:
nige2000 wrote:4.42>>>4.44
didnt like 4.44

seems like your trying to remove noise and taking detail, hf, vibration with it
just trying things out, this may have been a dead end.

Maybe you'll like 4.47, have done an sse2 version as well.
4.47 was improved, still short on micro micro detail,hf/LF, flow, macro detail is vgood very clear, however whispers in a crowd at a live recording sound like rustling of a bag in comparison to 3.39avx2

didnt test the others
sd card player, modded soekris dac, class a lifepo4 amp or gb class a/b amp, diy open baffle speakers based on project audio mundorf trio 10's
sbgk
Posts: 1950
Joined: Mon Oct 07, 2013 9:45 pm

Re: MQN

Post by sbgk »

nige2000 wrote:
sbgk wrote:
nige2000 wrote:4.42>>>4.44
didnt like 4.44

seems like your trying to remove noise and taking detail, hf, vibration with it
just trying things out, this may have been a dead end.

Maybe you'll like 4.47, have done an sse2 version as well.
4.47 was improved, still short on micro micro detail,hf/LF, flow, macro detail is vgood very clear, however whispers in a crowd at a live recording sound like rustling of a bag in comparison to 3.39avx2

didnt test the others
you can tell what type of chocolate bar people are eating, now that's hi fi.
Aleg
Posts: 1381
Joined: Thu Oct 10, 2013 8:26 pm

Re: MQN

Post by Aleg »

sbgk wrote:
nige2000 wrote:
sbgk wrote:
just trying things out, this may have been a dead end.

Maybe you'll like 4.47, have done an sse2 version as well.
4.47 was improved, still short on micro micro detail,hf/LF, flow, macro detail is vgood very clear, however whispers in a crowd at a live recording sound like rustling of a bag in comparison to 3.39avx2

didnt test the others
you can tell what type of chocolate bar people are eating, now that's hi fi.
If that's what is in the recording, it is the 'duty' of the player to let it be heard.
HDPLEX;picoPSU;ASUS Q87M;i7-4770T;PH SR7EHD;Server2012R2;Thesycon 2.24;
JCAT USB;Sonicweld DiverterHR2;Naim DC1;Chord Hugo;Morrow Audio MA6;Naim NAC-282,SuperCapDR;NAP-300;
AQ Cinnamon;GISO GB;Netgear Pro+XM21X;Cisco SG300;NAS-ZFS.
sbgk
Posts: 1950
Joined: Mon Oct 07, 2013 9:45 pm

Re: MQN

Post by sbgk »

Aleg wrote:
sbgk wrote:
nige2000 wrote:
just trying things out, this may have been a dead end.

Maybe you'll like 4.47, have done an sse2 version as well.

4.47 was improved, still short on micro micro detail,hf/LF, flow, macro detail is vgood very clear, however whispers in a crowd at a live recording sound like rustling of a bag in comparison to 3.39avx2

didnt test the others
you can tell what type of chocolate bar people are eating, now that's hi fi.
If that's what is in the recording, it is the 'duty' of the player to let it be heard.
yes, but it could have been a careless whisper.
Post Reply