Page 744 of 804

Re: MQN

Posted: Fri Sep 11, 2015 2:13 pm
by sebna
If it is not Haswell (an by the looks of it it is not) it will rather not work unless something has changed in that requirement that I do not know about.

Nevertheless there is no harm in trying. There was a time when my AMD X2 CPU was working fine with Haswell versions ;)

Re: MQN

Posted: Fri Sep 11, 2015 11:45 pm
by 2channelaudio
sebna wrote:If it is not Haswell (an by the looks of it it is not) it will rather not work unless something has changed in that requirement that I do not know about.

Nevertheless there is no harm in trying. There was a time when my AMD X2 CPU was working fine with Haswell versions ;)
OK.... thanks

So does anyone know which is the highest revision which will operate?

Re: MQN

Posted: Sat Sep 12, 2015 11:51 pm
by stephen.sokolows
Hi guys are any of the long time users able to answer this, Gordon?

Re: MQN

Posted: Sun Sep 13, 2015 4:46 am
by iori
Ivy Bridge cpu should be able to run sse, avx and avx2 versions until those specific for haswell cpu.
approximate milestones : 'cause I'm too old to catch up with Gordon's work and I missed some downloads
20141129 - mqnplay1644.exe 24 bit 8.91 normal
20141129 - mqnplay1644.exe 24 bit 8.91 avx2
20141224 - mqnplay1644.exe 24 bit 9.06 normal cuinas (dac hard coded ver)
20141224 - mqnplay1644.exe 24 bit 9.06 avx2 cuinas (dac hard coded ver)

20150118 - MQn_r910gk_1644_ks_avx2_399Control_ (my best reference before setting up a haswell)

viewtopic.php?f=15&t=2418&start=6750 page 676
20150212 - (haswell cpu only, test)
20150324 - mqnplay.exe v14 long device name (haswell cpu only)

20150708 - mqnplay.exe haswell v17 win 10.exehaswellv17win10 (haswell cpu and win10)
20150708 - mqnplay.exe haswell v17 2012 r2.exehaswellv172012r2 (haswell cpu and ws2012r2)

20150724 - mqnplay.exe sse4 w 2012 r2 v29.exesse4w2012r2v29 (special ver)
20150724 - mqncontrol.exe sse4 v20.exesse4v20 (special ver)
20150724 - mqnloader.exe sse4 v20.exesse4v20 (special ver)
"uploaded sse4 versions of control/loader v19 and play v29, play is for win 2012 r2 or win 8.1"

Re: MQN

Posted: Sun Sep 13, 2015 11:44 pm
by sbgk
made some progress with the multitrack, just need to tidy it up, but the basics are working.

next will be hirez.

Re: MQN

Posted: Mon Sep 14, 2015 6:12 pm
by sebna
Sounds very good. Hope you can make hires in both 32 and 24 bit flavors

Sent from my HTC One using Tapatalk

Re: MQN

Posted: Tue Sep 15, 2015 7:36 pm
by sbgk
uploaded control v29, loader v29 and play v96 for multitrack play.

bit of a noise at the end and cpu goes up, but it settles down and mqnplay can then be stopped via task mgr.

playback starts after 5 secs.

Re: MQN

Posted: Tue Sep 15, 2015 8:22 pm
by janh
Listened to a playlist of three (short) files, and the whole list played fine.
When finished, cpu usage quickly rise to c.25%, and after a few seconds was some noise. Close mqnplay, and no problem.
Big thanks to sbgk for this!!
For sq compared to v93, more listening.
Cheers

Re: MQN

Posted: Tue Sep 15, 2015 8:37 pm
by sbgk
that's good news.

v96 is the same code as v93, but the way it's called from control changed which improved it a bit.

Re: MQN

Posted: Tue Sep 15, 2015 9:10 pm
by Aleg
sbgk wrote:that's good news.

v96 is the same code as v93, but the way it's called from control changed which improved it a bit.
When starting a playlist I get a few short crackles during the time mqncontrol hasn't yet closed. When mqncontrol terminates the crackles are gone. As it takes only about 5 sec for mqncontrol to disappear I only get 2 or 3 crackles.
Playlist works fine and indeed after the last track the cpu goes up to 25.8% until mqnplay is killed.

No comments on SQ yet (but no immediate and obvious differences of course)