IPB

Welcome Guest ( Log In | Register )

8 Pages V  « < 2 3 4 5 6 > »   
Reply to this topicStart new topic
FLAC 1.3.0 has been released, the most popular lossless audio codec
testyou
post Oct 14 2013, 21:36
Post #76





Group: Members
Posts: 99
Joined: 24-September 10
Member No.: 84113



I'm not sure what you mean Wolf.

--ignorelength doesn't work?

This post has been edited by testyou: Oct 14 2013, 22:00
Go to the top of the page
+Quote Post
Werewolf6851
post Oct 15 2013, 01:01
Post #77





Group: Members
Posts: 10
Joined: 27-September 09
Member No.: 73521



QUOTE (testyou @ Oct 14 2013, 21:36) *
I'm not sure what you mean Wolf.

--ignorelength doesn't work?


idk, using ubuntu 12.04 so flac 1.3.1 has not been down ported to my system yet.

Wolf
Go to the top of the page
+Quote Post
eahm
post Oct 15 2013, 08:09
Post #78





Group: Members
Posts: 1063
Joined: 11-February 12
Member No.: 97076



QUOTE (Werewolf6851 @ Oct 14 2013, 17:01) *
idk, using ubuntu 12.04 so flac 1.3.1 has not been down ported to my system yet.

Wolf

1.3.1 is NOT out yet, read the comments back.


--------------------
/lwAsIimz
Go to the top of the page
+Quote Post
dfkt
post May 7 2014, 15:14
Post #79





Group: Members
Posts: 16
Joined: 16-January 07
Member No.: 39719



I'm late to the party, but anyways... didn't find any newer info about FLAC 1.3.0 builds for Windows, so I tested all the ones mentioned here.

CODE
system: i7-3770, 16gb ram, win8pro-x64
test method: flac.exe -8 input.flac -o output.flac
_________________________________________

input: bytes file MD5 fb2k verification MD5

flac 1.2.1, 3h22m, 44.1kHz/16bit/2ch 1,708,858,410 1092139DA14A43C016F614A1A070598C FC2B730F57920D49D4E41604FE14A623
_________________________________________

results: time ratio bytes file MD5 fb2k verification MD5 flac.exe MD5

1.3.0-case-x32 2:47 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 4ADD826644D303B57D220F866C67292B
1.3.0-case-x32-fb2k 2:47 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 4ADD826644D303B57D220F866C67292B
1.3.0-flacfrontend21-x32 2:49 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 8C4AE54CE1B243904CEAB30575534FB3
1.3.0-john33-x32 2:52 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 C31C314E8C1E8DC62EE0CDBE048C6112
1.3.0-john33-x64 2:38 0.998 1,705,098,763 49D7D89FAD8335F339915F7FED0C1265 FC2B730F57920D49D4E41604FE14A623 29546F37C3EB364155FADD43EC570DED
1.3.0-lamedude-sse1-x32 2:53 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 F08E20ADF2042919CA3BF8305E36CFF6
1.3.0-lamedude-sse2-x32 2:47 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 6C52EEA53F27CFEC697632194AA4D5D5
1.3.0-lamedude-x64 2:58 0.998 1,705,098,763 49D7D89FAD8335F339915F7FED0C1265 FC2B730F57920D49D4E41604FE14A623 174360529CEBF9371B41019D46F1FDC4
1.3.0-lrn-mingw32-x32 3:18 0.998 1,705,098,763 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 E87DC0158474E662637886797F36C0DA
1.3.0-lrn-mingw32-o2-x32 2:52 0.998 1,705,098,763 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 E0D5DAE8EF9BA49D0822901FE5D114A6
1.3.0-lrn-mingw32-o2-core2-x32 2:45 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 4D650372DBDB65DCC76E34DBAA42FB8F
1.3.0-wcd-x32 4:00 0.998 1,705,099,033 9BE9ABC5283909DC0F08A822F0B8FF46 FC2B730F57920D49D4E41604FE14A623 E5639F8BABD64836B85B4DB119279A7B

1.2.1-official-x32 3:12 0.998 1,705,099,033 2BA84F8B1B5514BE4BED536F341875C7 FC2B730F57920D49D4E41604FE14A623 16344C45643E41544A5C1C926A109C9F
1.2.1b-official-x32 3:15 0.998 1,705,099,033 2BA84F8B1B5514BE4BED536F341875C7 FC2B730F57920D49D4E41604FE14A623 B8396749F374C9F48CBE3B16C5333D49
Go to the top of the page
+Quote Post
The Sheep of DEA...
post Jun 29 2014, 20:57
Post #80





Group: Developer
Posts: 175
Joined: 16-April 06
Member No.: 29596



Bug? Using the -p option results in much larger files than without p (up to 2% larger!). This doesn't seem to happen when I use a blocksize of 512. I haven't tested it on many input files yet, so it might be limited to my input files -- more testing is needed.


--------------------
Copy Restriction, Annulment, & Protection = C.R.A.P. -Supacon
Go to the top of the page
+Quote Post
Maurits
post Jun 30 2014, 09:44
Post #81





Group: Members
Posts: 380
Joined: 30-September 05
From: London, Europe
Member No.: 24805



QUOTE (The Sheep of DEATH @ Jun 29 2014, 20:57) *
Bug? Using the -p option results in much larger files than without p (up to 2% larger!). This doesn't seem to happen when I use a blocksize of 512. I haven't tested it on many input files yet, so it might be limited to my input files -- more testing is needed.

Are you seeing this in 1.3.0 but not in older versions?
Go to the top of the page
+Quote Post
Dark_wizzie
post Jun 30 2014, 15:23
Post #82





Group: Members
Posts: 58
Joined: 5-October 13
Member No.: 110659



Noob question:
So if FLAC is itself updated, do these changes affect old FLAC files or only FLAC files made with the new version?
Go to the top of the page
+Quote Post
probedb
post Jun 30 2014, 15:32
Post #83





Group: Members
Posts: 1214
Joined: 6-September 04
Member No.: 16817



QUOTE (Dark_wizzie @ Jun 30 2014, 15:23) *
Noob question:
So if FLAC is itself updated, do these changes affect old FLAC files or only FLAC files made with the new version?


Only those created with the new version or older files you reencode with the new version.

How do you think they would affect older files unless you did something to them?
Go to the top of the page
+Quote Post
Dark_wizzie
post Jun 30 2014, 23:24
Post #84





Group: Members
Posts: 58
Joined: 5-October 13
Member No.: 110659



QUOTE (probedb @ Jun 30 2014, 07:32) *
QUOTE (Dark_wizzie @ Jun 30 2014, 15:23) *
Noob question:
So if FLAC is itself updated, do these changes affect old FLAC files or only FLAC files made with the new version?


Only those created with the new version or older files you reencode with the new version.

How do you think they would affect older files unless you did something to them?

No idea. Hence the "noob question".
Go to the top of the page
+Quote Post
GrieverV
post Jul 12 2014, 05:05
Post #85





Group: Members
Posts: 4
Joined: 12-July 14
Member No.: 116659



FLAC git 3194829 win32 and x64 compiles:
win32, x64

Built using MSVC 2013 Update 3 RC using the Windows XP toolset (v120_xp)

I decided to use the latest FLAC git instead of 1.3.0 due to various Windows/MSVC support improvements.
You can browse FLAC's git repository and view all the changes since 1.3.0 here: https://git.xiph.org/?p=flac.git;a=summary
Go to the top of the page
+Quote Post
Wombat
post Jul 12 2014, 15:11
Post #86





Group: Members
Posts: 1020
Joined: 7-October 01
Member No.: 235



Holy Moly! Just browsed the flac-dev Archives and seems like our fellow member "lvqcl" feels in love with the flac code smile.gif He is running amok with improvements. Impressive!
Go to the top of the page
+Quote Post
Emre
post Jul 12 2014, 20:01
Post #87





Group: Members
Posts: 18
Joined: 11-January 08
From: Turkey
Member No.: 50385



quick test
comp. level 8
Intel Q9400 o/c (3.2Ghz)

44100Hz 16Bit sample

~ 54.24x > flac 1.3.0 32bit (john33 icl13)
~ 54.24x > flac 1.3.0 64bit (john33 icl13)
~ 72.85x > flac (latest git) 32bit (GrieverV msvc2013u3rc)
~ 84.39x > flac (latest git) 64bit (GrieverV msvc2013u3rc)

48000Hz 24Bit sample

~ 27.85x > flac 1.3.0 32bit (john33 icl13)
~ 44.65x > flac 1.3.0 64bit (john33 icl13)
~ 50.22x > flac (latest git) 32bit (GrieverV msvc2013u3rc)
~ 52.43x > flac (latest git) 64bit (GrieverV msvc2013u3rc)

nice improvements.
Go to the top of the page
+Quote Post
Sparktank
post Jul 12 2014, 20:20
Post #88





Group: Members
Posts: 24
Joined: 31-July 12
Member No.: 101901



QUOTE (Emre @ Jul 12 2014, 12:01) *
nice improvements.


Indeed!

I only just downloaded them. I have not had the time to do any tests or setup my converters to use the git builds.

Nice to see development is continuing. So many nice improvements just looking at the git logs.


--------------------
I like to use "[i]HD audio[/i]" in PaulStretch. "HD audio", lol.
Go to the top of the page
+Quote Post
GrieverV
post Jul 13 2014, 15:10
Post #89





Group: Members
Posts: 4
Joined: 12-July 14
Member No.: 116659



Some basic benchmarks using PowerShell Measure-Command:

CODE
Command:        flac.exe -8

CPU:            AMD A6-5400K@4.0GHz
CD Image:       44.1/16/Stereo, 57:48

GrieverV    git 3194829    64bit: ~32.98s
GrieverV    git 3194829    32bit: ~44.70s
john33        1.3.0        64bit: ~63.30s
john33        1.3.0        32bit: ~64.33s
Case          1.3.0        32bit: ~66.13s


I hope they start thinking about tagging a release soon because these improvements are definitely worth it.

This post has been edited by GrieverV: Jul 13 2014, 15:41
Go to the top of the page
+Quote Post
polemon
post Jul 13 2014, 18:24
Post #90





Group: Members
Posts: 144
Joined: 1-April 09
Member No.: 68578



Great to see some new code here. I was considering FLAC to be "finished" to some extend, as in there is pretty much no room for further improvements.

Definitely glad to see someone take up the work.

I wonder when packagers start pushing builds into upstream, so my Fedora and Debian releases see them.


--------------------
-EOF-
Go to the top of the page
+Quote Post
audiophool
post Jul 13 2014, 19:17
Post #91





Group: Members
Posts: 25
Joined: 13-June 14
Member No.: 116389



QUOTE (polemon @ Jul 13 2014, 19:24) *
I wonder when packagers start pushing builds into upstream, so my Fedora and Debian releases see them.

Ubuntu Trusty (14.04) packages Flac 1.3.0, so I bet it's in Sid at the very least. I don't know about Fedora, though.
Go to the top of the page
+Quote Post
IgorC
post Jul 14 2014, 07:53
Post #92





Group: Members
Posts: 1560
Joined: 3-January 05
From: ARG/RUS
Member No.: 18803



Interesting.
The speed of encoding for the new alpha x64 build is now on par with FLACUDA/FLACCL.

foobar2000's converter w/ 8 threads:
FLACCL -6
FLAC (x64 alpha build) -8

Ivy Bridge quad-core i7, HD4000 graphics.

This post has been edited by IgorC: Jul 14 2014, 07:58
Go to the top of the page
+Quote Post
ktf
post Jul 14 2014, 08:31
Post #93





Group: Members
Posts: 360
Joined: 22-March 09
From: The Netherlands
Member No.: 68263



QUOTE (IgorC @ Jul 14 2014, 08:53) *
Interesting.
The speed of encoding for the new alpha x64 build is now on par with FLACUDA/FLACCL.

That depends very much on the ratio CPU-speed/GPU-speed of course.

On the flac-dev mailinglist itself, some performance checks have been posted as well.
http://www.audiograaf.nl/misc_stuff/FLAC-p...nux-GCC-4.8.pdf (system specs here: http://lists.xiph.org/pipermail/flac-dev/2...ly/004889.html)

This shows speedups around 60%

This post has been edited by ktf: Jul 14 2014, 08:32


--------------------
Music: sounds arranged such that they construct feelings.
Go to the top of the page
+Quote Post
skamp
post Jul 14 2014, 09:30
Post #94





Group: Developer
Posts: 1430
Joined: 4-May 04
From: France
Member No.: 13875



74 minutes album, flac -8, on a mobile Core i7 CPU @ 2.2 GHz:
99.81 seconds with 1.3.0, 52.69 seconds with git. That's insane ohmy.gif

This post has been edited by skamp: Jul 14 2014, 09:30


--------------------
See my profile for measurements, tools and recommendations.
Go to the top of the page
+Quote Post
dfkt
post Jul 14 2014, 21:59
Post #95





Group: Members
Posts: 16
Joined: 16-January 07
Member No.: 39719



Can't edit my post #79 above (some mod please delete it), so here's the updated list... yowzer.

CODE
system:                 i7-3770, 16gb ram, win8pro-x64
test method:            flac.exe -8 input.flac -o output.flac

flac 1.3.0 binaries:    http://www.hydrogenaudio.org/forums/index.php?showtopic=101082
________________________________________________________________________________
input:                                          bytes           file MD5                            fb2k verification MD5

flac 1.2.1, 3h22m, 44.1kHz/16bit/2ch            1,708,858,410   1092139DA14A43C016F614A1A070598C    FC2B730F57920D49D4E41604FE14A623
________________________________________________________________________________
results:                        time    ratio   bytes           file MD5                            fb2k verification MD5               flac.exe MD5

1.3.0-case-x32                  2:47    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    4ADD826644D303B57D220F866C67292B
1.3.0-case-x32-fb2k             2:47    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    4ADD826644D303B57D220F866C67292B
1.3.0-flacfrontend21-x32        2:49    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    8C4AE54CE1B243904CEAB30575534FB3
1.3.0-flac-git-3194829-x32      1:57    0.988   1,705,098,763   49D7D89FAD8335F339915F7FED0C1265    FC2B730F57920D49D4E41604FE14A623    3B67802CC3D2BD4572B9466763BF2E21
1.3.0-flac-git-3194829-x64      1:53    0.988   1,705,098,763   49D7D89FAD8335F339915F7FED0C1265    FC2B730F57920D49D4E41604FE14A623    27924BE6762C8CEC06CE811DF0513ECA
1.3.0-john33-x32                2:52    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    C31C314E8C1E8DC62EE0CDBE048C6112
1.3.0-john33-x64                2:38    0.998   1,705,098,763   49D7D89FAD8335F339915F7FED0C1265    FC2B730F57920D49D4E41604FE14A623    29546F37C3EB364155FADD43EC570DED
1.3.0-lamedude-sse1-x32         2:53    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    F08E20ADF2042919CA3BF8305E36CFF6
1.3.0-lamedude-sse2-x32         2:47    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    6C52EEA53F27CFEC697632194AA4D5D5
1.3.0-lamedude-x64              2:58    0.998   1,705,098,763   49D7D89FAD8335F339915F7FED0C1265    FC2B730F57920D49D4E41604FE14A623    174360529CEBF9371B41019D46F1FDC4
1.3.0-lrn-mingw32-x32           3:18    0.998   1,705,098,763   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    E87DC0158474E662637886797F36C0DA
1.3.0-lrn-mingw32-o2-x32        2:52    0.998   1,705,098,763   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    E0D5DAE8EF9BA49D0822901FE5D114A6
1.3.0-lrn-mingw32-o2-core2-x32  2:45    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    4D650372DBDB65DCC76E34DBAA42FB8F
1.3.0-wcd-x32                   4:00    0.998   1,705,099,033   9BE9ABC5283909DC0F08A822F0B8FF46    FC2B730F57920D49D4E41604FE14A623    E5639F8BABD64836B85B4DB119279A7B

1.2.1-official-x32              3:12    0.998   1,705,099,033   2BA84F8B1B5514BE4BED536F341875C7    FC2B730F57920D49D4E41604FE14A623    16344C45643E41544A5C1C926A109C9F
1.2.1b-official-x32             3:15    0.998   1,705,099,033   2BA84F8B1B5514BE4BED536F341875C7    FC2B730F57920D49D4E41604FE14A623    B8396749F374C9F48CBE3B16C5333D49
Go to the top of the page
+Quote Post
Wombat
post Jul 14 2014, 22:46
Post #96





Group: Members
Posts: 1020
Joined: 7-October 01
Member No.: 235



john33-x64 compile of 1.3.0 still has the fastet RG scanning with metaflac. No science but 5GB scanned in 1.06min against 1.12min with git x64.
Go to the top of the page
+Quote Post
Maurits
post Jul 14 2014, 23:30
Post #97





Group: Members
Posts: 380
Joined: 30-September 05
From: London, Europe
Member No.: 24805



QUOTE (GrieverV @ Jul 13 2014, 15:10) *
I hope they start thinking about tagging a release soon because these improvements are definitely worth it.

They are thinking of a 1.3.1: "This fix alone is worth a new release so its time to work towards one."

It would probably be useful if people gave this Git a good grilling and report any issues here or on the mailing list.
Go to the top of the page
+Quote Post
AiZ
post Jul 15 2014, 01:55
Post #98





Group: Members
Posts: 48
Joined: 4-February 02
Member No.: 1251



Hello guys,

Could you please test this one, it looks pretty good for a win32 compile.

CODE
-=-=-=-=-=-=-=-=-=-= Configuration Complete =-=-=-=-=-=-=-=-=-=-

  Configuration summary :

    FLAC version : ........................ 1.3.0

    Host CPU : ............................ i686
    Host Vendor : ......................... pc
    Host OS : ............................. mingw32

    Compiler is GCC : ..................... yes
    GCC version : ......................... 4.9.0
    Compiler is Clang : ................... no
    SSE optimizations : ................... yes
    Asm optimizations : ................... yes
    Ogg/FLAC support : .................... no

Quick test:
CODE
Command:        flac.exe -8

CPU:            Intel e7200@2.86GHz
CD Image:       44.1/16/Stereo, 58:44

GrieverV    git 3194829    64bit: ~50.20s
GrieverV    git 3194829    32bit: ~57.68s
john33        1.3.0        32bit: ~83.97s
AiZ         git 7251201    32bit: ~51.05s

Thanks,

AiZ


--------------------
AiZ stupid homepage - http://aiz.free.fr
Go to the top of the page
+Quote Post
GrieverV
post Jul 15 2014, 04:02
Post #99





Group: Members
Posts: 4
Joined: 12-July 14
Member No.: 116659



QUOTE (Maurits @ Jul 15 2014, 00:30) *
QUOTE (GrieverV @ Jul 13 2014, 15:10) *
I hope they start thinking about tagging a release soon because these improvements are definitely worth it.

They are thinking of a 1.3.1: "This fix alone is worth a new release so its time to work towards one."

It would probably be useful if people gave this Git a good grilling and report any issues here or on the mailing list.

That's awesome news! I'm not sure how I missed that on their mailing list.


QUOTE (AiZ @ Jul 15 2014, 02:55) *
Hello guys,

Could you please test this one, it looks pretty good for a win32 compile.

-snip-

Thanks,

AiZ

It's nice to see MSVC compiles are comparable to GCC and vice versa. 64bit builds would be nice to see since the 32bit is nearly the speed of the MSVC 64bit.

I'd be wary of using GCC 4.9.0. I've seen a few projects (FFmpeg, Musl) that have had some serious issues with it.

This post has been edited by GrieverV: Jul 15 2014, 04:05
Go to the top of the page
+Quote Post
Brand
post Jul 15 2014, 11:12
Post #100





Group: Members
Posts: 317
Joined: 27-November 09
Member No.: 75355



QUOTE (GrieverV @ Jul 12 2014, 05:05) *
FLAC git 3194829 win32 and x64 compiles:
win32, x64

Built using MSVC 2013 Update 3 RC using the Windows XP toolset (v120_xp)

I decided to use the latest FLAC git instead of 1.3.0 due to various Windows/MSVC support improvements.
You can browse FLAC's git repository and view all the changes since 1.3.0 here: https://git.xiph.org/?p=flac.git;a=summary

I get a nice encoding speed improvement with an i5 CPU, but with an older AMD X2 4000+ it performs the same as the old 1.3.0 from last year. Is this normal?
(Always using x64 on Win7.)
Go to the top of the page
+Quote Post

8 Pages V  « < 2 3 4 5 6 > » 
Reply to this topicStart new topic
2 User(s) are reading this topic (2 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 31st August 2014 - 06:05