Musepack Forums  

Go Back   Musepack Forums > Main > General

Reply
 
Thread Tools Search this Thread Display Modes
Old 27 August 2005, 10:24 am   #1
tonatona
Member
 
Join Date: Aug 2005
Posts: 3
Default quality suggestion

Hi,
i want to compress my whole huge collection (different music styles, sound effects, loops, grooves etc.) from Wave to Musepack using Monkey Audio 3.99. Is quality 5 good enough according to my configuration or shall i use a higher quality preset?

Sennheiser HD 650 on Vincent KHV 111 on DIY-DAC (around 300-400). And good ears.
tonatona is offline   Reply With Quote
Old 27 August 2005, 11:17 am   #2
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

"From Wave to Musepack using Monkey's Audio 3.99" is confusing.
If you mean you want to compress as lossless, APE files, that's Monkey's Audio format, not Musepack.
If you mean you want to convert from Monkey's Audio (APE) files to Musepack (MPC), read further:

How expensive and/or high quality your equipment is has little to do with choosing a preset to use. $30 headphones and a regular sound card headphone output can produce frequencies as high as ~20khz as well, what would allow you to notice possible artifacts just as well. Higher quality equipment can of course increase the detail level, but that difference doesn't always have a benefit for exposing artifacts.

The main factors are your hearing and your preferences. No one can recommend to you what to use and be 100% right. You don't need to, and shouldn't trust anyone but yourself when it's your hearing and preferences that are the judges.
Refer to my reply on this thread.

And regarding sound effects, loops etc., refer to this thread.
Shy is offline   Reply With Quote
Old 27 August 2005, 12:13 pm   #3
tonatona
Member
 
Join Date: Aug 2005
Posts: 3
Default

I'm using Monkey Audio as front end for mp3 and musepack. But i'm currently testing foobar2000, EAC and mpc batch encoder. I have both CDs and Ape files to compress.


Of cource the quality of your equipment has a lot to do with choosing a preset. You cannot hear differences between details without hearing details. And the more details you hear the more information is placed to your disposal.
"regular sound card headphone output". What's that? There are just few sound cards with a headphone output. That's why you have to use head amps. Do you connect your microphone to the line in input of your sound card?


PS: Does foobar2000 use the latest musepack encoder (1.15v)? Do you need --xlevel command in foobar2000?
tonatona is offline   Reply With Quote
Old 16 September 2005, 07:11 pm   #4
Lost Angel
Member
 
Join Date: Feb 2005
Location: Bremen, Deutschland
Posts: 11
Send a message via ICQ to Lost Angel Send a message via MSN to Lost Angel
Default

I'd go for braindead compression regardless of what equipment you have... simply because it is best quality while still not as big as the /lossless/ file.
__________________
Rain... Night... Musique...
Lost Angel is offline   Reply With Quote
Old 16 September 2005, 07:12 pm   #5
Lost Angel
Member
 
Join Date: Feb 2005
Location: Bremen, Deutschland
Posts: 11
Send a message via ICQ to Lost Angel Send a message via MSN to Lost Angel
Default

ps foobar is not using the latest codec to make mpcs, since it has been compiled much earlier, not sure about the recent (more or less) betas.
__________________
Rain... Night... Musique...
Lost Angel is offline   Reply With Quote
Old 16 September 2005, 08:42 pm   #6
tonatona
Member
 
Join Date: Aug 2005
Posts: 3
Default

Thank you for the replies.
tonatona is offline   Reply With Quote
Old 16 September 2005, 08:45 pm   #7
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

As I had mentioned in the thread I referred to, there is no definite answer, like the one you've set, and --insane is "the best overall quality to encode to if you require the very best transparency without a too high bitrate" and not braindead, as tested on countless samples.
foobar2000 doesn't come with the Musepack encoder, its converter comes with Musepack presets and you set the location of the encoder executable.

And please, no more "P.S.". The terms of service exist for good reasons.
Shy is offline   Reply With Quote
Old 20 September 2005, 04:28 pm   #8
Kamen
Member
 
Join Date: Sep 2005
Location: Bulgaria
Posts: 19
Default

Well, I usually use Xtreme and Insane, sometimes BrainDead or even Above BrainDead, more rarely Standard. It depends on the spectrum of your music and its profile and of course your wishes and sensitivity, and how important is the music to you. All is a question of musical and psychoacoustics. If the music has little presented high frequencies and/or is simple, I usually use a little lower quality setting. I use both ears and monitors sometimes and sometimes add manual switches for minor adjustments, thus making some settings between the quality presets in --quality and --[profile]. Let's say, for example, somewhere between Xtreme and Insane...
Kamen is offline   Reply With Quote
Old 20 September 2005, 09:35 pm   #9
xmixahlx
Musepack developer
 
xmixahlx's Avatar
 
Join Date: Nov 2004
Location: seattle, washington usa
Posts: 111
Send a message via ICQ to xmixahlx
Default

i usually use --standard, which i think would be transparent in ~99% of cases. if you want more room to wiggle, try --insane for ~99.9%

oh, and i pulled those numbers out of my ass.


later
__________________
-xmixahlx, the one they call "mike"
http://xmixahlx.com -|- http://rarewares.org
xmixahlx is offline   Reply With Quote
Old 20 September 2005, 10:08 pm   #10
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

Whistler, as I mentioned, there's very, very slight chance you'll notice a difference with --insane but not with --braindead. If there's any musical case you can demonstrate this on, we'd love to know about it.
Whether a piece consists of much high frequency content or not is often hardly related. The psychoacoustics handling of Musepack is much more complex than simply "lots of highs/transients = high bitrate/higher quality needed." In fact, such content is often "easy" to encode for Musepack, and music that sounds "simple" can in fact be very hard to encode. Sounds that seem "simple" and basic can in fact pose the biggest issues for audio codecs.
Shy is offline   Reply With Quote
Old 21 September 2005, 02:38 pm   #11
Kamen
Member
 
Join Date: Sep 2005
Location: Bulgaria
Posts: 19
Default

Quote:
Originally Posted by Shy
Whistler, as I mentioned, there's very, very slight chance you'll notice a difference with --insane but not with --braindead. If there's any musical case you can demonstrate this on, we'd love to know about it.
Whether a piece consists of much high frequency content or not is often hardly related. The psychoacoustics handling of Musepack is much more complex than simply "lots of highs/transients = high bitrate/higher quality needed." In fact, such content is often "easy" to encode for Musepack, and music that sounds "simple" can in fact be very hard to encode. Sounds that seem "simple" and basic can in fact pose the biggest issues for audio codecs.
Yes, and that's the reason I like Musepack - its very, very precise and perfect psychoacoustics handling, in comparison to all other lossy compression formats. I rarely use BrainDead - only for encoding very, very special albums to me. To try testing the differences, I gotta go to a sound engineering studio in my town (one of my friends works in such studio, so some day we can try, it will be interesting). I don't think regular sound card is enough, because such a card often implements distortion and noise in the sound (my home soundcard isn't good enough for such precise measurings by ear) and can change things.
I just talked generally. And not so literally. Of course, complexity of music doesn't depend on the HF content, it depends on the overall "picture", but usually HFs is the most problematic zone (at least for other encoders and not so for MPC!). Sometimes music only seems simple to the ear, but technically it can be more complex (that's why such judgements by ear aren't absolutely correct, as I mentioned above). In fact, differences in literally simple music often can be noticed easier by the ear. It's a question of psychoacoustics, as I said. And as you said, settings depends on the sensivity and preferences of the particular user.
Kamen is offline   Reply With Quote
Old 21 September 2005, 04:44 pm   #12
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

In fact even the simplest Sound Blaster Audigy and not so expensive headphones are good enough for testing. It's a big misconception that top of the line equipment is indispensable for testing audio codecs.
Shy is offline   Reply With Quote
Old 21 September 2005, 05:01 pm   #13
Kamen
Member
 
Join Date: Sep 2005
Location: Bulgaria
Posts: 19
Default

It is important what one has in mind by saying "regular soundcard". Now, it's better when you specified some model for orientation and I can say I'm more likely to agree here, but still not completely agree. Well, that's another thread which covers the technical parameters and specifications of soundcards. The top equipment isn't needed, but at least average quality or little above it is good. Audigy isn't bad, but If you take one AC'97, for example, it is not so suitable. This card sounds terrible to me.
Kamen is offline   Reply With Quote
Old 21 September 2005, 05:42 pm   #14
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

Many AC97-based cards may indeed be unsuitable because their frequency curve is pretty much ruined. I mentioned Audigy as an example of a cheap yet perfectly suitable card, as a response to the studio thing, as having access to a professional sound studio is definitely not needed We wouldn't like people to think they have to use only top of the line equipment to help test Musepack.
Shy is offline   Reply With Quote
Old 29 September 2005, 09:34 am   #15
Hellygator
Member
 
Join Date: May 2005
Posts: 2
Default

I have an Audigy2 myself. Owners of Creative Soundcards from SB Live! to Audigy 2 should mind the audible resampling artifacts that this cards produce. Especially Audio with 44.1Khz Samplerate (which is the most common i guess, because it's used with AudioCDs) gets distorted by those cards. Ofcourse, this is not a Musepack related problem.
Hellygator is offline   Reply With Quote
Old 29 September 2005, 10:10 am   #16
Seed
Musepack Nanny
 
Seed's Avatar
 
Join Date: Jul 2004
Posts: 168
Default

Some think they can just ignore it and it'll go away Some others think Creative is doing a better job than in the past. I think this issue is serious and no self-respecting music lover should touch their hardware.
Seed is offline   Reply With Quote
Old 29 September 2005, 03:35 pm   #17
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

Yes, I almost forgot about it. If anyone wants to do tests on a card that resamples to 48khz, they must perform those tests at 48khz to avoid the card's resampling. Samples can be resampled to 48khz using a high quality software resampler such as SSRC.
Shy is offline   Reply With Quote
Old 30 September 2005, 12:55 am   #18
Lost Angel
Member
 
Join Date: Feb 2005
Location: Bremen, Deutschland
Posts: 11
Send a message via ICQ to Lost Angel Send a message via MSN to Lost Angel
Default

hmm, posted ages ago, but looks like I am a bit more confused now comparing to how I was before. Saying that for quality preservation (keeping it closest to how original sounds exactly) braindead is most suitable, I was actually quite confident, yet it appears that insane is sufficient. Now my confusion is specifically regarding this "sufficiency" aspect. Is braindead in any way "more transparent" than insane? Or is insane less "transparent", but with smaller filesize? (I am not talking about what I would be able to hear with different headphones kind of difference, but about it being closer or more distant from the original - where do those extra kbs go in braindead if it is just like insane? or is is somehow better?) Hope I am still in the right thread with this point and I did read a thread discussing the optimal choice and suggesting to do blind hearing test for oneself to make up one's mind etc).

And my apologies for "P.S. thing above".
__________________
Rain... Night... Musique...
Lost Angel is offline   Reply With Quote
Old 30 September 2005, 07:21 am   #19
Seed
Musepack Nanny
 
Seed's Avatar
 
Join Date: Jul 2004
Posts: 168
Default

I can't think of a sample that I know that --braindead handles well and --insane fails at. In my opinion going above --insane is a waste of bits. If --insane can't handle a sample properly, there's a good chance that it'll be flawed with --quality 9 too and possibly even --quality 10.
Seed is offline   Reply With Quote
Old 06 October 2005, 12:52 am   #20
Vertigo
Member
 
Vertigo's Avatar
 
Join Date: Sep 2005
Location: Lakeland, FL
Posts: 10
Send a message via ICQ to Vertigo Send a message via AIM to Vertigo Send a message via Yahoo to Vertigo
Default

Seed, you know the code better than I, but I think that statement could also read "going above --xtreme is a waste of bits"...I've yet to encounter a real world sample that doesn't sound transparent to me at this setting. I have a quadraphonic setup, and this includes spatial imaging.
Vertigo is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Best Quality Setting? / Best CD Ripper? bbommarito General 2 27 March 2006 12:17 pm
MPC Quality presets audible difference? gerardobaez General 1 13 August 2005 01:24 am
How to re-encode mpc files at reduced quality? drmrbrewer General 6 31 July 2005 07:28 pm
sound quality improvements between versions? JEdwardP General 3 03 June 2005 01:29 pm
Quality Presets (Was: Really stupid question) MechaTama31 General 2 05 December 2004 01:19 pm


All times are GMT. The time now is 04:17 am.


Powered by vBulletin® Version 3.8.11 Beta 2
Copyright ©2000 - 2019, vBulletin Solutions Inc.