Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL (Read 52101 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #25
Wow those are some lengthy detailed posts there Chinch haha

This talk about Foobars bit-compare, can it determine whether a FLAC file is true CDDA or not? I check all my FLAC files with AudioChecker 2.0 (build 457) it would be nice to eliminate that step / piece of software.

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #26
Since you are a foobar2000 user, I am making the guess that you were converting from there.

Actually, when I encode from single WAV to FLAC, I usually do it in batch... and I simply use the Flac Frontend (the basic, generic one)... but even so, it still doesn't matter, as I just watched it again as I encoded from WAV to FLAC... and it took just as long as the other. Viewing the Resource Monitor, I saw FLAC.exe pop up once it started, and guess how many threads it was using? 1. The whole multi-thread option you're looking at will not bypass this limitation. What it *is* useful for is encoding multiple seperate tracks *in tandem* or in parallel, however you want to put it. So if I were encoding 10 WAV files to 10 FLAC files, then it'd likely fire off 4 to 8 copies of FLAC.exe, each core or HT handling its own track/thread. In doing this, the limitation can be surpassed, by simply executing several copies of the program at once and letting them run at the same time. That's what the threading is for there, I would have to assume.

If you have a look at Preferences --> Advanced --> Tools --> Converter, you will see an option to use multiple threads, and you will see an option to assign priority to conversion.

I actually already tried this earlier by using the START command, which you can use a priority switch to fire off a process... say... START /HIGH flac.exe blah.wav -8 or whatever. So I've already recreated that, and the priority doesn't affect the encode for any instances I have done.

I believe the default values are as on my setup: autodetect # of threads, and set priority to 2, on a scale from 1 to 7. I guess that explains the 15 percent figure. Try to set priority to 7 and redo the experiment. (Maybe it even helps to set fb2k at high or realtime priority first.)

Doesn't matter the priority of foobar.exe, since it opens the external FLAC.exe to do the encoding job. As I mentioned above, I have actually SET the priority via command line switches when starting up the actual FLAC.exe process, with no effect. You can look into the "start" command if you'd like, and its parameters... you will see the ability to set processsor affinity in there, but don't get confused, that is a RESTRICTION not an upgrade or whatever... you are telling it to run on "this" processor core ONLY.

As far as I know, you are right that FLAC.exe itself does not support multithreading. However, fb2k will simply start one for each number of threads, although no more than the number of files to convert. That means that big jobs are about equally efficiently done as if FLAC itself had supported multithreading.

I can finally agree with you here... I can empirically prove that it's not a multi-threaded process (flac.exe)... it's an old 32-bit DOS/command prompt/console based application, that like I was saying earlier... was compiled in 2007! A loooot has changed, hardware-wise and otherwise since 2007. It's definitely not optimally compiled for today's architectures.


fb2k has a decoding speed test: http://wiki.hydrogenaudio.org/index.php?ti...(foo_benchmark)
You can choose whether to buffer the files in RAM in order not to be constrained by drive speed. My few-years old laptop with a Turion TL-56 (dualcore at 1.8 MHz), decodes from RAM at slightly less than 200x realtime, meaning that the actual playback is already down to half a percent of max CPU load. And the differences between FLAC -8, FLAC -0, TAK and WavPack are quite uninteresting ...


Yep, I already have this one. I've been using foobar for years and years... so any of the plugins like that and the Binary Comparator (which I use a lot) I'm very familiar with. Buffering to memory isn't that big of a deal to me, I run a 4 drive RAID 10 array, with Western Digital Black model drives (which smoke even as a single drive)... Just for your interest, here is the result I have:

Code: [Select]
Total length: 1:18:56.600
Info Read time: 0:00.000
Opening time: 0:00.313
Decoding time: 0:10.564
448.367x realtime


That's still using TOOL - Lateralus... which as you see is 01hr:18m:56s -- why can't other bands fill up a CD like TOOL always manages to? I get tired of paying so much for 25-30 minute albums, it's garbage. They used to call that an "EP"... now they call them "LP"s and then wonder why sales of their $22 1/2 an-hour CD are low. Write more than 3-5 songs... make them more than the traditional 2 minute punk track... then re-test the waters.

But uhm... yeah the speed difference is fractions of a second difference whether I buffer it to memory or not, so it's not important here... but I am running an insane amount of apps right now, to include THREE different browsers, (this one has about 25-30 tabs open, just itself)... plus an audio track I've been mixing in Audacity that's like 20 tracks thick because I'm in the midst of editing and bouncing, chopping and joining samples and clips.

Yeah, task manager says 166 processes running, and it doesn't count the individual tabs in Firefox and IE9 (though it does for Chrome, which I'm  using now)... and I'm currently using 9GB of memory. Running 64-bit OS, obviously...

Ah yes, and I also know all of the technicals of ripping and ripping properly.... if that's in question at any point, I'll be pro-active:

Code: [Select]
Track  Status
01     Accurately ripped, confidence: 200.
02     Accurately ripped, confidence: 200.
03     Accurately ripped, confidence: 200.
04     Accurately ripped, confidence: 200.
05     Accurately ripped, confidence: 200.
06     Accurately ripped, confidence: 200.
07     Accurately ripped, confidence: 200.
08     Accurately ripped, confidence: 200.
09     Accurately ripped, confidence: 200.
10     Accurately ripped, confidence: 200.
11     Accurately ripped, confidence: 200.
12     Accurately ripped, confidence: 200.
13     Accurately ripped, confidence: 200.


I'd say 200 matches in the AccurateRip database is sufficient enough to call it "dependable". (I believe 200 is the max matches ARv1 will return without setting it to verbose mode?)

And trust, my tagging is complete and I take great care in maintaining my digital audio collection. I have personally ripped all of my discs, properly... and tagged them with nearly every detail you can imagine, as you see here...

and... oh yeah as I said, I do EAC->IMAGE+CUE+LOG... CUE is embedded and artwork is embedded as well. Lyrics are in there... actually... I have the LYRICS tag used... which of course means yes, I have synchronized lyrics in the puppy even... I know it's not showing the "real" cover of the album. I put that one there, so kiss it... If you have the CD you'll know the "cover" is clear cellophane type material, in layers... which is a PITA to scan and then have to fix the scan colors and artifacts, etc... (my scanner is a older flatbed and sucks pretty much... it works though). It can scan things up to the resolution of my HOUSE, LOL... like WTF is that?!? Oh... that's a pixel. Taking up my entire 1080p display. Just... 1 pixel? Damn. I think I set the resolution up a bit too much cause the JPEG saved at 10% quality is 900MB. Now it's just one gigantic, house-sized 900MB DISCOLORED pixel. Ahhhh... hyperbole. I love it. Do me a favor though, after all of my efforts, Porcus... can you put your trust and faith in the idea that I actually do know what I'm talking about and doing, and I'm not a n00b?

It seems like... though mostly you're carrying on a discussion with me about these subjects... I've gotten the feeling that since your first response to me, you've been trying to prove me wrong or catch me making some stuff up, haha... like every time I counter-evidence your doubts, you come right back with more things for me to try and prove! I ain't mad'atcha... It's all cool, I just was being honest with you about how it seemed from this side of the fence.

I had a good time spankin' ya though on each comeback you had for me, LOL. It's all in good fun... but I think that's the last time I will be able to explain myself for tonight, my brain is about to shut down... need rest! So does this forum, I think. I have laid some textual abuse down on it today. Once you get me blabbing about this stuff, it's tough for me to STFU... but I'll zip it right here. Talk to you later, my friend...



What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #28
Re your concern about a newer FLAC build; I knew I had seen one, because I actually downloaded the build, but luckily someone awoke one of those threads:
http://www.hydrogenaudio.org/forums/index....showtopic=91668


The speeds are such a minuscule difference it really doesn't matter to me I built my computer not to game but to keep my blood pressure down with patience and I have never ran out of patience or even close with anything FLAC related.

When you say newer FLAC build as in newer encoding/codec?

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #29
When you say newer FLAC build as in newer encoding/codec?


It was a reply to a complaint that it was old and didn't utilize multithreading, and wanted to compile one for multithread support -- I poined out that someone had done the job. I hae no indication that it produces different files.

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #30
Hello guys, I've been reading your posts as I was worried about which encoder to use in CUETools. Please help me decide I'm not well versed on this issues as I only knew of flac files weeks ago. I only want to re-encode the single flac (or ape) files into flac tracks without silence between the songs to be played mainly on Winamp. I own a regular PC not good, not bad I don't care if the re-encoding takes more provided I don't mess it up. I've recently been using the 11 level of compression but I've just read somewhere else it's not reccomended. What's the most common option for somebody like me who know practically nothing? Sorry for the silly question but I haven't been able to deduce it from your posts. Thanks.

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #31
What's the most common option for somebody like me who know practically nothing?

Use libflac. I've had compatibility issues with the cuetools encoder, even at compression level 8 (e.g. seeking not working).

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #32
What's the most common option for somebody like me who know practically nothing?

Use libflac. I've had compatibility issues with the cuetools encoder, even at compression level 8 (e.g. seeking not working).

Can you be more specific? Do you mean libflake or flacCL or even the older flaCuda? I use the GPU encoders from day one and neither my hardware devices nor my software players have problems with -8 at seeking.
It will be interesting if you provide more info or samples.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #33
What's the most common option for somebody like me who know practically nothing?

Use libflac. I've had compatibility issues with the cuetools encoder, even at compression level 8 (e.g. seeking not working).

Can you be more specific? Do you mean libflake or flacCL or even the older flaCuda? I use the GPU encoders from day one and neither my hardware devices nor my software players have problems with -8 at seeking.

Flake, what in 2.1.5 is now listed as 'cuetools'. The player is foobar2000. I've had several instances, where files gave seek errors (most do work though). Reencoding worked fine and the files were bit-perfect.

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #34
Flake, what in 2.1.5 is now listed as 'cuetools'. The player is foobar2000. I've had several instances, where files gave seek errors (most do work though). Reencoding worked fine and the files were bit-perfect.

I mostly still use Winamp but sometimes foobar. I don´t have that problem, rercently using flacCL. Maybe Grigory may have an idea what is different with seeking against libflac.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #35
Flake, what in 2.1.5 is now listed as 'cuetools'. The player is foobar2000. I've had several instances, where files gave seek errors (most do work though). Reencoding worked fine and the files were bit-perfect.

I mostly still use Winamp but sometimes foobar. I don´t have that problem, rercently using flacCL. Maybe Grigory may have an idea what is different with seeking against libflac.

I just tried to reproduce it, but with foobar 1.2.6 even level 11 is working for everything I tried.

Maybe the problem was/is not even related to the flac encoder. The seek table seems to be stored with the metadata, maybe foobar had problems with some metadata that was there, resulting in it not reading the seek table properly.

Unfortunately, I don't have any defective files lying around - everything was reencoded (and I did redo a lot of metadata).

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #36
Flake, what in 2.1.5 is now listed as 'cuetools'. The player is foobar2000. I've had several instances, where files gave seek errors (most do work though). Reencoding worked fine and the files were bit-perfect.

Thank you, there was indeed a problem with seek table in 2.1.5 cuetools flac encoder. Fixed in latest build.
CUETools 2.1.6

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #37
I also tried latest 2.1.5 because i am still with 2.1.2 in favour of its Tag handling, i don´t like the Comment from the CUE is in my flac file Tag.
Strangely when using the flacCL codec from within CUEtools it gives me different filesizes as when i use the cmd executable. This was not the case with older versions. While trying what swiitch may differ i even was able to crash my Windows 7/64 the first time using a group-size of 64.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #38
I think the difference in size is just padding. CUETools knows the approximate size of tags / cover art and calculates padding accordingly, so that it wouldn't normally have to rewrite the whole file.

There will likely be a separate option for comments tag in the final release.

Yes, be careful with low-level hardware settings for OpenCL
CUETools 2.1.6

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #39
What audioplayer do you use for your FLAC-files? Your screenshot looks nice.

What differences among CUETools outputs? libFlake/libFLAC/flake/FLACCL

Reply #40
What audioplayer do you use for your FLAC-files? Your screenshot looks nice.


Maybe I wrong, but it is a very minimalistic but with a good album/pertrack info interface from foobar2000 : )
loquor mee menti: factus de materia, cinis elementi...