Forum Replies Created
-
Posts
-
[quote author=”JesseG”]I would guess because it’s limiting inter-sample peaks. But… I didn’t think Breakaway Live limiter did that. :< Do you also have the high pass and low pass filters disabled?[/quote]The TT DR meter post Stereo Tool shows correctly -1.00 dB peak.
[quote author=”TT DR meter PDF manual”]In the area close to full scale, peak measurement is particularly complex and critical.
First of all, fixed point resolution can only show values up to full scale, since on the
digital level, no overs are possible. However, contiguous full scale words create audible
overs and so called interleaved sample overs. Secondly, floating point calculation makes it
possible to represent values well over 0dB. The measurement and display of peak values in
4x over sampling leads to a display of overs so frequently that we have found a middle of the
road solution. The peak values are measured "normally" and provided numerically.
In the case that two contiguous bit words show full scale (all 16 bits at 1 without any
oversampling) and at the same time a value of over 0 dB is detected via oversampling metering
(run as a parallel process), then the peak display shows "OVER."[/quote]
It is caused by the BA Live high pass filter.
I had not changed this setting since BA Live installation, so it was still activated with the default value (-> HPF: 20 Hz).
Now I have disabled it (HPF: Off) and everything is ok. ❗But I wonder if it is normal that as long as the HPF is active you have to lower the effects output level to -2.5 dB (75 %) to avoid that peak limiter activity ? ❓
[quote author=”JesseG”]I think you understand what I’m getting at, right?[/quote]Aha, ok now I understand.
Undo is not meant as a standalone dynamic remastering (wonder) tool.
It is only one step of (pre-)processing to prevent further damages to already heavy limited material by adding more headroom.
➡ You should not eat a meal before the whole cooking (process) is done ! 😆Jesse, thank you very much !
I don’t like the original version, because the drum sounds have too much compression (compressed and limited to death – Keith Moon is crying in his grave).
Sorry Jesse, but the UNDO version I don’t like, too.
Now the drums (especially snare) are much too aggressive and dominant IMO (it really hurts my ears).I guess if you would use half of the expansion factor it could sound much better.
I know Izotope RX Declip and SeeDeClip.
But the audible results are very different than what we hear with this UNDO.Most stunning of UNDO processing in magic.WAV is the new punch and presence of drum sounds.
Jesse, please I would like to hear what this UNDO can do with ‘Greenday – American Idiot’.
This album has horrible clipping and a DR5 (dynamic range) value analyzed by the TT DR offline meter.
So, please UNDO track 1 (‘American Idiot’) in the same manner as the magic.WAV:
60 seconds original + 60 seconds with UNDOThis would be a very interesting torture test for UNDO.
(maybe in combination with declipping before UNDO)I guess it is based on this kind of technical concept !?:
❓You need two envelope follower circuits with different attack times to detect the transients of the input signal and two envelope follower circuits with different release times to detect the sustain of the input signal.
The difference between both envelope follower circuits (one for transients and one for sustain) generates two control signals (one for transients and one for sustain).
Then you have to clone the input signal and multiply one instance with the control signal for transients and the other instance with the control signal for sustain.
This way you split the original input signal in two different signals.
One signal contains only transients and the other only sustain.So, finally you can re-mix these two signals.
For a dynamic (upward-) expansion you add more gain to the transient signal.That’s it (in my theory).
By the way:
Compare the dynamic range of the original (first 60 seconds) and the processed version (e.g. with TT DR VST meters or the offline tool – important: you have to split it and normalize both signals to approx. 0 dB) !
This kind of dynamic (upward-) expansion processing proofs that such a (green !) signal sounds much better (more vivid and punchier) than a traditional compressed and peak limited signal ever can.[quote author=”JesseG”]Please upload any builds you have of it too, and people can see if it’s faster for their particular CPU.[/quote]
➡
ftp://ftp.tmn.ru/pub/windows/misc/lame-3.93.1.zip
This is build Dec 3 2002 that I use.
It is faster with my Athlon X2 6000+ than the John33 build (Dec 12 2009).[quote author=”Dj Buik”]I was wondering how did you make this graphics?
What software (or hardware?) did you use?[/quote]1. I have used the lame.exe 3.93.1 build Dec 3 2002.
You can find many downloads for this build with google – though it doesn’t matter if you use this build or the initial build from Dec 1 2002 or Jesse’s build.
You always get the same encoded MP3 (only difference: on my system Jesse’s build is slower than the other builds !?).2. I have even used only the standard settings (not Jesse’s advanced settings) !
This command line in dMC R14:
"-s 44.1 – [output]"Using the addtional CLI plugin for dMC (I prefer to use dMC R14 because it supports multi core encoding and VST plugins).
Important: the CLI plugin is the only way to use LAME versions prior to version 3.98 in dMC R14 !
Replacing the 3.98.4 lame.exe with an older version and the usage of the standard LAME encoding GUI leads to completely distorted files (full of noise).
And ACM LAME encoding leads to 1 KB WAV files !
3. I opened the MP3 files with Adobe Audition 1.5 (based on Cool Edit Pro) and then took screenshots with an older freeware version of FastStone Capture.
4. I copied the screenshots to good old Picture Publisher 10 and used the automatic masking tool to adjust waveform colors and to cut out the waveform of the encoded MP3 files and pasted these to the screenshot of the original WAV.
5. Of course it is important that you paste at the exact X/Y pixel positions.
But it is relative easy when you look at the spikes of the waveforms to merge it perfectly.That’s it.
Of course you cannot judge the quality only based on these waveforms.
But the visual impression of the waveform comparisons matches actually the results of my (and Jesse’s) blind ABX (listening) testings.I had (and have) not the time now to do more waveform comparisons, but I remember that my tests last year lead to the conclusion that even the usage of low VBR modes (matching 128 kbps) with newer LAME versions result to inferior MP3 files than you can create with LAME 3.93.1 @ CBR 128 standard encoding.
So, no doubt, for 128 kbps MP3 streaming or files for your mobile MP3 player LAME 3.93.1 is the best choice.
It is absolutely astonishing what high grade of (near) transparency you can create with LAME 3.93.1 and (even already with) standard CBR128.
Last year I have tried almost every LAME build (all official versions from 3.90 to 3.98.2) and 3.93.1 is definitive the best version for CBR128.
IMPORTANT:
You have to use LAME 3.93.1 !!!
Do not use LAME 3.93 (because it has severe CBR and other bugs) !!![quote author=”Audio”]I was wondering one thing… in one of your previous (and since delete posts) you mentioned that nothing beats 3.93.1 for contemporary music. What would you say are the major differences between 3.93.1 and 3.98.4? We are currently using 3.98.4 and I was wondering in which ways 3.93.1 would improve our sound…[/quote]
[quote author=”JesseG”]I did extensive blind ABX testing to find that 3.93.1 seems to sound the closest to the original input, at 128kbps cbr. This was done when 3.97 was out, so it’s possible that 3.98.x would beat 3.93.1, but I highly doubt it.[/quote]
➡
viewtopic.PHP?f=5&t=1749[quote author=”camclone”]LOUD and at the same time … AS CLEAN AS POSSIBLE[/quote]Yes, you can !
But even when it is better (louder and cleaner than the big boxes in the market) than it is still not good.[quote author=”camclone”]and please listen for more than 5 minutes[/quote]LOL
That’s the point.
It is as if you can drive with your car faster (without extra danger of damages) than others.
But is it really an advantage to drive constanly with 250 kmph instead of 200 kmph ?
Ok, you are faster (=louder).
But it is fatigueing and boring, too.Why is it fun to drive a car ?
➡ To pick up speed !
But you cannot pick up speed when you constantly drive with maximal speed.
So what acceleration/slowing down is for driving a car, that is dynamic range for music.[quote author=”Soumya”]Well leaving the DSP running when I am not around or am sleeping would not quite serve my purpose as I won’t have 100 five minute discrete tracks rather a single gapless track[/quote]This is no problem.
You need a cue sheet to automatically cut the resulting big file (for MP3 you can use mp3directcut) in small pieces.
I have just found a free tool that can convert playlists (.m3u) to cue sheets (and vice versa):
MusicList (or Music Lister or Music List Converter or Music List Manager – obvious the author is not sure what the correct name is of his own program !?)
http://www.audiography.com.au/Software/Downloads.htmSo the only thing you have to do manually is to cut the exact starting and the exact end point of this big file with mp3directcut (for other formats you need other cutting tools of course) to match the exact original playtime (duration called in foobar2000) of the whole transcoded playlist !
[quote author=”mp3directcut manual”]When loading a cue sheet the program reads titles and artist names and shows them in the graph area. If you split a file by using a Cue sheet you can create filenames with titles and ID3v1.1 tags for each file.[/quote]
Besides mp3directcut there are some other free tools to cut big files into small pieces based on cue sheet information.[quote author=”Soumya”]Another thing, is there any way for performing automation for encoding tracks in mass, rather than playing them in realtime.[/quote]No way !
This would be another product (Breakaway Batch – but this is Breakaway Live = realtime).
You are not the first one with this wish of batch transcoding processing.
Let’s wait for the next V1 release, maybe we will get something like this fictive Breakaway Batch.
In other discussions Leif has said that he thinks about this request, but no definitive yes or no when or if at all this will be available.Batch processing is something Breakaway does not support (maybe Leif has his own private tool ❓ , but nothing for the public).
So you have to think about alternative Winamp DSPs for this batch purpose.
You won’t have your favourite Breakaway sound, but maybe you can find something else (a DSP) that comes close.
The whole live is a compromise.
I won’t go in details here about alternative DSPs, because this is the Breakaway product forum.
Though realtime is slow when you are used to work with batch processing, you have to change your perspective.
You can transcode music tracks with a running time of more than 16 hours every day (while you sleep and while you work) in realtime !@ Jesse and celar
Your suggestions about blind ABX tests and recording the different outputs are very rational.
I will do some bit by bit comparisons and spectral analysis.
In 2008 I have done this for VAC 4.08, Total Recorder’s old user-mode and the newer kernel-mode virtual device driver.
And surprisingly Total Recorder’s old user-mode driver was the only of these 3 to record a WAV bit true.
On Total Recorder’s homepage you can read:
"The system mixes audio streams being played before it arrives in the kernel-mode driver. Audio streams being played can only be recorded in mixed form."
This explains why the newer kernel-mode virtual device driver cannot record bit true.
But on VAC homepage you can read:
"All transfers are made digitally, providing NO sound quality loss (a bitperfect streaming)."
http://software.muzychenko.net/eng/vac.htm
Maybe I was too dumb to set the VAC driver properly inside Total Recorder, but all my tests lead to the conclusion, that the usage of VAC is not "a bitperfect streaming" connection, because it uses the same kernel system as Total Recorder’s virtual device ("Audio streams being played can only be recorded in mixed form." TR homepage).Anyway.
Even if ABX tests and/or bit by bit comparisons will show that there is no difference and everything is based on subjective phenomena than it is still there.
Subjectivity is objective (Woody Allen in "The Last Night Of Boris Gruschenko") !
So all denying ABX tests are more or less irrelevant.
Human brains are not bit true calculating computers.
Everybody knows (you don’t have to be a neuro scientist) that different colours have different effects to the subjective feelings of different human beings.
It makes a difference e.g. if you are using a blue or red skin for your player.
It is irrelevant to compare (to record streams) and to prove that it is always the
same audio (bit true independent if you use a red or a blue skin).
You can hear the difference (that does not exist in the audio stream) !
It is same thing as to love somebody desperately.
You can explain that it is all based on biological, chemical reactions inside your body.
But does this scientific knowledge help you to control your feelings ?
No !@ timmywa
Sonique 1.96 does not support Winamp DSPs, but this is irrelevant, because you can use Breakaway Live 0.90.96b (even an unregistred trial in bypassed mode !) to host Winamp DSPs (Sonique using Breakaway Pipeline as audio output)!
Only V7 ?
Then it must be buggy.We need a V8 or a V12 release !
Breakaway V8 with wide base tyres !
[quote author=”syamkumar”]I expect valuable information from Leif also….[/quote]Good luck !
➡
search.php?keywords=&terms=all&author=Leif&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search
…busy, busy, busy…..
It seems as even Pope Benedict XVI stops by here more often during the last months.
But IMO it is better that he (Leif – not the Pope) spend his limited time to finish some very important things (for the final V1 releases and other business things) instead of wasting it here.[quote author=”Orio”]and the Final way is… is to listen to it, from inside the brain-heart of the Artist : you posses the Artist to listen to his Imagination[/quote]Unbelievable !
If you had not already written this idea, it would have been my next level of response.
Great, we understand each other in a perfect way. 😀Now I leave this damn’ internet (plastic world) for today and will read some scores of Bach, Mozart and Beethoven (the best sound you can get).
But tommorow I will try to feed these scores (the sheets !) into VL and enhance these with a sublime touch to the sound from FM Magic preset !Have a nice day !
