Is there a difference between the two? Which one is better? This is probably gonna sound stupid but I'm curious. :/
+ Reply to Thread
Results 1 to 30 of 38
If it is done correctly...
More channels isn't necessarily better. If the source was stereo adding fake rear channels doesn't improve it.
In the old days, we had "Quad", which introduced front and rear fading into a stereo system. We thought that was a tremendous improvement.
"Dolby" was never popular back then because hiss lives in the same frequencies as high hats, tambs, shakers, etc. So your music gets a haircut in that area.
To me, Surround Sound is like 3D. Yeah, it's more "in your face", but meh, it's kind of blingy too, and gets on the nerves when overdone.
A good meatloaf is defined by it's basic attributes like hamburg-to-breadcrumbs ratio and tangy/sweet bark. When you try to bling it up, it's rarely better. Same with music. When the bling stands out, it's just bling, and gratuitous bling doesn't fly on its own.
Swirling helicopter blades are about as enjoyable as a trash truck emptying a dumpster and panning the dump high/low/left/right.
What's my friggin point? Yeah, I like EDM too. hahaha
Last edited by budwzr; 25th Jul 2013 at 10:44.
MeGUI says Downmix and Wiki says Upmix. I thought maybe Dolby Pro Logic II will make it sound better but guess stereo is the best I'll get. ...but Dolby says
Dolby Pro Logic II transforms movie soundtracks and stereo music into 5.1 channels of surround sound.
If you have multiple channels to start with (studios often record each instrument/voice/effect on a separate track so they can be placed wherever they want them) the effect works as intended. But if you only have a mono or stereo track to start with, simulating those extra channels doesn't necessarily improve the overall sound.
I think,therefore i am a hamster.
I've never thought about the Dolby Pro Logic downmixing option before so I thought I'd try it using MeGUI. So far the result isn't great but I'm not sure if it's an encoding or decoding issue. Anyone know if ffdshow's Dolby Pro Logic filter is any good?
I downmixed a dts file to Pro Logic II while converting it to flac with MeGUI. I then used MeGUI to convert the flac Pro Logic file to a multichannel flac file and ffdshow to apply Prologic decoding and output 5.1ch. Finally I loaded the 5.1ch flac file into Audacity and listened to the channels individually while comparing them to the original dts. The surround channels were a mess compared to the original (I compared sections where the surround channels contained music).
The object of the exercise was to downmix to a lossy codec using Prologic II and decode that to a multichannel wave file to see how much the lossy encoding messed with the Prologic surround sound, but before I can do that I've got to find a lossless method which works..... or is Dolby Prologic II just really poor quality begin with? After encoding/decoding and listening to the surround channels individually they sounded like they'd had a phasing effect applied to them. And a lot of it.
If Pro Logic II should produce reasonable sounding surround sound I could try a different encoding method but I'm not sure of another way to decode that back to a multichannel file, aside from using DirectShow and ffdshow. If my test is as good as Prologic II gets, I wouldn't bother with it.
I have a PLII receiver and it sounds fine to me. The way I watch movies, it decodes to stereo to send over SPDIF then the receiver gets it and can matrix it into surround using PLII.
I never thought it sounded bad. But when I connect to the one upstairs that is a bit better of a receiver, it can decode DD 5.1 and DTS 5.1 and passthough via SPDIF to the receiver sounds better than decoding to stereo and then using PLII to rebuild the surround.
Last edited by hogger129; 3rd Aug 2013 at 22:35.
Originally Posted by hogger129
Pro Logic 2 simulates the full 5.1 experience.
I don't have a pro logic 2 amp so I can't comment on how well it works.
But yes as you point out hogger129 using the original 5.1 track is best when available. But there is nothing wrong with playing with virtual surround sound for non 5.1 sources. Though they may tend to sound hollow and "fake" at best compared to a properly produced "real" 5.1 track.Donatello - The Shredder? Michelangelo - Maybe all that hardware is for making coleslaw?
I think,therefore i am a hamster.
So I assume it's because 5.1 has metadata that tells it how to decode rather than PLII where it estimates to simulate surround sound, that 5.1 sounds better?
I have a Phillips/Magnavox receiver that has PLII (wish it had DTS also), but I guess I could never tell a huge difference. My speakers are not exactly high end so that could be why I can't tell the difference.
I think when I get out into my own place soon here, I am going to track down a nice Pioneer receiver and some decent speakers for my movie/music setup.
There is no such thing as Dolby Pro Logic or Dolby Pro Logic II encoding. They are decoding circuits, that rely on certain frequency, phase, and timing cues in the decoder to steer appropriate content to the created rear and center front channels. These decoders have processor-driven logic to create a convincing 4 channels out of 2 original ones on the DVD, CD, VHS tape, etc. Note that, in math you can't solve for 4 unknowns out of merely 2 variables, but as far, as Dolby Pro Logic is concerned, with enough cues, convincing matrixes can be made out of the two channels.
This all started out as 4 analogue channels in the master (Lf, Rf, Cf, rear), and they were matrix encoded as Dolby Surround for playback media. I don't know if ffdshow or MeGUI or whatnot that can do the encoding is clear about this.
So it's silly talking about DPL II encoding this and that when clearly it is not (even if so labeled and causes confusion); a choice should be offered only as Dolby Surround encoding for 4 original tracks available in your source or NLE. Then on decoding, this is where choices can be presented for Dolby Pro Logic/II, etc.
Encoding for Dolby Surround is fairly straightforward it can even be done with Audacity if you have the 4 original channels in an NLE situation; a search on the web can give the equations. Basically, the Cf channel is lowered by -3dB, then mixed equally with Lf and Rf.
The rear channel is low-passed at 7KHz, given a certain delay (I don't know the precise millisec figure), split into two, one the original and the other phase-inverted, and each also mixed with the Lf and Rf. So basically composite left = original Lf+0.707Cf+rear, and composite right is = original Rf+0.707Cf-rear.
All this mangling is now to be decoded by your favorite wretched Dolby Pro Logic decoder, h/w or s/w, to partially (but convincingly) recover the four original Lf, Cf, Rf, and rear. So it's basically true, that since the gumption is in the decoding, there can be better, best, and just so-so Dolby Pro Logic/II decoders.
Occasions rose where unencoded conventional stereo audio would be forced through a DPL decoder, and the result is just absolute pits. Since a lot of energy in an ordinary music track (like low frequencies) is mono, and therefore equally present in the L and R channels, a DPL decoder would assume they were part of the center channel being attempted to decode. This often resulted in a lopsided situation where the center channel would be blaring away on playback, but almost nothing from L, R, and rear. To correct this, sophisticated formulas and complex processor circuitry were yet cooked up to give us the next version Dolby Pro Logic II, so that whatever played out from the four channels sounded mostly equitable, whether or they were duly-encoded Dolby Surround tracks, or just plain jane stereo music tracks.
Of course, this was all decades ago and is now largely mute. We don't have to worry about matrixing and shit anymore because discrete-channel Dolby Digital and DTS (and their HD counterparts) are here.For the nth time, with the possible exception of certain Intel processors, I don't have/ever owned anything whose name starts with "i".
I think,therefore i am a hamster.
Donatello - The Shredder? Michelangelo - Maybe all that hardware is for making coleslaw?
Handbrake and others, IMHO, add to the confusion by giving a choice for DPL II encoding.
Ancient Dolby Surround is an analogue encoding scheme of the 80s and 90s that attempts to put 4 discrete channels into 2, and there is only one way to do that scheme as Dolby implements. It can also be decoded in the analogue domain with a Dolby Surround decoder. To heighten perceived channel separation, etc, Dolby Pro Logic and Dolby Pro Logic II decoding schemes were developed. The Logic part of that moniker is the processor circuitry that minutely detects the differences in the channels; for example, if there is a difference if, say at least 3dB between the Cf channel and any of the others, then gain in Cf is increased even more and all the others decreased, all on the fly.
In the original setting, an encoder would have the 4 discrete audio channels with which to encode into a Dolby Surround 2-ch track, with a view to it being passed later through a Dolby Pro Logic/II to retrieve imperfect but passable versions of those 4 original audio channels. One thing I infer from the OP of this thread, and of what Handbrake and others purport to achieve, is that they have a conventional stereo 2-ch track, and of how it can be treated so that deliberately passing it later through a DPL II decoder can make it sound, well, "better".
It's a bit circuitous but I can imagine, the 1st steps would probably be to wring out a mono track from those two channels (to make the Cf track), then an out of phase track (for rear), apply appropriate levels, frequency response characteristics, and delays, then remix the lot back into two using a Dolby Surround scheme. This may or may not be how Handbrake et al is achieving it. If it is, IMHO it's just silly and needless because genuine DPL/II decoders, at least those who adhere to specs
and more, are sophisticated enough to know when a conventional (non-Dolby Surround encoded) stereo track is being fed to it, and therefore produce the four channels "better" than what older Dolby Surround and Dolby Pro Logic encoders could with the same.For the nth time, with the possible exception of certain Intel processors, I don't have/ever owned anything whose name starts with "i".
I assume the OP was referring to taking multichannel audio and downmixing it to stereo/Prologic. I can't speak for all conversion programs but MeGUI does have options to "upmix" stereo audio to 5.1ch using SOX, but I assume it "upmixes" to discreet 5.1ch and not Prologic.
Anyway I was a little curious to see how MeGUI downmixes 5.1ch to Prologic II so I ran a quick encode. It uses AVISynth for the work so I copied what I assume is the relevant part of the script:
# 5.1 Channels L,R,C,LFE,SL,SR -> Dolby ProLogic II
function c6_dpl2(clip a)
fl = GetChannel(a, 1)
fr = GetChannel(a, 2)
fc = GetChannel(a, 3)
sl = GetChannel(a, 5)
sr = GetChannel(a, 6)
ssl = MixAudio(sl, sr, 0.2818, 0.1627)
ssr = MixAudio(sl, sr, -0.1627, -0.2818)
fl_fc = MixAudio(fl, fc, 0.3254, 0.2301)
fr_fc = MixAudio(fr, fc, 0.3254, 0.2301)
l = MixAudio(fl_fc, ssl, 1.0, 1.0)
r = MixAudio(fr_fc, ssr, 1.0, 1.0)
return MergeChannels(l, r)
I understand the gist of what you're saying regarding decoding circuits relying on certain frequency, phase, and timing cues to decode the audio properly, but are those things only ever found in genuine Dolby-created Prologic audio?... As opposed to a Prologic II downmix such as the one above. Could the above downmix ever be decoded "properly" by a Prologic II decoder? I assume from what you're saying the answer is no, but I'd like to understand the process a little better.
The problem is you can't have four full bandwidth channels encoded simultaneously. For example, you can't have four different mono songs playing simultaneously and have each come out of only one speaker.
Last edited by jagabo; 9th Aug 2013 at 08:30.
I'm not sure if it's ffdshow's Prologic decoder or my convoluted method for converting your above file to discreet multichannel audio, but anyway.....
I can't seem to decode a wave file using ffdshow no matter how much I beg, and even if I convert it to flac first, getting MeGUI to use DirectShow to decode it doesn't seem to want to happen. For the moment I've converted it to flac, stuck it in an mka and then used a DirectShow script to convert it to multichannel wave with ffdshow decoding.... that at least works. The audio is clear but there's a lot of bleed.....
Here's what the front stereo and centre channels looked like after I decoded that to a multichannel wave file with ffdshow's Prologic decoder (the other channels were there, just not in the screenshot). If you recall the order (FL, C, FR, RL, RR), you'll see the front left and right channels are still well separated, but centre bled into left and right and a lot of rear left and right ended up in the front channels too.
Likewise there was ample front left and right in the rear channels, although the centre channel kept out of the rear.
Is that sort of thing as good as it gets?
I'll try to work on a better way to decode/re-encode or maybe ffdshow's Prologic decoder isn't terribly good......
It worked as expected on my audio receiver (four discreet channels) with the receiver set to Pro Logic mode.
I suppose, it would make more sense to monitor waveforms as the soundtrack plays and record them at the output of a DPL/II decoder to quantify what is going on. Like, resolving for mono content (speech, for example) will still make it appear equally on both Lf and Rf through an analogue Dolby Surround, but for a noteworthy Dolby Pro Logic/II design, I expect Lf and Rf to be greatly attenuated (>-10dB), and center channel boosted (>3dB).
I have little idea what handbrake and ffdshow's take on all this is.For the nth time, with the possible exception of certain Intel processors, I don't have/ever owned anything whose name starts with "i".
Dolby pro logic are matrix encoded and then decoded,that's like saying dts aren't encoded but decoding schemes,how are dolby pro logic created in the first place?I think,therefore i am a hamster.
I'll play around again sometime in the next day or so.... although unless ffdshow's Prologic decoder does what it's supposed to do there's probably not much point in my messing around as that's the only way I can decode it..... Proloically.
The sample you uploaded jagabo..... That's supposed to be decoded using Prologic II and not Prologic? You said the rear channels are mono so maybe I can't decode it properly.
I've re-encoded countless multichannel audio streams as Prologic in the past.... that's what AutoGK does when it converts to MP3 and it doesn't let you disable Prologic (probably no point). I've never decoded one of those encodes using Prologic though so I'm still curious regarding the encoding process. Does encoding with a lossy encoder (ie MP3) mess with Prologic much?
Anyway.... does anyone have any experience using ffdshow's Prologic II decoder.... ie does it work correctly? I'm not sure trying to find a more direct way to decode/re-encode the sample wouldn't be a waste of time until I know.
Originally Posted by hello_hello
if its just a curiosity than go for it.
However it might be more worth it in the long run to get a surround amp that can decode aac 5.1 properly. That should be the maximum savings you could get over ac3 and maintain discrete 5.1 surround.
Fyi if you have a tablet with hdmi out that can run xbmc you already have a way to properly play aac 5.1. I can do that on my ouya console that runs android and xbmc. It reads as ac3 with full surround sound (ie the blue light is on on my sony amp and its more than ten years old and is sd only with no hdmi in).
And fyi I run the hdmi from the ouya to my hdmi in on my hdtv and the optic out to my receiver from the hdtv and I get full surround sound. My hdtv actually passes the surround sound through to the amp.Donatello - The Shredder? Michelangelo - Maybe all that hardware is for making coleslaw?
Last edited by jagabo; 10th Aug 2013 at 09:57.
A curiosity thing definitely, as I don't like surround sound. I'd offer my spiel on how I find surround sound annoying but my spiel occasionally has the same effect on posters that surround sound does on me.
ffdshow audio decoder can do the same. Not that I've tried it myself.... only having 2 speakers and an inability to understand why the majority of the world thinks surround sound is such a wonderful thing.
People are strange creatures..... half of them will swear black and blue every MP3 audio file ever encoded sounds like crap while the a similar number will surround themselves with speakers and swear black and blue it sounds better.
But that's just me of course.... each to their own.
Not to mention the time I tried a similar test using surround sound which contained music in the rear channels. I thought it sounded terrible after decoding, but maybe that was ffdshow's fault.
Analogue domain matrixing decoding: Dolby Surround
Analogue domain matrixing decoding with steering and individual channel level control through h/w logic (proprietary circuitry depending on manufacturer, with, if at its most basic can decode Dolby Surround as is, then permitted to use logo): Dolby Pro Logic
Analogue domain matrixing decoding with steering and individual channel level control through h/w logic with different emphasis on center and rear for reproducing conventional non-Dolby Surround encoded stereo music tracks: Dolby Pro Logic II
Digital domain discrete (non-matrixed) 5-channel compressed lossy encoding & decoding: Dolby Digital & dts
Digital domain discrete 5-channel (or more) lossless encoding & decoding for HD versions of above: Dolby TrueHD & dts-HD Master Audio