Welcome to the Invelos forums. Please read the forum rules before posting.

Read access to our public forums is open to everyone. To post messages, a free registration is required.

If you have an Invelos account, sign in to post.

Invelos Forums->Posts by mediadogg Page: 1 2 3 ...9  Previous   Next
Message Details
Quoting mediadogg:
Quote:
Quoting MehLeh:
Quote:
Hello,

I do not know anybody in the US. Sadly.

Now you do.

No response here or to my PM. Probably a troll.
Posted:
Topic Replies: 7, Topic Views: 391
Ok, I am finding mostly good correspondence, but I just hit one that will make it relatively easy for me to sort out bugs. The IFO scan shows all tracks and languages, dead equal and the same as the profile in DVDP. But my GUI drop downs are blank. Clearly a bug there. But it shows that the hard part is correct - the IFO parsing. The mono channels, in this case, were Commentary. Little Miss Sunshine, 024543403319, US.
Posted:
Topic Replies: 47, Topic Views: 379
Never mind everybody regarding examples. The nose on my face got in the way of the obvious.

A simple filter search on my own collection, identified a whole bunch of DVDs and BDs with MONO audio tracks. I will just run a few through DVDPca and compare what it finds to the existing DVDP profiles. If I find a difference, I wil examine the code and the actual IFO bits.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
For what it's worth: All discs I looked at, whether 1 or 2 channels, were identified as AC3 by DVDPca. But that was only maybe half a dozen discs, so that could be just coincidence.

Barring a good way to display channel difference (l-r), it still seems to me that an educated guess is the best way to identify 2-channel mono. Not sure how well that would fare with voters/scanners though.


Quote:
I was going to say that I do have several 1 channel discs, bur the problem would be to find them.
But now that AiAustria has provided an example maybe I don't have to?
If you need further examples, mediadogg, I can certainly take a look.

Not sure how to interpret this. If you used some 1 channel discs, then you obviously could find them.  What did I miss?
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
I was going to say that I do have several 1 channel discs, bur the problem would be to find them.
But now that AiAustria has provided an example maybe I don't have to?
If you need further examples, mediadogg, I can certainly take a look.

Thanks, let me start with that one. But if DVDPca already reports mono, I don't need to ...

Note to AiAustria, Thanks, but I need all the IFOs. Is that the only one?

I'll take a look anyway, but it would be interesting also to know: what does a DVD player say?

I just looked at the profile. Both tracks are "Dolby Digital - Mono".

I will be very eager to see the actual bits in the metadata! 
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
For what it's worth: All discs I looked at, whether 1 or 2 channels, were identified as AC3 by DVDPca. But that was only maybe half a dozen discs, so that could be just coincidence.

Barring a good way to display channel difference (l-r), it still seems to me that an educated guess is the best way to identify 2-channel mono. Not sure how well that would fare with voters/scanners though.

I would only be concerned if DVDPca produced different metadata than other trusted tools. If you find such a case, then let me know, and send me the IFOs.

Also, you have actual 1 channel discs? Wouldn't mind taking a look at those IFOs as well. Thanks.

In terms of displaying (L-R), a simple hardware circuit will do it. I have an old Radio Shack voice remover that was used for Karaoke, if you want to play with it. Pay for shipping and you can borrow it.

Or, I could build you a little custom circuit that works with headphones. let me know.

But again, if during mastering, the mono signals are passed thru a Dolby encoder, there is pseudo surround info added to the 2 channels. Check the Wiki for Dolby for a full discussion. So then you go from M->L and M->R, to stereo AC3 Lt and Rt, where Lt and Rt are the augmented pseudo surround channels. In that case, I think the subtraction trick still works, but I am not 100% sure.
Posted:
Topic Replies: 47, Topic Views: 379
Gunnar, I have to issue a retraction. DVDPca, as I think you know, is an augmented mashup of a bunch of other programs, such as DVDInfo, BDInfo, two different IFO scanners, two different movies players, and a bunch of other snippets. During the testing, which a bunch of people contributed to, I experimented with various mixes and matches, fixed some bugs and added some new features. I stopped when I was getting mostly reports that the metadata matched exactly other tools such as PowerDVD, and running times were computed exactly. Extra features included extracting other metadata from the IFO files and off the disk.

As a result, when I looked at the code the other day, I posted a snippet containing "DualMono". That code, as far as I can tell, will never be hit in the current version of DVDPca, whose code is a jumbled mess.

DVDPca, for DVDs will only designate "Mono", if there is only 1 channel. period.

I am currently looking at some updates, and I might experiment with at least putting in the logic that is capable of outputting "Dual Mono." Whether there is a DVD out there, with metadata that will trigger the code, is anybody's guess.

By the way, the IFOs you sent, when parsed, identifies an AC3 codec (Dolby) and 2 channels.

I was trying to find an analogy for what we are dealing with. This is as close as I could come:
Quote:

A car is registered as a 5 passenger car, if it can legally carry a maximum of 5 passengers. It retains that designation, no matter how many people are actually traveling.

However, in a "high occupancy" lane, you are required to have at least 2 passengers, otherwise get a ticket from police.


There is no policeman that prevents, while mastering the DVD, two identical mono sources from being passed to the Dolby AC3 encoder as a "stereo" signal. You get pseudo surround, unless you turn off the Dolby processing of the receiver. In this case, if the cover says "Mono", then depending on your point of view, both the cover and the metadata are correct.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
Quoting mediadogg:
Quote:
The "it" you are talking about is the amalgamation of the codec being used, and your receiver settings, including manual adjustments (e.g. "balance", or a "mono" switch, or down-mix settings and the list goes on).

That may well be so, but the bottom line is - is listening to the center speaker vs the left/right ones a reliable way to determine if it's mono or stereo. And my understanding is that it is not, if you're looking for a quick determination.

Correct. Although there is a "DIRECT" setting that some A/V receivers have that i think turns off the codecs. That might be a way ...
Posted:
Topic Replies: 47, Topic Views: 379
The "it" you are talking about is the amalgamation of the codec being used, and your receiver settings, including manual adjustments (e.g. "balance", or a "mono" switch, or down-mix settings and the list goes on).

Based on all the discussions, I'm coming to a very simple way of thinking about it. If there is one channel in the audio stream, the source is mono. If there are 2 channels, the source is stereo (seems to borne out by the code I am looking at - that Dual Mono code seems never to be hit). And so forth.

Now, what you HEAR in which speaker is totally dependent on setup , configuration, and actual channel content.
Posted:
Topic Replies: 47, Topic Views: 379
I think we are getting some (admittedly very confusing) things conflated.

1. The encoding and processing of the actual bit stream representing the audio that is sent to the speakers.
2. The encoding and processing of the bit streams that represent the metadata headers that describe the format of the contents.

I think they are almost entirely independent. The digital processing that creates, say Dolby, don't require any knowledge derived from metadata to know what to do.

And for another example, the process that creates ATMOS effects works regardless of whether the audio bit streams are coming from a disc or from the network.. All of the necessary content headers and metadata are encoded within the stream, to tell the codecs in the Xbox or A/V receiver what to do.

The metadata we have been talking about is different, albeit encoded in a complex bit stream on the DVD or BD, this data is not actually a part of the audio stream itself, and does not have any relevance to the actual codec processing - it is information for humans (and programs such as players) only.

Now that is the way I understand it to work. There may be other ways for it to work. If someone has actual technical knowledge or experience that would inform me better on this, it would be appreciated. Otherwise, any other reasonable conjecture would be just that - and I am pretty comfortable with my own, semi-informed version.
Posted:
Topic Replies: 47, Topic Views: 379
Thanks. Your question is the type of thing I would be looking for in the metadata. The code is really detailed - parsing bits from the IFO headers, shifting bits, etc. I suspect that since "everybody" makes the same mistake, there is no way to catch errors in the metadata. Even the IFO bit map I referenced has question marks in some places by the authors. But I don't have a better source for that bit map.

By the way, I had fun playing with the circuit I mentioned to you. I agree you shouldn't try it. While listening to a stereo FM station, voice goes dead during a commercial, or when the receiver is switched to MONO. That reminded me that most Karaoke machines have a "voice or vocal removal" function. Basically, it subtracts out the common information between the stereo channels. So, in the case of true mono, you get silence. For a center imaged singer, you get the music without the vocals - great for Karaoke. Some apps have this feature, and in some versions of windows, you can actually set the Speaker/Headphone properties to include voice cancellation. That would be a super simple way for you to do what you want to do - play a DVD and instantly check for dual mono. My version of windows and the apps I have installed somehow killed that option. Maybe you still have it enabled.

I also read that Audacity has the option, but I have not tried it yet.

Oh, and I fundamentally agree with you and T!M. I was just asking the philosophical question: whether if in this case, the metadata is in fact the data. Do you trust the metadata when it says ATMOS or DTS, etc., or do you go and play the DVD to see what is actually being delivered to the speakers? It is that exact same IFO bit map that is supplying that information (different structures for BD, but same principle).
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
By the way, wouldn't a player like PowerDVD also use the metadata to display its audio information? Isn't that why we see so many profiles with 2-channel stereo where it ought to be mono; people enter what the player tells them?

Correct!!!!

So, the question remains, what does it mean to (correctly) profile a DVD? To me, that means recording what the DVD metadata says. Just as we believe what we see as the credits roll, even if the rolling credits have been created in error (nobody goes and interviews the original staff to validate what has been painted on the screen for credits.). So why should we have to listen to audio tracks? I'm just saying ...

Anyway, please post a few UPCs and give me a link to the IFO files that you have for DVDs that are known to have Dual Audio, and I will examine the metadata. There are other places in DVDInfo where the correct info is deduced by comparing related, but sometimes conflicting metadata. Maybe we can do  the same for this.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
By the way, I took a few titles that had been profiled as 2-channel stereo, but that I knew would be mono (and at least one actually said mono on the back cover), and I tested them with DVDPca. It reported them as 2-channel stereo. So if that info comes from the metadata, then the metadata is obviously not reliable.

Ok, cool. Let me get those profile IDs from you, and when I update DVDPca, I will double check and make sure that the error is indeed in the metadata, and not the code. Just FYI, here is the IFO bit map that I use to compare with the code I stole from DVDInfo.
Posted:
Topic Replies: 47, Topic Views: 379
Ah, OK.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
By the way, I just found out that VLC player actually can display waveforms. And I agree that it is impossible to say that ttwo waveforms are identical and therefore determine that it must be mono. But it is possible in many cases that waveforms are sufficiently different that you can tell with a fair degree of certainty that it is stereo.

Not sure how useful it is, though, since you would have to look for places where the tracks really are quite different.

You have got to be kidding! But if you really want to try using the waveforms, just wire a headphone plug to bridge across the L and R signal wires, ignoring the ground. Plug into the headphone jack of your amp, and you will hear the L-R content. If you hear silence, then both channels have the identical instantaneous content. This method should be electrically safe and offer no harm to you or your equipment, but to be super safe, you could add a small resistor 10 - 100 ohm in series.

The program Audacity can do just about anything with a waveform, including aligning the L and R channels horizontally next to each other. Very useful program for many purposes. I think you would enjoy it.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
Quoting mediadogg:
Quote:
how about check what is on the actual DVD

Eh? You've got me confused now. I thought I covered metadata ("would the mastering engineer really care enough to create different metadata for mono and for stereo?"). Or are you talking about something else?

I don't understand your point at all. My only point was (discounting the motives of the recording engineer), is that the IFO file on a DVD has a spot to encode Dual Mono, so any program designed to read that metadata on the DVD could be used to check it. Of course, it might not be there, for whatever reason.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting MehLeh:
Quote:
Hello,

I do not know anybody in the US. Sadly.

Now you do.
Posted:
Topic Replies: 7, Topic Views: 391
My point that you forgot: how about check what is on the actual DVD (not the cover)? There is a placeholder in the specifications for somebody to tell you "Dual Mono". But there is no requirement for them to do so. If they used it, DVDPca and DVDInfo will display it. In those cases, no need to guess. Hmm, well I don't know if it is still working. Been a long time since the last update. Sorry about that. Guess I should do an overhaul and see what happens.
Posted:
Topic Replies: 47, Topic Views: 379
Yes, I found code in DVDPca/DVDInfo that handles the "AudioMode" descriptor on the DVD, which can have a value of "Dual Mono" for a channel count of 2. My understanding of Dual Mono, is that the same signal is designated for the both left and right channels.

                if (ChannelCount == 2)
                {
                    switch (AudioMode)
                    {
                        case TSAudioMode.DualMono:
                            description += " / Dual Mono";
                            break;

                        case TSAudioMode.Surround:
                            description += " / Dolby Surround";
                            break;
                    }
                }

Then there is "ChannelLayout" descriptor. Here is the code for that:
                  switch (ChannelLayout)
                    {
                        case TSChannelLayout.CHANNELLAYOUT_MONO:  // Send to center channel speaker.
                            description += "1.0";
                            break;
                        case TSChannelLayout.CHANNELLAYOUT_STEREO:
                            description += "2.0";
                            break;
                        case TSChannelLayout.CHANNELLAYOUT_MULTI:
                            description += "5.1";
                            break;
                    }

Now, you ask: is there such a thing as channel layout of "mono" for channel count = 2? Here is that code:

                if (ChannelLayout == TSChannelLayout.CHANNELLAYOUT_MONO &&
                    ChannelCount == 2)
                {
                }

Seems like even the original coders of DVDInfo punted on that one! 

I would encourage you to try DVDPca if you want to see what is encoded on the DVD. Back when I was testing, I think it was found to be very accurate.

By the way, just to be clear, the "encoding" I am talking about is the encoding of descriptive information (metadata) in the IFO files on the DVD, not the audio stream encoding.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting GSyren:
Quote:
Surely, the encoding doesn't differentiate between 2-channel mono and 2-channel stereo? It's the content that differs, not the encoding. ....

I think in fact it does. I will double check.

And as far as contributing, I can't imagine that Ken envisioned the community actually monitoring wave forms to make a contribution. Take the word of the recording engineer and call it a day. Geez, we can't even agree on the spelling of a name, or how to handle Asian names.       
Posted:
Topic Replies: 47, Topic Views: 379
Well it is a little bit what everybody said.

There is not necessarily a 1-1 correspondence between:

(1) What is actually on the DVD audio tracks
(2) What is said to be on the tracks (cover info, track info encoded on the DVD)
(3) What signal is actually sent to each speaker (see iPasta comments)
(4) What you can see by viewing the signals in the time domain

In fact:

(1) Only the recording engineer knows
(2) Can be wrong on the cover, or even on the DVD. Software such as DVDInfo (which DVDca is derived from), actually reads the encoding from the track info on the DVD. This information would likely have been supplied by the recording engineer.
(3) Depends on your home theater wiring and receiver settings.
(4) It would be virtually impossible to visually compare two audio waveforms in real time (I mean with your eyes - yes, software and/or hardware that can sample the signal can do it.)

An example of (3): You could incorrectly determine that a signal was "mono" if an error in your wiring sent both left and right channels the same signal (say the left channel), or if there is a "mono" setting on the receiver, artificially mixes a stereo signal down to mono by adding the two channels, and sending the sum to both speakers.

If you want a simple way to tell whether left and right signals coming from a stereo output are exactly the same, you can subtract them. If you have simple electronics skills, this is easy to do. Probably a ton of circuits on Google. If Left and Right are EXACTLY the same in amplitude at every instant in time, you will hear dead silence when the signals are subtracted ( a waveform will show a flat line). Theoretically it would be possible to do this in software while the DVD was playing, as Gunnar suggests.

Legacy receivers used to have a switch for this, and this method was sometimes used to synthesize pseudo surround effects; e.g,,  Left surround got (L-R), right surround got (R-L), and center got (L+R) in cheap equipment.

Ah, I almost forgot: the real issue is: what do the rules require that you submit? I think it is (2) - what is encoded on the DVD.
Posted:
Topic Replies: 47, Topic Views: 379
Quoting MehLeh:
Quote:
Thank you but that page says "unavailable".

Here is an idea: Is there any way to repay somebody in USA for a "Gift Voucher"?

If so, then you could pay that person (maybe you have a friend in USA), and then they would purchase a "Gift" voucher for DVD Profiler. Then, when you redeem the voucher, you would have a subscription.

Could that work for you?
Posted:
Topic Replies: 7, Topic Views: 391
I am a big fan of Kpop / Jpop

Here are a couple of samples:

Ailee: Song stylist right up there with Beyonce, Christina Aguilera, Whitney, Ariana Grande, and a that whole crew of amazing female crooners. Aliee is one of Korea's best.



Shinee x SNSD: Of the hundreds of Korean boy and girl groups, Shinee and Girl's Generation (SNSD) represent a couple of the best. This is a classic collaboration, called "Juliette".



Double: Probably the most authentic "R&B" singer from Japan, she was trained in the LA style, bravely continuing her career, retaining the name "Double" after the tragic death of her twin sister. If you like this, you might also enjoy the singers "MISIA" and Amuro Namie, who retires this year. Also check out the iconic band "Dreams Come True" for knock your socks off vocals, solid guitar and drums.



Exile: All boy group known for the best Japanese adaptation of US hip-hop style. Crisp and intense dance moves and smooth vocals. This clip is a classic version of "24Karats" in collaboration with the sultry singer / dancer "Sowelu".



E-Giirls: Just as I was beginning to tire of Japan's super girly, cutesy girl groups like Morning Musume and AKB48, 
Exile blasts out of the chute with their protégé group, E-Girls, consisting of 3 sub-groups: Dream, Happiness, and Flower. This group produces torrid dance moves in the Exile style, with solid vocals. The perfect combination of feminine beauty, strength and diverse expressions of fashion. Love this group.


(sorry, can't seem to avoid the annoying ads from YouTube)
Posted:
Topic Replies: 5, Topic Views: 223
Good find. I tried adding OCR to DVDPca a while back and failed. Wish I had known about this. Window has closed for me pick that one up again.
Posted:
Topic Replies: 7, Topic Views: 370
Quoting GSyren:
Quote:
I have found a couple of problems:
- Save Flag Set does not work correctly for profiles with alternate IDs
- Save Flag Set does not work correctly for Non-US region profiles
- Edit Profile comes up with an empty profile

Flag sets should contain ID, but are saved with ID_Base
(assuming that the API uses the same terminology as in the exported XML).

I think for flag sets, <ID> should be used, as it also includes the Locality. Corresponds to "Profile ID" internally. But I'm sure Mark knows that. Must have been an oversight.
Posted:
Topic Replies: 571, Topic Views: 30626
Invelos Forums->Posts by mediadogg Page: 1 2 3 ...9  Previous   Next