|
|
|
Welcome to the Future
|
|
Senior Member
|
22. August 2003 @ 02:04 |
Link to this message
|
|
Firstly, et me say a big thank you to the boards administrators for starting this new section. I firmly believe that high resolution is here to stay as long as people understand it properly, and AfterDawn is the best place to start.
Secondly, let us turn this section into one where people can learn, the same as happens in the other boards.
Finally, welcome to the future.
|
|
Advertisement
|
 |
|
|
|
Praetor
Moderator
|
22. August 2003 @ 10:04 |
Link to this message
|
|
Me too! Because I know absolutely nothing about audio hehe... if i can hear it i'm happy hehe. Damn I hate that "stupid" feeling ;-)
|
Senior Member
|
23. August 2003 @ 01:19 |
Link to this message
|
On DVD-Audio, you have 2 types of sound files. There is either "standard" wave files, at anything between 16/44.1 to 24/192, and a compressed format called MLP. Multichannel audio is anything from 16/44.1 to 24/96. However, multichannel at 24/88.2 or 24/96 exceeds the available bitrate which is 9.6 MBPS for a DVDA player, so MLP packing becomes compulsory. This is a lossless compression and it works by repacking the stream in a more efficient manner.
This would have the file extension .mlp as opposed to the more common .wav
DVDA uses both of these formats, and the decoder for MLP is built into the players. It's not really a conversion, it just "unpacks" the audio again. MLP will also give you a huge space difference as it achieves an average of 50% filesize reduction. In effect, it IS a wave or PCM file. There are no conversions taking place at all.
|
|
Praetor
Moderator
|
23. August 2003 @ 06:23 |
Link to this message
|
|
Hehe I need a tech dictionary already ;-)
|
Senior Member
|
23. August 2003 @ 08:22 |
Link to this message
|
|
It seems to be instantaneous, which infers it unpacks as it's playing. There's no noticeable lag on my player. I'll check it out (another item in the rapidly growing "to do" list!!)
|
Senior Member
|
23. August 2003 @ 10:37 |
Link to this message
|
|
No, it doesn't seem to eliminate the 2 sec gap either. I'll get the textbooks out & post back.........
|
|
tigre
Moderator
|
23. August 2003 @ 13:00 |
Link to this message
|
|
On playback of lossless codecs like MLP, FLAC, Monkey's Audio etc. (the same applies to lossy) a small part of the file is buffered but generally it's decoded on demand, not the whole file at once. You'd need an extremely powerful processor and a huge ammount of RAM (immagine a full 192KHz/24bit DVD-A consisting of 2 tracks -> several GBs of RAM needed).
This message has been edited since posting. Last time this message was edited on 23. August 2003 @ 13:00
|
AfterDawn Addict
|
23. August 2003 @ 17:08 |
Link to this message
|
|
Yeah, I had ruled out the buffering as an option, basically because i thought that if you fast forwarded or skipped a few tracks, that the processing and RAM needed would be collosal(spelt wrong) and not really feasible. If thats so, then maybe thats why the damn units costs so much :-(
|
Senior Member
|
24. August 2003 @ 02:00 |
Link to this message
|
It's not as dear as you might think. Panasonic do one for £150. I'll try to find the link and model number & post it back. There's also the one I use, the Limit DVD9900SE, which is available at Richer Sounds for around £200. This one does it all. MP3, VCD, DVDA, DVDV, JPEG, DVD-R, CD-R, CD-R/W, DVD-R/W, built in DTS and Dolby Digital decoders and to cap it all is fully multi-regional too.
|
AfterDawn Addict
|
24. August 2003 @ 06:15 |
Link to this message
|
|
If it has component out on the back i may well buy it. I need y, PB/PR, CB/CR for connection to my projector.
|
AfterDawn Addict
|
24. August 2003 @ 06:18 |
Link to this message
|
|
Senior Member
|
24. August 2003 @ 06:25 |
Link to this message
|
It's not the one I have, but it's certainly a full fledged DVDA player.
Mine is the DVD9900SE, distinguishable by the centre loading tray with display underneath it.
it does Progressive or interlaced, Pal/NTSC, DD, DTS, MP3, CD, PCM, full 96KHz output - no downsampling unless you switch it in, but look for yourself http://www.richersounds.com/index.php?f=itemdetl.php&p=300260
and it has £100 off. Only £149.95 - bargain.
Might go get me another one. You should get the region hack codes too. If you get this and not the codes, email me and I'll send you the info.
Outputs are:
L,R,C,LFE,Ls,Rs, L&R stereo, Component colour out, video out, PbCb, PrCr, Svideo, Co-axial, Optical & SCART.
This message has been edited since posting. Last time this message was edited on 24. August 2003 @ 06:28
|
AfterDawn Addict
|
24. August 2003 @ 06:33 |
Link to this message
|
That does look like a quality player alright - better than my Plu2 i would say anyway. The Progressive scan is a handy function except last time i tried to turn it on for my projector, the screen disappeared and i couldnt get it back until my mate in richer sounds fixed it. What i was wondering, is the remote control for it any good? I hate these lovely looking players that come with stupid big clumsy remotes.
|
AfterDawn Addict
|
24. August 2003 @ 17:31 |
Link to this message
|
|
Senior Member
|
25. August 2003 @ 03:41 |
Link to this message
|
|
That's a whole new can of worms. I'll post over there later today when I have a half-hour or so. Basically, you're right though as anything much above 20(24 in extreme circumstances) is inaudible. Are we hearing aliasing or artifacts then? anyway - I'll be over ther later.
BTW, the remote on the limit ain't big & clunky. It covers the DVDV & DVDA functions plus the memory/programming functions.
|
AfterDawn Addict
|
25. August 2003 @ 05:11 |
Link to this message
|
It sounds like a great player - i am seriously considering it now, especially since it has Progressive scan as well.
Quote:
Are we hearing aliasing or artifacts then?
I suppose thats the question the producers are eciding upon when setting the frequencies.
|
Senior Member
|
25. August 2003 @ 05:37 |
Link to this message
|
|
Not to mention the whole Psychoacoustics issue!
My gut feeling is to use 96KHz if you can, but anything above this is probably pointless! Also, I still think you get better results if you work at the frequency of your target media.
However, there is some interesting research going on suggesting that if you put vibration transducers directly in contact with the mastoid bone behind the ear, it does show perception at least as high as 90KHz!! this is percieved and pitched by listeners in the 8-16KHz range though. It appears to be "almost" certain that the perception of these tones is due to subharmonic and "difference-tone intermodulation" than direct perception, so who knows. A good example of this are hard-driven compression drivers, which generate these subharmonics in the ultrasonic range, but you actually hear the difference within the traditional audio band.
The AES has decided that an upper sampling rate of 60-64KHz will cover all sound "in air", and any audible differences are probably down to filter design. The main areas here seem to be
1/. - steep sloped filters near the passband having undesired amplitude & phase effects
2/. passband frequency-response ripple, and
3/. "pre-echo" due to the phase response of linear phase filters could also be audible.
The superb "5.1 surround sound - up & running" by Tomlinson Holman has many more discussions on this, also see Bob Katz for more real world observations. The one certain thing is that this whole area is very much open to discussion.
Finally, how many people actually have the equipment to monitor these frequencies? I wonder just how many of those who claim audible differences have got an aliasing problem in their loudspeakers?
|
AfterDawn Addict
|
25. August 2003 @ 07:30 |
Link to this message
|
Quote:
subharmonic and "difference-tone intermodulation", than direct perception, so who knows
Does that mean we are hearing it without actually hearing it as such, or that we are not able to distinguish what we are hearing? Am i also right in saying that the sound measured from a driver, will change frequency as it travels through the air?
Quote:
I wonder just how many of those who claim audible differences have got an aliasing problem in their loudspeakers?
I doubt very many could claim differences without having the right equipment to measure it. Therefore, it would be assumed that the audible differences would be down to problems with the Loudspeaker.
|
Senior Member
|
25. August 2003 @ 08:11 |
Link to this message
|
|
I don't think it's that we can't distinguish what we are hearing, more like we don't always actually hear what's there. The brain is wonderful at interpolating data in a number of ways and seeing or hearing what it thinks it should be, which is why camouflage works and clever illusionists are able to convince you they can do magic.
A good one, although not strictly speaking an audio one - but related to this discussion, is the oddity in the relationship between picture and sound. There has been a lot of work done that suggests a better sound quality will make you think the picture quality is better.
There's also Heisenberg's principle. The actual act of measuring something causes changes in that being measured. I wonder if this applies here too?
As far as frequencies changing as sound travels, this must be true as we all know about the decay in the top end the further away you are from the source.
For me, the most important and audible changes in hi-res come from using a larger word. Once you get above 96KHz sample rate there really doesn't seem to be much, if any, audible difference. We are talking about 99% of the population 99% of the time now. I'd even go as far as to say with properly designed filters and converters, most people won't hear any changes going above a 48KHz sampling rate.
How many people do we know that really can hear above 22-24KHz? Most people apparently cannot really hear the difference between PCM and MP3.
(see various posts in the Audio board for this)
I guess the usual motto should hold - if it sounds good to you, then it is good enough for you. Personally, I hope the high resolution formats are here to stay, but we need to decide on the sample rates for once and all - the current trend for doubling every other year is bad for studios and worse for the consumer who won't buy anything in case they get stuck with the next betamax. 24/96 is as far as I go until I get a client that wants 24/192 stereo. If this happens I'll go buy the converters but until then i'm staying where I am. It's going to be the limit for multichannel fpr a while anyway as DVDA only allows for 24/192 in stereo and I for one haven't had any enquiries about even this. DVDA players seem to cut off at 96 - some don't even go that far so we need to be careful there too.
This message has been edited since posting. Last time this message was edited on 25. August 2003 @ 08:20
|
AfterDawn Addict
|
25. August 2003 @ 08:29 |
Link to this message
|
Quote:
There's also Heisenberg's principle. The actual act of measuring something causes changes in that being measured. I wonder if this applies here too?
That is something i have never considered before. As far as im aware Heisenberg's principle relates to uncertainty and more to the point, measurement of position which disturbs a particle's momentum. In our case the particle's momentum would be the frequency of sound. Does this mend that tools used to measure the sound actually cause the sound to be different to its original form. In saying that, doesnt that mean that the ears, brain and sensors used by humans to understand what they are hearing, is actually changing the original sound also. In that case, what we hear is not what is produce, but an interpretation of what we think is there. Am i off the mark a bit with that, or am i following you ok :-(
Quote:
DVDA players seem to cut off at 96
If DVDA players cut off at 24/96, then would there ever be a erason to go to 192? Doubling it doesnt seem to make any sense to me.
Suppose I have a time-varying signal such as a sound wave, and I want to know the exact frequencies in the signal at an exact moment in time. I presume to determine the frequencies accurately, you need to sample the signal for a short time, therefore lose time precision which is needed for sampling sound. In other words, a sound cannot have both a precise time, as in a short pulse, and a precise frequency, as in a continuous pure tone. If this is correct, how do we actually measure sound. Seems to me like a circle of hate lol
This message has been edited since posting. Last time this message was edited on 25. August 2003 @ 08:32
|
Senior Member
|
25. August 2003 @ 08:40 |
Link to this message
|
|
Agreed with you here. The uncertainty principle does indeed apply to particles in motion, but what I was asking in light of research done into audio perception is "does something similar apply". Badly worded, what!.
I do think that in the ultrasonic range we are hearing an interpretation as opposed to the actual sound. Also, what are the filters our brain uses like in the aliasing department?
With the DVDA players, 192 KHz is the upper end for stereo tracks in the DVDA format specs. I guess there must be players out there that will do this, but I've not yet seen one. You could just about get this information down an optical line, but you'd be using all 8 streams to carry 2 streams so the recombining of them could be interesting.
As far as the last point goes, I just don't know the answer. Certainly going to make a few calls tomorrow and try to find out though. I guess - and emphasize the guess - that it depends on exactly what sound is - a particle or a wave, or more likely a quanta, the same as light is. It's an extension of the electromagnetic spectrum, as light is, but I really don't know.
|
AfterDawn Addict
|
25. August 2003 @ 13:29 |
Link to this message
|
|
Can sound be classified as a quanta? I wouldve though the main question would be is sound a wave or a particle. If its a particel, then measurement of sound would be taken on each particle. Since to measure sound it has to be sampled, then sound cant be a particle, since a particle is only one. However, if its a wave, and you sample it, it no longer becomes a frequency, but instead a range of frequencies, so it cant be that either. Well actually, my guess is that it can, but im generalising too much. Who really knows for sure? Yuo probably do Wilkes, i unfortunately do not :-(
|
Senior Member
|
26. August 2003 @ 02:03 |
Link to this message
|
|
Afraid I don't know either!
I'm going to try and find out though. Could be fun, and I'll almost certainly learn something.
|
AfterDawn Addict
|
26. August 2003 @ 02:50 |
Link to this message
|
|
When you do, teach me so i know as well :-)
|
|
Advertisement
|
|
|
|
IMRANmel
Senior Member
|
3. September 2003 @ 11:54 |
Link to this message
|
|
this is an amazing thread
jus like to highlight
mEwZ...ExPeRiEnCe
mAkE wAy 4 tHe BaD GuY
MEWZONE (GANG STAR!!!!!)
THE GREATEST TRICK THE DEVIL EVER PULLED WAS CONVINCING THE WORLD HE DIDN'T EXIST!!!
AND LIKE THAT.... HE'S GONE!!
|
|
