Player decoding or Receiver decoding, that is the question

[welwynnick]

Of course I've been itching to dive back into this debate for some time, but first I want to go back to this excellent post by EWL5 about amplifier decoding (which I've paraphrased). I whizzed through it first time, but only read it thoroughly yesterday, and something leapt out at me:

Originally Posted by EWL5

This question ranks right up there with "Which BD player should I get?" and "Can I hear the new lossless audio using my current receiver?" ......

On a purely theoretical basis, the decoding done in the player should be identical to the decoding done in the AVR/prepro.......

Shouldn't they sound the same given the fact that decoding is just a simple unwrapping of PCM from the original codec? Yes and No and here's why:

Yes, decoding will sound the same
When the following conditions are met

1) The crossover in the BD player is the same as the one in the AVR/prepro

2) Delay/Level settings for each speaker channel approximate the settings arrived at when calibrating for the AVR/prepro

3) For analog out, it is possible for the LFE to be boosted by +15dB in the downstream amp whenever speaker settings in the BD player are set to SMALL.

4) Identical DAC's used b/w the BD player and the AVR/prepro

5) ..when a player decodes and transmits the PCM via HDMI 1.1+ as the DAC's are used in the AVR/prepro....

No, decoding will not sound the same

1) Most BD players have a fixed crossover (~100-150Hz)... AVR/prepro's on the other hand, allow a multitude of crossover settings to best match your situation..... Analog users are at the mercy of the fixed crossover.

2) Most decent AVR/prepro's come with automatic calibration functions that attempt to correct for errant frequencies, nulls, etc. (examples include Audyssey, MCACC, etc.)....

3) Boosting of LFE is always appropriately handled when the AVR/prepro accepts the bitstream. If you are working with multichannel analog out connections, you better have the capability to boost that LFE by +15dB whenever any speakers are set to SMALL in the BD player.

4) DAC's play a large part in the final output. In most cases, AVR/prepro DAC's will be superior to a BD player's...

5) Some naysayers claim that jitter from PCM transmission over HDMI audibly affects the signal. Can be minimized or solved using technologies like Pioneer's PQLS or Denon's DenonLink.

Almost all the issues with amplifier decoding relate to D to A conversion, and not to bitstream to LPCM conversion. The exception is the final one, in bold. ELW5, if you think that any of the other nine issues apply to decompression, could you explain why, as I don't see it?

This leaves me with jitter as the explanation for the difference. There may be other issues, which have been discussed at AVS and AVF, but these seem to boil down to noise and interference on the LPCM signal from the player, and I would categorise those as jitter-like anyway.

There was some uncertainty expressed here about whether it was really as simple as that. That related to differences heard between the PS3 and stand-alone players when using a Pioneer SC-09TX receiver. That uses TI/BB SRC4192 sample rate convertors to buffer the digital audio stream. Similar SRCs in the SC-07 were shown to be very effective in jitter reduction, and it was presumed that the 09TX would perfrom at least as well. HiFi News tested both in the last few months, and found that the older but more expensive model actually had much higher jitter - 710ps vs 50ps, so this explianed why it would be sensitive to LPCM jitter from the PS3 in this example.

To reinforce this from my own point of view, I'm quite familiar with the effect of jitter on stereo replay. I've had two players and three amplifiers with i-link connections, and they all clean up music reproduction in ways that are very similar to amplifier decoding. The effects are familar and recognisable, and while there could be other explanations, I'm pretty convinced that it is indeed down to jitter.

Nick

[axe79]

Originally Posted by welwynnick

Of course I've been itching to dive back into this debate for some time, but first I want to go back to this excellent post by EWL5 about amplifier decoding (which I've paraphrased). I whizzed through it first time, but only read it thoroughly yesterday, and something leapt out at me:

Almost all the issues with amplifier decoding relate to D to A conversion, and not to bitstream to LPCM conversion. The exception is the final one, in bold. ELW5, if you think that any of the other nine issues apply to decompression, could you explain why, as I don't see it?

This leaves me with jitter as the explanation for the difference. There may be other issues, which have been discussed at AVS and AVF, but these seem to boil down to noise and interference on the LPCM signal from the player, and I would categorise those as jitter-like anyway.

There was some uncertainty expressed here about whether it was really as simple as that. That related to differences heard between the PS3 and stand-alone players when using a Pioneer SC-09TX receiver. That uses TI/BB SRC4192 sample rate convertors to buffer the digital audio stream. Similar SRCs in the SC-07 were shown to be very effective in jitter reduction, and it was presumed that the 09TX would perfrom at least as well. HiFi News tested both in the last few months, and found that the older but more expensive model actually had much higher jitter - 710ps vs 50ps, so this explianed why it would be sensitive to LPCM jitter from the PS3 in this example.

To reinforce this from my own point of view, I'm quite familiar with the effect of jitter on stereo replay. I've had two players and three amplifiers with i-link connections, and they all clean up music reproduction in ways that are very similar to amplifier decoding. The effects are familar and recognisable, and while there could be other explanations, I'm pretty convinced that it is indeed down to jitter.

Nick

Hi guys, whilst I am at a disadvantage to many of you here who take a far more Lab measured approach, (and maybe have the facilities) I do have many years of domestic real world experence.
Many on this board appear to be USA based.
In the UK, Meridian are regarded as making some of the best High End equipment and have been cutting edge with much of it. I own a 207CD/pre amp, (2 box) regarded as the best CD player in the world when 1st brought out. CD players ALL have issues with the clock reading info off the disc and 'superclocks' with better circuit design have been made to improve things.
Years ago with the help of Russ Andrew I modified this player to a high degree. Incidently I obtain the same Hi End sound off a Hard Drive today!

There is a difference in the interaction between the PS3 and the Receiver on LPCM v Bitstream OTHERWISE THEY WOULD BOTH SOUND IDENTICAL and they don't. It looks like what is going on is very similar to what was/is still happening with CD players.
Logic tells me that if the PS3 (with all the updates installed) has a variable in the delivery of the information down the HDMI cable the issue is 'inside the player'!
Now I have a Totally seperate Dedicated Mains spur connected up to Two
Mains transformers/mains capacitors/surge protection circuits (British Telecom ones for Main Frames etc) (The Russ Andrews ones are £2750 each!) which feeds the equipment in as clean manner that is possible.
No RFI, no Mains noise off other circuits, no voltage surges or spikes.
Fed with a 'perfect feed' clear differences can be heard.

Sony cannot by software updates ever make the PS3 put out in full Bitstream like its newer players can, which on Vision & Sound show improvements. So it is really about getting the most out of the various settings available.
And that is a big can of worms!

[EWL5]

Originally Posted by welwynnick

Of course I've been itching to dive back into this debate for some time, but first I want to go back to this excellent post by EWL5 about amplifier decoding (which I've paraphrased). I whizzed through it first time, but only read it thoroughly yesterday, and something leapt out at me:

Almost all the issues with amplifier decoding relate to D to A conversion, and not to bitstream to LPCM conversion. The exception is the final one, in bold. ELW5, if you think that any of the other nine issues apply to decompression, could you explain why, as I don't see it?

This leaves me with jitter as the explanation for the difference. There may be other issues, which have been discussed at AVS and AVF, but these seem to boil down to noise and interference on the LPCM signal from the player, and I would categorise those as jitter-like anyway.

There was some uncertainty expressed here about whether it was really as simple as that. That related to differences heard between the PS3 and stand-alone players when using a Pioneer SC-09TX receiver. That uses TI/BB SRC4192 sample rate convertors to buffer the digital audio stream. Similar SRCs in the SC-07 were shown to be very effective in jitter reduction, and it was presumed that the 09TX would perfrom at least as well. HiFi News tested both in the last few months, and found that the older but more expensive model actually had much higher jitter - 710ps vs 50ps, so this explianed why it would be sensitive to LPCM jitter from the PS3 in this example.

To reinforce this from my own point of view, I'm quite familiar with the effect of jitter on stereo replay. I've had two players and three amplifiers with i-link connections, and they all clean up music reproduction in ways that are very similar to amplifier decoding. The effects are familar and recognisable, and while there could be other explanations, I'm pretty convinced that it is indeed down to jitter.

Nick

I personally have not heard the difference b/w jitter from LPCM over HDMI vs bitstreaming as I don't own a receiver that can decode the new codecs. At this time, I believe current jitter reduction circuitry only applies to stereo and not multichannel (more for CD's than movies). The Denon A1UDCI may be the first BD player to incorporate jitter reduction for multichannel using DenonLink+HDMI simultaneously but the language is not clear.

Read about it here:

http://www.usa.denon.com/DVDA1UDCI.blurayrel.Final.pdf

It was hoped that the Pioneer 09 flagship BD player would have multichannel PQLS but I think it is also restricted to 2-channel.

[welwynnick]

Originally Posted by EWL5

I personally have not heard the difference b/w jitter from LPCM over HDMI vs bitstreaming as I don't own a receiver that can decode the new codecs. At this time, I believe current jitter reduction circuitry only applies to stereo and not multichannel (more for CD's than movies). The Denon A1UDCI may be the first BD player to incorporate jitter reduction for multichannel using DenonLink+HDMI simultaneously but the language is not clear.

Read about it here:

http://www.usa.denon.com/DVDA1UDCI.blurayrel.Final.pdf

It was hoped that the Pioneer 09 flagship BD player would have multichannel PQLS but I think it is also restricted to 2-channel.

For the record, here's a bit more info on the new Denon and DL4:

DENON LINK 4th Edition Further Enhances Audio and Video Performance
In yet another major advancement, the DVD-A1UDCI features the new DENON LINK 4th Edition digital interface, which significantly enhances both video and audio quality. DENON LINK 4th fully supports all digital audio transmission specifications, including SACD as well as CD and DVD-Audio. The 4th Edition of DENON LINK adds the exclusive Denon Flow Control feature; when using the HDMI interface for A/V output, Denon Flow Control takes over the Master Clock and Jitter control – a feature that will be especially useful when used with upcoming Denon A/V receivers.
DENON LINK is a proprietary, high-quality digital audio transmission technology that utilizes a real-time balanced transmission system to protect the signal from external noise. With the new advanced DENON LINK 4th used in an HDMI connection, as well as for HD audio read from Blu-ray discs, the master clock that operates the D/A converter of the A/V surround amplifier is transmitted to the player, enabling the circuitry to be operated while sharing the same clock, thereby achieving digital audio transmission with virtually no jitter. Sound localization becomes more precise, and a greater sense of space is produced in the sound images. When combined with a Denon A/V surround sound receiver that supports DENON LINK 4th, users will now be able to enjoy the absolute maximum level of sonic quality possible with HD audio. Denon will make available a DENON LINK 4 upgrade for customers of the AVP-A1HDCI and AVR-5308CI; details to be announced shortly after the launch of the DVD-A1HDCI.

http://www.avsforum.com/avs-vb/showt...1048326&page=5
http://www.hometoys.com/news_detail.php?id=10884091
http://www.avsforum.com/avs-vb/attac...7&d=1228519346
http:///www.hometheaterhifi.com/foru...ead.php?t=7085

Pioneer, Sony and Denon all have jitter-reducing interfaces, but are not as clear as they might be about how they do it. Its pretty clear that Sony and Pioneer use HDMI V1.3a Audio Rate Control, which is essentially an HDMI implementation of what i-link did with firewire. Denon's DL4 sounds like the same thing, but they are rather ambiguous.

Note that all of these schemes only work with audio sources, not DVD or blu-ray video sources, and theres a very good reason for this. Pioneer suggest they will be able to get roud this with their upcoming BDP-23FD, which will have multi-channel PQLS over HDMI, which is seriously clever. Until then, their SC-07 seems to have the only effective jitter-reduction capability that can be used with any player, achieved simply by re-sampling.

Nick

[EWL5]

Originally Posted by welwynnick

For the record, here's a bit more info on the new Denon and DL4:
[FONT='Arial','sans-serif']
http://www.avsforum.com/avs-vb/showt...1048326&page=5
http://www.hometoys.com/news_detail.php?id=10884091
http://www.avsforum.com/avs-vb/attac...7&d=1228519346
http:///www.hometheaterhifi.com/foru...ead.php?t=7085

Pioneer, Sony and Denon all have jitter-reducing interfaces, but are not as clear as they might be about how they do it. Its pretty clear that Sony and Pioneer use HDMI V1.3a Audio Rate Control, which is essentially an HDMI implementation of what i-link did with firewire. Denon's DL4 sounds like the same thing, but they are rather ambiguous.

Note that all of these schemes only work with audio sources, not DVD or blu-ray video sources, and theres a very good reason for this. Pioneer suggest they will be able to get roud this with their upcoming BDP-23FD, which will have multi-channel PQLS over HDMI, which is seriously clever. Until then, their SC-07 seems to have the only effective jitter-reduction capability that can be used with any player, achieved simply by re-sampling.

Nick[/font]

It does appear that the war against jitter for BD movies is not full blown yet. Pioneer's got some nifty ideas but the company as a whole is consolidating.

[Sir Terrence]

Originally Posted by welwynnick

Of course I've been itching to dive back into this debate for some time, but first I want to go back to this excellent post by EWL5 about amplifier decoding (which I've paraphrased). I whizzed through it first time, but only read it thoroughly yesterday, and something leapt out at me:

Almost all the issues with amplifier decoding relate to D to A conversion, and not to bitstream to LPCM conversion. The exception is the final one, in bold. ELW5, if you think that any of the other nine issues apply to decompression, could you explain why, as I don't see it?

This leaves me with jitter as the explanation for the difference. There may be other issues, which have been discussed at AVS and AVF, but these seem to boil down to noise and interference on the LPCM signal from the player, and I would categorise those as jitter-like anyway.

There was some uncertainty expressed here about whether it was really as simple as that. That related to differences heard between the PS3 and stand-alone players when using a Pioneer SC-09TX receiver. That uses TI/BB SRC4192 sample rate convertors to buffer the digital audio stream. Similar SRCs in the SC-07 were shown to be very effective in jitter reduction, and it was presumed that the 09TX would perfrom at least as well. HiFi News tested both in the last few months, and found that the older but more expensive model actually had much higher jitter - 710ps vs 50ps, so this explianed why it would be sensitive to LPCM jitter from the PS3 in this example.

To reinforce this from my own point of view, I'm quite familiar with the effect of jitter on stereo replay. I've had two players and three amplifiers with i-link connections, and they all clean up music reproduction in ways that are very similar to amplifier decoding. The effects are familar and recognisable, and while there could be other explanations, I'm pretty convinced that it is indeed down to jitter.

Nick

Can you show me any evidence that the PS3 has issues with jitter? Because I have read alot of your theories online, but never seen any evidence visually that supports this notion. I am not saying the PS3 is jitter free, because I have no evidence it is, but I hate theories that are not supported, and so far this one isn't. A scope picture would be nice, a valid measurement taken from a valid source, anything that would give your theories credibility.

[Sir Terrence]

Originally Posted by axe79

Level matched precisely, under controlled acoustical conditions in a very quiet room, of course.
But why not THX, what format would you process?

How about no processing whatsoever. THX DSP's are signal degrading circuits that do not belong in any listening test. The pure direct mode with no processing is much better.

No other settings sound better on my 875 given its supply.
But if you haven't addressed a very clean Mains supply to your separates the noise off your power rails will be greatly effecting the Burr Brown DAC's on the signal path.

Whatever THX processing does to clean up your sound, it undoes by manipulating the mix in a way that it is not supposed to be. THX processing is for THX speakers, that is what the processing is designed for.

Is this the most important issue here? Really? How do you know I haven't addressed it? Making a lot of assumption aren't you?

The number of toroidal transformers in your system along with 2 independent audio and video transformers do demand a totally clean feed or the interaction is terrible. No mention of you addressing this.

Do I need to explain in detail what is under the dress of all my equipment? Sorry, I don't do that because then the conversation becomes a pissing contest, and that is not my gig.
I didn't mention this because it is a pile of crap compared to a 20db deviation because of a room resonance. Once again, you continually address equipment issues, but never address room issues. This is why I cannot take what you state seriously. You are majoring in minors here.

[axe79]

Originally Posted by Sir Terrence

How about no processing whatsoever. THX DSP's are signal degrading circuits that do not belong in any listening test. The pure direct mode with no processing is much better.

It might be on your equipment, fed with a virgin clean Mains supply ALL the processing channels are as clean as one another. Musically nothing is being lost in the extra processing circuits.
The signal degradation being put through so many chains in your equipment is not ideal at all. I have for years had Hi End Hi Fi systems where you had no tone controls, equaliser settings etc and the fine tuning was done by careful selection of components, cables, mains conditioners, and mods to internal circuitry. Oh and tuning the room with furnishings, wall treatment, seating position, equipment set up etc.
My brief when I went over to building an AV system was to develop it so it would play music as well as very decent Hi Fi set up. Have enjoyed that standard since the 70's. I didn't buy a AV amp before the Onkyo 875 because there was nothing on the market which had the build design I was looking for to work on. Full class A, massive transformer, Burr Browns, etc. That's why the unit weighs in at around 57lbs. It was a good starting point.

I have now achieved a system that is much more than the sum of its parts.



Whatever THX processing does to clean up your sound, it undoes by manipulating the mix in a way that it is not supposed to be. THX processing is for THX speakers, that is what the processing is designed for.

Its a good job then that I have THX speakers with 80hz crossover settings apart from the front pair that are set up Full range for Music and balanced out on the equalizer settings.

Is this the most important issue here? Really? How do you know I haven't addressed it? Making a lot of assumption aren't you?

Not really, if you had you wouldn't need a Berhinger Feedback Destroyer to process things before hitting the Horns! If you had really addressed this you would be sure to mention it in your specs as it costs a darn site more to sort out to reference levels than a couple of pieces of your kit!


Do I need to explain in detail what is under the dress of all my equipment? Sorry, I don't do that because then the conversation becomes a pissing contest, and that is not my gig.
I didn't mention this because it is a pile of crap compared to a 20db deviation because of a room resonance. Once again, you continually address equipment issues, but never address room issues. This is why I cannot take what you state seriously. You are majoring in minors here.

In your opinion.
I very much know how to get music set up to foot tapping perfection for my room. Am well aware of how sound waves travel, wave lengths, resonance etc. I don't need a meter to tell me either, just like a piano tuner doesn't.
By what you say you do have a particular problem with your room that has/is taking a bit of sorting out. Not something that can be easily and fairly be discussed on a board with so many variables.
I also know how to modify, and how to tweak for particular effects by the choice of components in the design. Schottky Diodes, Os-cons, DNM Design T-Network, Silmic, Polypropylene, capacitors. Different makes/models have different chararistics.
Much of this work is a 'black art' and tuning sound so it has musicality with depth, detail, clarity, balance, texture is something that develops with time.

I'm not here to fight, Sir Terrence, (I don't use my title on the boards so I assume you don't have one) but do report on listening observations that are quite real!

Regards

[welwynnick]

Originally Posted by saprano

There is no reduced quality, loss of information, or anything else missing when ps3 decodes the HD formats.

Originally Posted by Sir Terrence

Timing issues within HDMI as far as video have already been identified. I have yet to hear anyone complain about jitter through HDMI.

It still has not been demonstrated that you get better performance via bitstreaming to the receiver over the player doing the decoding.

Originally Posted by Sir Terrence

Where is your proof the PS3 suffers from jitter? I would like to see a picture of the measurements you have that support this theory. From what I understand from another insider, the PS3 has some of the most accurate decoding of the advance codecs outside of the high end.

Originally Posted by Sir Terrence

I don't think it is fair. If you want to do these tests properly, you have to eliminate all differences. You would probably need two identical players hook to your receiver, both using a chip based decoding solution.

Originally Posted by Sir Terrence

Almost all references to jitter I have read about come from SDPIF based connections, and comments on jitter from HDMI have not been supported with visual accompaniment, so are just theories until visual representations show up.

My room has only a 3db variation from 20-40khz that it imparts on the output from my speakers which vary only 1.5db from 20-40khz in a anechoic chamber, are time aligned and phase correct, and I have had my hearing tested and still cannot hear a difference from bitstreaming and player decoding. I even have the ability to precisely volume match all of my sources via my HD digital switcher/processor, and still cannot hear a difference.

Originally Posted by Sir Terrence

Can you show me any evidence that the PS3 has issues with jitter? Because I have read alot of your theories online, but never seen any evidence visually that supports this notion.

I think we're looking in the wrong direction on a couple of fronts.

Firstly, all the attention and blame seems to be directed at the PS3, but the problem is really a more generic one and relates to the HDMI interface, and what receivers and processors do with it. The PS3 probably does have more jitter than other players, but this is a supposition from lots of other people's accounts, and I'm not aware that any jitter measurements have been done on it. I'm only aware that Stereophile and HiFi News that; not on the PS3 as yet, but they have produced these jitter measurements that show HDMI is generally a far worse culprit than SPDIF or HATS. Many people (myself included) have commented that HDMI audio sounds worse than other, better connections (including analogue!) and there is some proof of this here:

Jitter with SPDIF input:
15ps DCS Scarlatti (£18k)
37ps Pioneer SC-LX81
40ps Cambridge DACMagic
183ps Yamaha RX-V3900
250ps Denon 2500/AVP-A1
430ps Onkyo PR-SC886
470ps Onkyo TX-NR906
485ps Audiolab 8000AP
560ps Denon 3808A

Jitter with HDMI input:
50ps Pioneer SC-LX81
200ps Sony STR 5400ES with HATS
2000ps Sony STR 5400ES without HATS
2200ps Denon 2500/AVP-A1
3700ps Denon 3808A
3860ps Onkyo TX-NR906
3900ps Onkyo PR-SC886
7660ps Yamaha RX-V3900
8490ps Audiolab 8000AP

In the context of amplifier decoding, the sensitivity to jitter really lies in the amplifier, and we shouldn't worry about the player too much. The real issue is that bitstream allows much less sensitivity to jitter because of the different replay architecture. Its much the same as with DD & DTS, only with higher bit rates, and more quality at stake.

I always stood on the fence, saying that amplifier decoding MAY make a difference, but I heard it for myself three weeks ago, and now its quite clear to me that with the players and amplifiers available to me, it does make a difference. My doubt was over the way the amplifier generated the audio clock. With LPCM, the clock is derived from the video pixel clock on HDMI, but there's no reason why this shouldn't have been used with bitstream as well. Had that been the case, then bitstreaming probably wouldn't have made any difference. But its obvious that Onkyo at least doesn't do this, and can capitalise on the opportunities with bitstream. It could be that other manufacturers will have different architectures, and will use video clock decimation for audio clock generation with bitstream inputs as well as LPCM. They would presumably gain no benefit from bitstreaming.

Secondly, there's a lot of focus on spectral response and performance, and I think this is a big red herring. Jitter does have a slight impact on brightness etc in some instances, but those aren't the sort of differences that I hear and look out for. This is how I described the differences I hear:

My wife also tended to listen to the tonal sound of the codecs, which isn't what its all about to me. I don't think in terms of the instantaneous tonal characteristics, so much as temporal and spatial characteristics.

Bitstream, like other jitter-reducing methods, gives much the same sound, but better resolution and imaging. Its especially good at retaining the tiny, vanishing, decaying notes and echos that analogue audio was always good at. I think this has a number of advantages:

• You get better focus on each subject.
• There is greater depth in the image.
• You can hear musical instruments playing continuously and independently of everything else - you hear them concurrently, not sequentially.
• Musical instruments become joined-up and coherent, instead of independant and unrelated.
• The timing of music is improved, because the gaps between notes are filled with delicate decays and reverberations.
• They were there on purpose when the piece was produced, played and recorded.
• Without this small-signal fidelity, you just hear sound - pause - sound - pause etc, which is empty and boring, there's no flow, pace or variety.

These were the things that I was looking for (I hope some of you at least understand what I'm prattling on about) and I found it quite easy to hear the differences. These differnces were also quite apparent with all soundtrack, not just with music!

So getting the speakers or the room equalised seems quite irrelevant to me, notwithstanding the fact that both sources were heard in the same environment. The benefit to me is in better fidelity as the signals get smaller, not larger, and particularly in the depth and space of the soundstage that is produced, be it stereo or multi-channel, music or dialogue. Everything benefits, even if the "sound" is exactly the same. If someone was specifically listening to the "sound" - the spectral envelope or something like that, they would probably be disappointed. Bitstream doesn't give you more treble or anything like that, it gives you space.

Originally Posted by EWL5

I will have to do an LPCM shootout b/w my $600 PS3 and my $2000 Denon 3800. It's quite possible that the jitter from the PS3 is bad enough to be significantly different from the 3800.

Originally Posted by EWL5

I don't have any proof of jitter because I haven't done the shootout yet.
Since they both fully decode and output PCM over HDMI, the comparison should be fair, yes?

Originally Posted by EWL5

Since the PS3 decodes in software, does that mean the Denon 3800 will have the audio advantage or does that remain to be seen (or in this case, heard)?

I guess the only fair comparison for a PS3 would be something like a HTPC that decodes the codecs in software as well.

Darn, it'll be 2 weeks before I get to do this shootout.

Those will truly be interesting comparisons, and I really look forwards to hearing about what you find. You are fortunate to have a Denon 3800, as thats a really great player, and allows you to hear the difference between player and amplifier decoding when using the same player. A comparison with the PS3 would still be interesting though, and i expect you would find that the 3800 sounds slightly better than the PS3 when decoding is done in the player in both cases.

best regards, Nick

[Sir Terrence]

Originally Posted by axe79

In your opinion.
I very much know how to get music set up to foot tapping perfection for my room. Am well aware of how sound waves travel, wave lengths, resonance etc. I don't need a meter to tell me either, just like a piano tuner doesn't.
By what you say you do have a particular problem with your room that has/is taking a bit of sorting out. Not something that can be easily and fairly be discussed on a board with so many variables.
I also know how to modify, and how to tweak for particular effects by the choice of components in the design. Schottky Diodes, Os-cons, DNM Design T-Network, Silmic, Polypropylene, capacitors. Different makes/models have different chararistics.
Much of this work is a 'black art' and tuning sound so it has musicality with depth, detail, clarity, balance, texture is something that develops with time.

I'm not here to fight, Sir Terrence, (I don't use my title on the boards so I assume you don't have one) but do report on listening observations that are quite real!

Regards

Wow, I have never heard anyone say they don't need a meter to equalize any volume mismatches between source. IMO, this test you performed is completely invalid, and cannot be used to advance ANY argument on bitstream versus LPCM. Level matching is of upmost importance when trying to conduct any listening test between two sources. Our ears are extremely insensentive to small volume mismatches, and cannot be used to match volume between two source.

[Sir Terrence]

Originally Posted by welwynnick

So getting the speakers or the room equalised seems quite irrelevant to me, notwithstanding the fact that both sources were heard in the same environment. The benefit to me is in better fidelity as the signals get smaller, not larger, and particularly in the depth and space of the soundstage that is produced, be it stereo or multi-channel, music or dialogue. Everything benefits, even if the "sound" is exactly the same. If someone was specifically listening to the "sound" - the spectral envelope or something like that, they would probably be disappointed. Bitstream doesn't give you more treble or anything like that, it gives you space.

best regards, Nick

Another invalid observation because the room plays such a big part of what you hear. Having controlled room acoustics is an extremely important aspect of doing any listening test. as room resonances can contaminate anything you hear.

If you guys are going to do these kinds of tests, and then spread your observation all over the internet, they are going to have to be conduct in much less an amateur kind of way, or they are just not going to be taken seriously. Anyone who dismisses level matching, and room acoustic is not going to be taken seriously. This is why these kinds of test can only be conducted in a lab, or a studio, not somebody's house with no testing equipment to eliminate variables.

[welwynnick]

Its not a two-man campaign by Axe79 and myself to corrupt every free-thinking person on the net, we are simply agreeing with the majority of people who have compared player and amp decoding where they have the facility to do that. Of course some of the anecdotal accounts are not reliable because people don't level-match properly, and I'm well aware of that. Which is why I DID level match to within 0.5dB - as close as I could get it with the SC886 - and is why I DID do blind testing to back up my own observations. I only came to this thread once I had done that, because I knew what the reaction would be.

Notwithstanding all that, even though the rest of the World are doing comparisons that might not be scientific by your chosen standards, it doesn't mean that the results are wrong. They're not - bitstreaming does sound better in certain systems, and just because somebody cannot hear any difference does not mean that there isn't a difference.

However, it could be that with some combinations of player and amplifier, the digital audio replay architecture may be similar with both schemes, and there could be no reason for them to sound different. What I do know now is that with the Onkyo amp that I used, there was a difference, and from what I hear elsewhere from reliable sources there are differences with many other amplifiers as well.

I've put a great deal of effort into performing accurate and reliable comparisons over two weeks to reach this conclusion, and if you care to look around, you will see that I am being taken seriously.

regards, Nick

[axe79]

Originally Posted by Sir Terrence

Wow, I have never heard anyone say they don't need a meter to equalize any volume mismatches between source. IMO, this test you performed is completely invalid, and cannot be used to advance ANY argument on bitstream versus LPCM. Level matching is of upmost importance when trying to conduct any listening test between two sources. Our ears are extremely insensentive to small volume mismatches, and cannot be used to match volume between two source.

Oh boy, what's your problem?

I am not advancing a argument on anything. But it obvious by your remarks that you cannot think outside the box. Furthermore you insist that everything must be quantified under lab conditions to have any weight. The ears are the perfect vehicle for discerning the weight and balance of music for Christ sake!
What do we listen to music for, to watch it on a meter?

You have not responded to the detailed observations made in my last post. It appears it is you that has the problems and delimers with your own set up in your home and I am getting the impression you are clueless in sorting them out or building a High End system. Do you have more than ABC 'lets measure it' knowledge to work on?
If things were that easy everyone would be enjoying perfect Hi Fi.

Futhermore loose the pretentious handle, you are not entitled to it unless dubed by a royal court of Europe.

[saprano]

How is this turning into an argument? theres no need for that!

Bottom line is this, the fact that not EVERYBODY agrees that bitstreaming is better means that it is in no way a fact. i guess it depends on the system, the person, and the room. do you guys know that some people even think the opposite? doesn't mean its true. i still stand by that there is no difference. i have yet to see that bitstreaming is better is universally a fact. as of right now its all an opinion, and that right there is a fact.

[welwynnick]

I think this has been one of the hottest and most debated topics in most AV forums since the format war ended. People have adopted strong positions based on their own listening experience; the experience of others (professionals and users); and on theories that try to explain what may or may not cause differences.

What complicates things are comparisons suggesting that amplifier decoding is better, when other factors are causing the differences, and null results that don’t prove it one way or the other. I do take a scientific approach to this, and try to understand and explain what makes audio sound good, but I also trust my ears for various good reasons. I suppose I’ve spent half a lifetime trying to reconcile analytical and empirical assessment, and its not easy.

I’ve sat on the fence on this issue for along time, in as far as I couldn’t answer it to my own satisfaction. Last month I finally had the opportunity to get a lot of equipment together that would allow me to find out for myself what I had heard many other people saying. Some of them may have been reliable, but ultimately I only trust myself, and if the benefits of bit-streaming were insignificantly small, or only applicable to certain combinations of mis-matched equipment, then I wouldn’t worry about it.

And that was what I came to think for a few hours about three weeks ago – I actually agreed with sir terrence, crackinhedz and the other skeptics. I had shown to myself that I couldn’t tell the difference between LPCM and bitstream. That didn’t mean to me that there wasn’t a difference – just that I couldn’t hear one with my kit (well, somebody else’s kit actually). My conclusion was either that there wasn’t any difference, or that my hearing wasn’t good enough, or that my equipment wasn’t good enough.

No problem with any of that – if other people wanted to pursue the n’th degree of perfection, that’s OK by me, but I could now be happy with what I had got, and stop worrying about it. As far as I was concerned, the answer was in. Having said that, I was also slightly disappointed, as I thought there MAY be a difference, which is what I’ve been saying for a year or so.

However, I soon found out that the reason I couldn’t hear a difference was because there wasn’t a difference – I had been comparing LPCM with LPCM without realizing it (just used the wrong BD). This was kind of gratifying, because I had shown to myself with a clear conscience that I wasn’t making up differences that I didn’t know weren’t there!

Once I corrected and repeated the test it became quite clear I had been mistaken, and in the system I was using and with the discs I used, there was a clear difference. For the avoidance of doubt, it wasn’t an enormous difference, not as great as getting room acoustics right, but it was pretty fundamental and well worthwhile.

Now to the important point. Just because I heard it in my system doesn’t mean that everyone should hear it as well. Its very easy to make mistakes in set-up and spoil the result – believe me, I know. Other systems may have a different digital audio replay architecture that will suppress any differences. For example, a very good player like the Sony S5000 may be good enough to offset the problems with LPCM transmission over HDMI, and give a null result in the same test. Alternatively, a good receiver like the Pioneer SC-07 may be insensitive to jitter, and will also make LPCM sound as good as bitstream.

I have no idea about the outcome of those situations, and I’m all ears to those have the opportunity. What I did hear was a smaller difference with bitstream when using an expensive player (3800BDCI) against a cheaper player (BH200). I had already pre-supposed that bitstream would be dependant on the amplifier being used, but now it seems, its also dependant on the player as well. And that kind of makes sense.

So all I’m saying is that bitstreaming DID make a difference in a level-matched, blind test using a particular selection of discs, players and amplifiers, but those differences may not be apparent AT ALL with other combinations of equipment.

So while we can say that bit-streaming can make a difference, we can't say that it can’t make any difference.

Regards, Nick

[axe79]

Here is a breakdown of the Vastlane chip family from Silicon Image

SI Vastlane 9132: (Gaming Transmitter) Supports Deep Colour but not the latest high-quality audio formats.
SI Vastlane 9133: (Dual Input Receiver) Enables HDTV's to display Deep Colour but does not support the latest high-quality audio formats.
SI Vastlane 9134: (Dual Input Transmitter) same abilities as the 9133 with additional support for the latest high-quality audio formats.

It is clear to see the evolution on these Chips from the 9132 put in the PS3 to the 9134 found in the newer crop of players.
It is now seen on the dedicated Blue ray players that they all do Bit streaming on the high quality audio formats.
Manufactures in the development research must have concluded that this path was one that needed to be offered.

"The flagship Sony S5000ES is also the first Sony player to feature a Precision Clock Conditioner, designed to reduce the effects of jitter."

Now if it wasn't a problem why address this?

Perhaps it needed to be sorted if the player was to be doing the processing itself and then sending it out!

"All the components used are of the highest quality and include high-performance capacitors and digital-to-analogue convertors."

In England this player is currently regarded as the best that has been reviewed.
I would be surprised if there was differences between Bitstream and LPCM on this player as the problems appear to have been designed out.

Another player that is flying off the shelves at the moment is the Panasonic DMP-BD55.
"Like the entry-level ’BD35, this player is also Blu-ray Profile 2.0, DivX and x.v.Colour compatible.So why does it cost more? The extra outlay gets you upgraded audio parts, including new capacitors, a new 192kHz/24bit DAC and a couple of tweaks to the HDMI output to help reduce jitter."

So Manufacturers are addressing jitter issues. Well, well.

[Sir Terrence]

Originally Posted by axe79

Oh boy, what's your problem?

No problem at all, I just do not agree with your opinion. That is all. The methods you used to come to your conclusions are not sound.

I am not advancing a argument on anything. But it obvious by your remarks that you cannot think outside the box.

You are thinking out of the box because you do not have the equipment to do an sound, accurate listening test that can stand scrutiny. You are thinking out of the box because you cannot follow the rules in the box, you are not equip to do so. The methods and protocols for DBT(not single blind which means nothing) listening test have long been established and effectively used. There is no need to think out of the box on this one. You are only doing so because you have to.

Furthermore you insist that everything must be quantified under lab conditions to have any weight. The ears are the perfect vehicle for discerning the weight and balance of music for Christ sake!

If you are going to get on the internet with your conclusions, they better be able to stand up careful scrutiny. Yours didn't. The ears are sensitive to small volume changes between 1-4khz, and everything above and below that, they become increasingly insensitive(that's about 9.5 octaves of various sensitivities). Have you ever heard of the Fletcher/Munson curve? What you don't seem to fathom is that when you decide to compare two different sources, any differences or variables(such as volume differences) must be taken out of the test, or the test is contaminated. THX circuitry does not belong in any test, plain and simple. I do not care what it does to your amp, you don't use a signal mangling circuit to correct another supposed issue within your amp. That is not the way these tests should be conducted, and why I say they cannot be conducted outside a lab.

What do we listen to music for, to watch it on a meter?

This wasn't about just listening to music, you were comparing bitstreaming to LPCM, or player decoding or receiver decoding. That REQUIRES a meter to equalize any volume mismatches between the two decoding routines. Since I have done this test already, I know the bitstreaming usually ends up being louder between the two methods of decoding, so you have to eliminate that variable if you are going to even TRY to do this kind of test. If this was just a casual listening session, then I wouldn't have wasted my time responding.

You have not responded to the detailed observations made in my last post.

That is because they are minor issues, not major ones. Do you really think you are the only one that understands that you need power regulators and filters in your system???. Well, if it will make you sleep better, power regulation and filtration starts right at the box in my house. My music listening room, and my dedicated theater have their own regulated and filtered power box completely separate from the main house one, which is also regulated and filtered. Now, why this is nothing. Even if I didn't, do you really think you will hear that effect over an untreated room full of slap echos, poor modal distribution, loud uncontrolled resonances, lack of proper absorption and diffusion, and early arriving reflections? I don't think so, as these effects can effect impart frequency domain changes of over 40db between 20-40hz, and 10-20db at frequencies above that to beyond our hearing. That is a major issue, not whether there is power regulation or filtration within the system. Majoring in minors!

It appears it is you that has the problems and delimers with your own set up in your home and I am getting the impression you are clueless in sorting them out or building a High End system. Do you have more than ABC 'lets measure it' knowledge to work on?

Have you ever SEEN or HEARD my setup? Nope. What you are doing is called deflection. You are trying to deflect away the issues that I have pointed out in your so called test that invalidate it, and trying to create issues within my system that don't exist. This is rather poor form my friend, because it doesn't really change the issues much. You have never seen let alone heard my system, so you have no grounds for drawing any conclusions about its capabilities or lacks. Very poor form.

A little lesson for ya. You use the ABC "let's measure it" knowledge and process to get good accurate balance between channels(I bet you do that by ear), to get a flat as flat a frequency response in your system as you can, and to deal with any acoustical issues that contaminate your audio signals once they leave the speakers. This requires a "lets measure it" mentality. Once you have achieved an accurate foundation, you use your ears to tweak those results. Anyone who would use their ears to do this is a fool from day one, because your ears will never be as accurate at balancing channels as a level meter or a RTA would be.

If things were that easy everyone would be enjoying perfect Hi Fi.

Apparently things are much harder than even you perceive.

Futhermore loose the pretentious handle, you are not entitled to it unless dubed by a royal court of Europe.

I guess by this order you must be from the Royal court of Europe. Seems to me I can adopt any name, moniker, handle I choose. I live in America, and am therefore not under the jurisdiction of the Royal court of Europe

[Sir Terrence]

Originally Posted by axe79

It is clear to see the evolution on these Chips from the 9132 put in the PS3 to the 9134 found in the newer crop of players.
It is now seen on the dedicated Blue ray players that they all do Bit streaming on the high quality audio formats.
Manufactures in the development research must have concluded that this path was one that needed to be offered.

Playing devils advocate here because things are not always as they seem, and some of us are not as gullible as others to marketing schmeel.

Or that it was another way marketing can help sell the product for more money. The PS3 was not made to stream because you lose something in the process of streaming. The PS3 is optimized for the Bluray format in which streaming of the audio results in the loss of audio on extra content. Just because something cannot do something(that is out of spec by the way) does not mean something is wrong with the PS3. You don't put chipsets that stream in products that are not designed to stream data. Its a waste of money.

"The flagship Sony S5000ES is also the first Sony player to feature a Precision Clock Conditioner, designed to reduce the effects of jitter."

Now if it wasn't a problem why address this?

Because it could BECOME a problem, or is a problem that may, or may not exist. Just another perspective. When a audiophile sees this, they salivate even if it is not really needed at all. This is just another perspective.

Perhaps it needed to be sorted if the player was to be doing the processing itself and then sending it out!

Or maybe not! Some things are added for marketing purposes.

"All the components used are of the highest quality and include high-performance capacitors and digital-to-analogue convertors."

In England this player is currently regarded as the best that has been reviewed.
I would be surprised if there was differences between Bitstream and LPCM on this player as the problems appear to have been designed out.

The PS3 does not do A/D conversion, and therefore does not need the highest quality D/A analog converters. Those are only needed if one uses the analog outs on the player, not the HDMI out.

Another player that is flying off the shelves at the moment is the Panasonic DMP-BD55.
"Like the entry-level ’BD35, this player is also Blu-ray Profile 2.0, DivX and x.v.Colour compatible.So why does it cost more? The extra outlay gets you upgraded audio parts, including new capacitors, a new 192kHz/24bit DAC and a couple of tweaks to the HDMI output to help reduce jitter."

So Manufacturers are addressing jitter issues. Well, well.

Is it jitter they are addressing, or shielding the audio signals from the other processes going on within the HDMI cable? Only they know, not you and I. The bottom line here is that all those fancy new part may or may not do anything to help the player sound better, but only the marketing folks really know right? They may do nothing to help the sound, but might be an effective tool to help sell the player for more money. Given the way player prices have dropped(and profit margins as well) all of these manufacturers are looking for ways to increased those margins, even if it is a fancy marketing blitz that does it.

Just playing devils advocate, because all may not seem as it looks.

[Sir Terrence]

Originally Posted by welwynnick

I think this has been one of the hottest and most debated topics in most AV forums since the format war ended. People have adopted strong positions based on their own listening experience; the experience of others (professionals and users); and on theories that try to explain what may or may not cause differences.

Agreed, it is the hottest topic post war. And what makes it worse is that every Tom, Dick, and Harry is conducting their own little amateurish unscientific listening tests, and chiming in on the subject which does nothing more than muddy the waters on the issue even more. Much like the Dts vs DD shebang of the DVD days. Which one sounded better than the other was always subjective to the personal listening characteristics of the individual listeners, and nobody could really decide which sounded better.

What complicates things are comparisons suggesting that amplifier decoding is better, when other factors are causing the differences, and null results that don’t prove it one way or the other. I do take a scientific approach to this, and try to understand and explain what makes audio sound good, but I also trust my ears for various good reasons. I suppose I’ve spent half a lifetime trying to reconcile analytical and empirical assessment, and its not easy.

I love this

Once I corrected and repeated the test it became quite clear I had been mistaken, and in the system I was using and with the discs I used, there was a clear difference. For the avoidance of doubt, it wasn’t an enormous difference, not as great as getting room acoustics right, but it was pretty fundamental and well worthwhile.

Nick, here is my rub, and thank you for pointing it out. I have always thought that most people chase after improving equipment more than chasing after the real culprit in most rooms. Its acoustics. So if you get greater improvement from improving room acoustics, then it seems rather silly to me to keep chasing after a minuscule improvement that a equipment change can give. What most folks do is put the cart(the equipment) before the horse(which is the room) because they understand and hear a small difference, and have no knowledge whatsoever of the complexities of speaker/room interaction. They take the path of least resistance because it is easier, and does not require an understanding on a very complex issue that does more improvement overall.

Now to the important point. Just because I heard it in my system doesn’t mean that everyone should hear it as well. Its very easy to make mistakes in set-up and spoil the result – believe me, I know. Other systems may have a different digital audio replay architecture that will suppress any differences. For example, a very good player like the Sony S5000 may be good enough to offset the problems with LPCM transmission over HDMI, and give a null result in the same test. Alternatively, a good receiver like the Pioneer SC-07 may be insensitive to jitter, and will also make LPCM sound as good as bitstream.

Another very astute observation.

I have no idea about the outcome of those situations, and I’m all ears to those have the opportunity. What I did hear was a smaller difference with bitstream when using an expensive player (3800BDCI) against a cheaper player (BH200). I had already pre-supposed that bitstream would be dependant on the amplifier being used, but now it seems, its also dependant on the player as well. And that kind of makes sense.

Another reason why Tom, Dick, and Harry's conclusions based on their own amateurish testing at home makes any conclusions invalid. The interactive complexities of the player and receiver must be examined, and variables must be taken out of the procedure in order to validate and lend any legitimacy to the test. In order to identify these interaction, you have to have a great deal of knowledge on the process of processing, which most folks don't. You have to have equipment that can identify these interactions, and most folks don't. Even if all equipment variables have been answered, they take this equipment in to rooms that are too noisy to effectively evaluate any differences, and they never repeat the test to effectively see if they can identifiy any differences over and over again. A noticeable difference that cannot be repeated over and over again, is not a noticeable difference. And since nobody I hear chiming that bitstreaming is better than LPCM has conducted this as a DBT, and never reported repeatable differences, you can now understand my skepticism.

So all I’m saying is that bitstreaming DID make a difference in a level-matched, blind test using a particular selection of discs, players and amplifiers, but those differences may not be apparent AT ALL with other combinations of equipment.

So while we can say that bit-streaming can make a difference, we can't say that it can’t make any difference.

Regards, Nick

This is a null conclusion, which still makes any claim of superiority suspect. So in the end, the "decoding" might not be a player or receiver superiority argument, but the compromises in each's design might be more of an issue.

[Sir Terrence]

Originally Posted by welwynnick

Its not a two-man campaign by Axe79 and myself to corrupt every free-thinking person on the net, we are simply agreeing with the majority of people who have compared player and amp decoding where they have the facility to do that. Of course some of the anecdotal accounts are not reliable because people don't level-match properly, and I'm well aware of that. Which is why I DID level match to within 0.5dB - as close as I could get it with the SC886 - and is why I DID do blind testing to back up my own observations. I only came to this thread once I had done that, because I knew what the reaction would be.

Nick, what equipment did you use to level match within .5db? A pre-pro cannot really match volume within this tolerance without external help, no matter how good it is. Was it an electrical level match at the source, or an acoustical level match at the speakers output? Did you do a single or double blind test? A blind test is not really valid because somebody has to be switching, and you are going to know they are as well. Electronic random switching is much more accurate a method than manual switching that is for sure.

Notwithstanding all that, even though the rest of the World are doing comparisons that might not be scientific by your chosen standards, it doesn't mean that the results are wrong. They're not - bitstreaming does sound better in certain systems, and just because somebody cannot hear any difference does not mean that there isn't a difference.

It doesn't mean their results are wrong, but their results cannot be considered definitive and conclusive. If somebody cannot hear a difference, then there is no difference. If the difference is not repeatable tomorrow, the day after tomorrow, or the day after that, then there are no differences. I can convince myself of anything, but can I convince others becomes the question.

However, it could be that with some combinations of player and amplifier, the digital audio replay architecture may be similar with both schemes, and there could be no reason for them to sound different. What I do know now is that with the Onkyo amp that I used, there was a difference, and from what I hear elsewhere from reliable sources there are differences with many other amplifiers as well.

Then what this means to me is that which "decodes" the signal is not the issue, but the equipment design is the issue. Since the programming of the decoding is standardized, which decodes is a non issue. Its how the equipment is handling the audio before, and after decoding that is more the issue based on this statement.

I've put a great deal of effort into performing accurate and reliable comparisons over two weeks to reach this conclusion, and if you care to look around, you will see that I am being taken seriously.

regards, Nick

Let's be more precise here. SOME are taking you seriously, not everyone. However, you may have uncovered the real issue here, and that issue may not be what is doing the decoding, but how the audio is handled before and after that. It seems to me you have identified flaws in the equipments design, and not that bitstreaming is better than LPCM