INTERACT FORUM

Please login or register.

Login with username, password and session length
Advanced search  
Pages: [1]   Go Down

Author Topic: Ripped sound is different from the original Audio CD  (Read 10506 times)

AndreaT

  • World Citizen
  • ***
  • Posts: 120
Ripped sound is different from the original Audio CD
« on: May 07, 2012, 05:56:45 am »

Hi Folks,

I am testing the demo of MC 17 last build .144 and I have to notify that the sound reproduction I get is not as good as I was expecting.

I ripped one of my last purchased Audio CD. I used APE format, the default setting proposed by MC 17.

Then I run an AB comparison using my original CD via SPDIF to ASIO interface and MC 17 player via ASIO.

My setup is so made that my DAC just receives two ASIO streams on the same outputs.
So,  to run my comparison I just have to play/pause one at a time my two players (CD and MC17).

What I got, is that the two sound different!

Why?
Anybody can help me?

Thanks and regards,
Andrea
Logged

JimH

  • Administrator
  • Citizen of the Universe
  • *****
  • Posts: 72379
  • Where did I put my teeth?

Please follow the link to Getting Started in my signature.  Audio Setup in that topic covers WASAPI Event Style and ASIO, your best choices for quality.
Logged

jgreen

  • Citizen of the Universe
  • *****
  • Posts: 2419
Re: Ripped sound is different from the original Audio CD
« Reply #2 on: May 07, 2012, 01:08:07 pm »

Andrea--

Your ripped CD should not sound different, unless you want it to sound different, which you would do with setting in the DSP Studio (there's a button on the top right). 

If I understand correctly, you are comparing MC with a hardware CD player?  Is this right?  Or is it a different software player, playing the CD, which may have presets which you prefer (and which you can imitate in MC)?

Okay, reading it again you're doing ASIO on both sides, but I'm still betting that someone is changing the sound on you:  Either MC is, through settings in DSP  Studio, or your player is, and you prefer the altered sound.

Or, there's a problem hooking up to ASIO, on one side or another. 
Logged

mbagge

  • Galactic Citizen
  • ****
  • Posts: 255
Re: Ripped sound is different from the original Audio CD
« Reply #3 on: May 07, 2012, 01:34:45 pm »

It might also be worth a try to rip your CD in a losless format, eg. flac.
Logged

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #4 on: May 07, 2012, 03:06:05 pm »

It might also be worth a try to rip your CD in a lossless format, eg. flac.

Well - APE is lossless - but sure - rip to FLAC as well and and compare. Logically the two files (flac and ape) should be identical - as should the CD.

Over here - using ASIO for both CD and FLAC - there is no difference whatsoever in the "sound" of my output.

VP
Logged

AndreaT

  • World Citizen
  • ***
  • Posts: 120
Re: Ripped sound is different from the original Audio CD
« Reply #5 on: May 09, 2012, 03:34:23 am »

Dear JGreen,

yes, I am comparing hardware CD Player with MC17.
I also did it in the past comparing my HW CDP with Foobar2000 1.11, so it is a verified "testing chain and method".

Exactly to avoid big errors (like comparing apples and pears), in both cases the HW CDP is just used as a pure digital source and the audio stream is sent to my PC via SPDIF and then transferred to Allocator XOver (running on my PC) and then to the DAC via the DAC's ASIO driver.

As I stated, the audio CD was ripped by MC17 in the lossless "ape" format (as per your default parameters, I didn't change any).

I verified the DSP Studio settings and it is flat. I mean, I had to enable 5.1 output mode to have Allocator plug-in running, but except that, all other is off.
Of course, Allocator plug-in is in the chain and running.

The differences in sound I noticed are mostly related to the bass.
Where should I have made my mistake in the settings?
Can I provide you my configuration for your review?

Thanks again and kind regards,
Andrea.

PS for JimH: I am sorry but I haven't understood where to continue the discussion. Clicking on the link you gave in your re: subject, I couldn't browse away from the current page.
Sorry. Regards, Andrea.
Logged

JimH

  • Administrator
  • Citizen of the Universe
  • *****
  • Posts: 72379
  • Where did I put my teeth?
Re: Ripped sound is different from the original Audio CD
« Reply #6 on: May 09, 2012, 06:47:15 am »

PS for JimH: I am sorry but I haven't understood where to continue the discussion. Clicking on the link you gave in your re: subject, I couldn't browse away from the current page.
Sorry. Regards, Andrea.
Please try a different browser.  Here's the direct link:

http://wiki.jriver.com/index.php/Getting_Started
Logged

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #7 on: May 09, 2012, 07:23:17 am »

Exactly to avoid big errors (like comparing apples and pears), in both cases the HW CDP is just used as a pure digital source and the audio stream is sent to my PC via SPDIF and then transferred to Allocator XOver (running on my PC) and then to the DAC via the DAC's ASIO driver.

Andrea,

To ensure a fair test - I would suggest removing that Allocator XOver plugin from any playback - since as a frequency EQ - it can easily be imparting it's own will over the CD stream. I have hardware and software EQ's here that make playback sound vastly different - just by placing the plugin in the audio chain and leaving all the sliders or controls untouched.

With a plugin inline - the CD source is not being auditioned as a true bit perfect. Then ensure that MC is playing back it's audio in full bit perfect mode (no DSP at all - the little Audio Tools button in the upper right - looks like a set of EQ sliders - should be glowing "blue" during playback)

After you clear out all possible sources of coloring - report back. You should not be hearing any difference - especially in the bass area. You also did not specify what audio interface is being used that requires ASIO... I have never heard of a DAC that uses or requires ASIO?

Cheers,

VP
Logged

AndreaT

  • World Citizen
  • ***
  • Posts: 120
Re: Ripped sound is different from the original Audio CD
« Reply #8 on: May 10, 2012, 11:46:38 am »

Reply for JGreen

Dear JGreen,

today I used a Foobar tool, "Bit Comparator" to analyze if the cause of my "perceived" sounding "not as the source" could be in the generated compressed APE track.
Well, I don't know if what reported by this tool can be 100% true, but it reported that the APE track generated by ME17 is errored.
Then I set ME17 to generate a FLAC, and also the FLAC was reported to be errored.

To crosscheck the matter, I generated the same FLAC using Foobar driving the flac.exe directly.
Well, in this case, no errors were detected.

Herein are the diagnostic ouputs of the "Bit Comparator"

Test 1
=====
Using Foobar2000 1.1.12 beta5 to generate the FLAC and then testing it directly against the source CD
Codec used: Official FLAC 1.2.1b

All tracks decoded fine, no differences found.

Comparing:
"C:\Temp\Temp2\FOLLIA D'AMORE.flac"
"cdda://01979DC2" / index: 1
No differences in decoded data found.


Test 2
====
Using ME17 last build 147 to generate the APE (default ME17 settings) and then testing it directly against the source CD
Codec used: ME17 own APE codec

Differences found in 1 out of 1 track pairs.

Comparing:
"C:\Parcheggio\Raphael Gualazzi\Reality and Fantasy\01 - Follia D'Amore.ape"
"cdda://01979DC2" / index: 1
Differences found: 18693670 sample(s), starting at 1.3568254 second(s), peak: 1.9999084 at 208.0399547 second(s), 2ch



Test 3
====
Using ME17 last build 147 to generate the FLAC (set to level 8) and then testing it directly against the source CD
Codec used: ME17 own FLAC codec

Differences found in 1 out of 1 track pairs.

Comparing:
"C:\Temp\Raphael Gualazzi\Reality and Fantasy\01 - Follia D'Amore.flac"
"C:\Temp\Temp2\RAPHAEL GUALAZZI\REALITY AND FANTASY\01. FOLLIA D'AMORE.flac"
Differences found: 18693670 sample(s), starting at 1.3568254 second(s), peak: 1.9999084 at 208.0399547 second(s), 2ch

Any idea about these errors?



Reply for Vocalpoint

Dear Vocalpoint,

actually I cannot get rid of Allocator because my stereo system is fully DSP 3way Xovered and no passive filter (except the loudspeaker wires) is in place.
I need to have 6 outputs to get something "normally" playing out of my loudspeakers.
My DAC is a MOTU 896 mk3 Hybrid. ASIO is its "standard" interface to PC... unless having a Studio Console like PC with optical ATAPI or SPDIF/TOS interfaces....
But I have just a Dell D830 with USB and Firewire...

Kind regards,
Andrea.

Logged

JimH

  • Administrator
  • Citizen of the Universe
  • *****
  • Posts: 72379
  • Where did I put my teeth?
Re: Ripped sound is different from the original Audio CD
« Reply #9 on: May 10, 2012, 11:56:46 am »

Make sure you have no DSP on.
Logged

Scolex

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 1116
  • Cheers
Re: Ripped sound is different from the original Audio CD
« Reply #10 on: May 10, 2012, 12:21:56 pm »

Just a couple of thoughts/ideas.
Enabling 5.1 through SPDIF can reduce quality because the SPDIF pipe is rather small requiring the audio to be compressed the rate of compression varies depending on encoding type (DD, DTS, etc).
If you have Do not play silence enabled in tools>options>audio>track change disable it and see if the comparator changes.
Logged
Sean

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #11 on: May 10, 2012, 12:47:20 pm »

Actually I cannot get rid of Allocator because my stereo system is fully DSP 3way Xovered and no passive filter (except the loudspeaker wires) is in place.
I need to have 6 outputs to get something "normally" playing out of my loudspeakers.

Could you give me more details on exactly what kind of stereo system are you running -that would have a configuration like this? I have never heard of needing 6 inputs either to get a stereo signal working. Most us just have:

Computer->DAC->Input on Integrated amp or Receiver.

The audio from MC is bit perfect and logically is the truest possible representation of the source material. But if your stereo system is "coloring" the CD playback in some way - then you will never solve this. Whatever kind it is - there cannot be ANY DSP in-line on your system - to make a proper comparison.

As mention by Scolex - having any kind of 5.1 processing or filtering on during stereo playback can/will also skew the results. It is not the same as serving up a pure 2 channel source feed.

Over here for testing -I put my OPPO BDP-83 (CD Playback) directly to my Music Hall DAC (via TOSLINK) and the computer audio stream directly to the DAC (via SPDIF - from an MC based computer).

Playback of CD and FLAC of the same source simultaneously while switching inputs on the Music Hall - I get absolute indistinguishable playback. It's impossible for me to tell which is which - outside of staring at which DAC input is active.

My DAC is a MOTU 896 mk3 Hybrid. ASIO is its "standard" interface to PC... unless having a Studio Console like PC with optical ATAPI or SPDIF/TOS interfaces....
But I have just a Dell D830 with USB and Firewire...

Now - I am familiar with this. However - it's a digital audio interface rather than a pure DAC.

Can you elaborate on how you are connecting the outputs of the MOTU to your "stereo system" and what kind of connectors/cables are being used for this?

VP
Logged

Phil LD

  • Junior Woodchuck
  • **
  • Posts: 75
Re: Ripped sound is different from the original Audio CD
« Reply #12 on: May 10, 2012, 01:18:41 pm »

Well, I also did serious testing using a Benchmark Dac1 Pre and Sennheiser HD 800. No DSP Effects turned on.

I used an Onyko SACD player with a digital connection to the Benchmark DAC1 and the computer over usb.
The SACD Player and the computer were running the same track at the same time, I switched sources with the Dac1.

I found no differences when using MC 17 with Wasapi in Exclusive mode. --> Pure music
I found a annoying kind of harshness compared to the Onkyo SACD player when using ASIO4All --> rubbish
I found normal WASAPI to sound much better than WASAPI Event Style, don't know why... Maybe because the Benchmark is not Asynchronus? --> Not working for me.
I found MC17 to sound better and much closer, even similar, to the SACD Player than MusicBee or foobar using WASAPI. --> MC 17 wins over all others when audio quality is required.
By the way, I also tested the iPad 3 with B&W P5 headphones --> complete rubbish, not the headphones, but the iPad in terms of sound.
Logged

jgreen

  • Citizen of the Universe
  • *****
  • Posts: 2419
Re: Ripped sound is different from the original Audio CD
« Reply #13 on: May 10, 2012, 02:01:41 pm »

Excellent point that the Comparator may be seeing bit differences due to audio options regarding "leading silence..." or fadein/fade out.  These settings are under audio properties in tools/options/etc.  In any case, these settings would NOT affect the sound difference you heard. 

As JimH says again, please spend some time in the DSP studio rooting around for any FX that might be enabled.  Also, please allow for the "possibility" that there is some audio FX on your comparison path, and that you prefer this effect. 

As for the testing path you have constructed, many others here are more qualified to judge the suitability of this--I have trouble getting a lightbulb screwed in properly.  But I too am a little concerned over the introduction of 5.1. 

Good luck.  MC outputs unaltered audio when set to do so.
Logged

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #14 on: May 10, 2012, 02:09:25 pm »

Also, please allow for the "possibility" that there is some audio FX on your comparison path, and that you prefer this effect.

This is exactly what I was thinking.

This "stereo system" may very well have it's own unique sonic signature due to "something' being in-line during playback and if that's been the case for a long while - this specific signature has now become expected.

Then to suddenly compare this "sound" to MC's pure bit-perfect output - can be dramatic - but the fact is there is a difference - comparing one signal with some sort of DSP to another signal with none.

VP
Logged

AndreaT

  • World Citizen
  • ***
  • Posts: 120
Re: Ripped sound is different from the original Audio CD
« Reply #15 on: May 11, 2012, 02:50:21 pm »

Let me first thanks all for your replies.

However, I am sorry, I see that the core discussion gets lost in the fog.

Anyway, if you are sure that MC17 is encoding and decoding perfectly, at the opposite I still have some dubts about.

I carefully read all the intros and faqs (following JimH link), and, as I can verify, I didn't enabled any DSP effect: absolutely nothing.

I just had to tell MC17 to manage output in 5.1 because Allocator needs 6 outputs to go, and setting different it doesn't play.
MC17 hasn't any other way to manage the ASIO output mapping.

Unfortuantely I cannot test WASAPI because my DAC has only ASIO sw drivers.

And, unfortunately twice, MC17 hasn't any off-line (post rip) generated track to original source comparator.

So, whatever we can say more is absolutely opinable.

Regards and good development to all the MC17 team.
Andrea.
Logged

pcstockton

  • Citizen of the Universe
  • *****
  • Posts: 1261
Re: Ripped sound is different from the original Audio CD
« Reply #16 on: May 11, 2012, 03:07:32 pm »

I am not surprised in any way that you are getting variable SQ between the two methods given that you are re-feeding signals back through a digital crossover.  In one case you are using taking the SPDIF from CDP.  In the other you are generating a 6 channel from a media player.

Question... Assuming your CDP outputs 2 channel audio, why is it necessary to make MC output 6 channels?  Why can't your digi crossover take 2 channels from MC?

This is not any way I would want to run an "active" set-up.  Does your speaker mfr not make an external (and presumably active) crossover?

You have far too many things in the replay chain that can create errors and/or impart something that changes the sound.

Just my 2 cents,
Patrick
Logged
HTPC (ASRock Mini PC 252B: i5 2520M Sandy Bridge/HD3000 - 2.5 GHz - 8GB RAM - 256GB Intel SSD - Win7 Home) > MF V-Link 192 > Wireworld Ultraviolet > Naim DAC > Naim NAC 102/NAPSC/HiCap (PSU) > Naim NAP 180 Amp > Naim NACA-5 Speaker Cables > Naim Ariva

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #17 on: May 11, 2012, 03:28:08 pm »

Anyway, if you are sure that MC17 is encoding and decoding perfectly, at the opposite I still have some dubts about.

While I have no doubts at all that MC is serving up bit perfect audio everytime - I must agree with pcstockton  - your systems are too far apart to allow for a proper, fair comparison.

If you had a simple method of getting your CD audio stream going to a clean standalone DAC and a computer stream coming into the same DAC and could switch between them - you would at least be in the ballpark to compare - but as it stands - it's just not apples to apples at this point.

VP
Logged

Scolex

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 1116
  • Cheers
Re: Ripped sound is different from the original Audio CD
« Reply #18 on: May 11, 2012, 04:42:23 pm »

I am confused how are you getting the reference disc from the CDP to the comparator.

This is my assumption on how things flow
CDP SPDIF-->MOTU 896-->USB-->MC (encode 5.1)-->Allocator-->comparator
MC (encode 5.1)-->Allocator-->comparator

If the above is not the process then it is not a fair comparison, since the mostly likely causes for a change in sound quality are the 5.1 encode and/or Allocator

This is my assumption for playback
CDP SPDIF-->MOTU 896-->USB/FW-->MC (encode 5.1)-->Allocator-->USB/FW-->MOTU 896-->Analog XLR outputs-->Multi-channel inputs on a receiver or separate amplifiers
MC (encode 5.1)-->Allocator-->USB/FW-->MOTU 896-->Analog XLR ouputs-->Multi-channel inputs on a receiver or separate amplifiers

If my assumptions are incorrect please list the flow in a similar manor. I am just trying to figure out what it is exactly you are doing.
Just an FYI MC17 has native features that allow you to remap channels and apply crossovers.
Logged
Sean

natehansen66

  • World Citizen
  • ***
  • Posts: 239
Re: Ripped sound is different from the original Audio CD
« Reply #19 on: May 11, 2012, 07:40:57 pm »

Is there any DSP in the Motu input that might be mucking things up? You said you did the comparison with Foobar and it came out good, maybe something changed in the Motu's settings before you did the comparison with J River?The only variable here is passing the audio from the cdp through the Motu into the PC (besides the MC rip)....

For those not aware how active systems work Scolex's description is right on, MC splits the original 2 channels into 6 for AndreaT's active speakers. It will do this for the signal from the cd player, as well as from the ripped file, so MC is doing this in both instances in exactly the same way. Also, Allocator is a plugin that runs within JR's DSP section, not an external crossover. I'll venture a guess that he's got diy 3-way speakers, so each one needs 3 channels of amplification. MC 5.1 encoding is not the reason for any perceived differences.

Even if there was any other DSP enabled, as long as it was the same for both comparisons (no changes made) the sound should be the same. It'll be interesting to see the result of this........
Logged

pcstockton

  • Citizen of the Universe
  • *****
  • Posts: 1261
Re: Ripped sound is different from the original Audio CD
« Reply #20 on: May 11, 2012, 09:25:52 pm »

*
Logged
HTPC (ASRock Mini PC 252B: i5 2520M Sandy Bridge/HD3000 - 2.5 GHz - 8GB RAM - 256GB Intel SSD - Win7 Home) > MF V-Link 192 > Wireworld Ultraviolet > Naim DAC > Naim NAC 102/NAPSC/HiCap (PSU) > Naim NAP 180 Amp > Naim NACA-5 Speaker Cables > Naim Ariva

pcstockton

  • Citizen of the Universe
  • *****
  • Posts: 1261
Re: Ripped sound is different from the original Audio CD
« Reply #21 on: May 11, 2012, 09:43:20 pm »

Quote from: natehansen66 link=tpic=72004.msg487814#msg487814 date=1336783257

For those not aware how active systems work

That isn't how "active systems" work in my world.  That is just one convoluted way to run "active speakers" without real crossovers.

An "active system" takes the signal from the preamp and sends it to a powered (ACTIVE) external crossover where the electronics separate the frequencies needed (the active crossover is dialed in for the specific speaker it is used with), with very fine precision respecting proper grounding and stout linear power supplies.  This could be 2 or 3 (or more i suppose) signals.  Then the signals are sent along to amplifiers (3 stereo or 6 monoblocks in the OP's case) which are then coupled to the corresponding drive units on the speaker.




Logged
HTPC (ASRock Mini PC 252B: i5 2520M Sandy Bridge/HD3000 - 2.5 GHz - 8GB RAM - 256GB Intel SSD - Win7 Home) > MF V-Link 192 > Wireworld Ultraviolet > Naim DAC > Naim NAC 102/NAPSC/HiCap (PSU) > Naim NAP 180 Amp > Naim NACA-5 Speaker Cables > Naim Ariva

AndreaT

  • World Citizen
  • ***
  • Posts: 120
Re: Ripped sound is different from the original Audio CD
« Reply #22 on: May 12, 2012, 12:16:06 pm »

Dear Frieds,

OK, let me try to reply to all of your questions. ... if only this forum could allow me to attach docs and pictures it will be all a lot easier...

Here is the schema of the setup.

Chain 1): CDP used as source.
My Audio CD in my CDP => CDP via SPDIF 44.1 kHz => my PC => Allocator with my XOver setup => MOTU ASIO Driver => MOTU 896 mk3 DAC 6 analog outputs

Chain 2): MC17 used as source.
MC17 reading a 44.1 kHz FLAC/APE encoding of the same my own CD Audio => DSP Studio plug-in just to have 5.1 settings for ASIO 6 output enabling => Allocator plug-in with the same my XOver setup => MOTU ASIO Driver => MOTU 896 mk3 DAC 6 analog outputs

No other settings on MC17. All as "clean as possible".
If somebody tell me how to copy and provide you all the config, I will do!
Volume is set to be the same (carefully checking the DAC VU and peak meters).

To compare AB the two chains I simply start (with my CDP remote) the CDP and the MC17 players concurrently and then I mute one or the other (this becasue ASIO support mutiusers and Allocator is multi instance so I can run two instances of Allocator separately).

Now, listening to one and to the other, my impression is that MC17 has some difference in the details (bass mostly).
It may be just my mistakes... and this is why I posted here this new thread.

However, I also did the same with Foobar2000 1.1.12-beta5, and in this case no difference is detectable by my "hears".

I don't want to say that MC17 is bad or not sounding better than Foobar, but I would understand why I am having these results and why doing a bit comparation of the track ripped by MC17 I get errors.

That's all.
Again, kind regards to all,
Andrea.
Logged

Frobozz

  • Citizen of the Universe
  • *****
  • Posts: 641
  • There is a small mailbox here.
Re: Ripped sound is different from the original Audio CD
« Reply #23 on: May 12, 2012, 12:31:06 pm »

Dear Frieds,

OK, let me try to reply to all of your questions. ... if only this forum could allow me to attach docs and pictures it will be all a lot easier...


You can upload pictures to a host like ImageShack or imgur.  Then link to those images in your post.
Logged

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #24 on: May 12, 2012, 12:33:01 pm »


Allocator with my XOver setup

DSP Studio plug-in just to have 5.1 settings for ASIO 6 output enabling => Allocator plug-in with the same my XOver setup

All as "clean as possible".

Andrea,

Apologies - but having any of the above does not represent "as clean as possible". With ANY DSP or ANY plug-in in the chain whatsoever - you are immediately skewing the results and you cannot make a clean and fair comparison.

Q: Are you really sending the CD feed "into" the PC? Can you not send it directly into the MOTU?

NOTE: As soon as you turn on anything in DSP Studio in Media Center - you immediately negate the bit perfect audio stream.

A true "clean" audio path for comparison to me would be:

1. CD->Digital Cable (SPDIF or TOSLink)->DAC->Amp->Speakers
2. Computer (MC sending bit perfect stream - which is NOT possible with ANY DSP active BTW)->Digital Cable (SPDIF/TOSLink/USB)->DAC->Amp->Speakers

A true comparison can only be attained with NO DSP anywhere. I would even go as far to say a true comparison can only be made without ANY multichannel activity anywhere (5.1 etc)

Your system is full of DSP - and while it's unfortunate that you cannot seem to get a normal 2 channel stereo feed going without adding in 5.1 plugins and Allocator and all that other stuff - but your question and concern cannot ever be properly discussed with all the audio going through all this "stuff".

All I can say is - when a "clean" chain as suggested above is used - there is no difference whatsoever with my CD source (via Oppo BDP-83SE) and Media Center (Bit perfect).

VP
Logged

Scolex

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 1116
  • Cheers
Re: Ripped sound is different from the original Audio CD
« Reply #25 on: May 12, 2012, 01:10:53 pm »

Chain 1): CDP used as source.
My Audio CD in my CDP => CDP via SPDIF 44.1 kHz =>WHAT DEVICE IS HERE=> my PC => Allocator with my XOver setup => MOTU ASIO Driver => MOTU 896 mk3 DAC 6 analog outputs

You state "But I have just a Dell D830 with USB and Firewire" with that in mind there has to be some device used for the SPDIF input.

To rule out any possible outside influences why not use the optical drive in the Dell instead of the CDP when running the comparator.

Have you gone in and disabled Don't play silence, one other thing that can be the cause is if you used "Copy Mode: Secure" when ripping the disk.
With secure enabled the rip can be different because if MC encounters any blemishes in the surface that effect disc readability it will rescan that
area of the disc attempting to get a good read. The CDP does not have this technology and will just send what is read.
To rule this factor out you have 2 options:
1 - Create an ISO of the reference disc and mount it with daemon tools then rip the ISO image
2 - Rip using "Copy Mode: Normal"
Logged
Sean

SamuelMaki

  • World Citizen
  • ***
  • Posts: 115
Re: Ripped sound is different from the original Audio CD
« Reply #26 on: May 12, 2012, 01:38:08 pm »

Have you tried the option play 2.0 in 5.1 container? Just for debugging, does it sound different than the audio-cd?
Logged
AMD Phemon II N930 Quad 2.0ghz, AMD Radeon HD 5650, 4GB RAM, Windows 8 RP x64

RC23

  • World Citizen
  • ***
  • Posts: 125
Re: Ripped sound is different from the original Audio CD
« Reply #27 on: May 12, 2012, 04:28:46 pm »

Here is the schema of the setup.

Chain 1): CDP used as source.
My Audio CD in my CDP => CDP via SPDIF 44.1 kHz => my PC => Allocator with my XOver setup => MOTU ASIO Driver => MOTU 896 mk3 DAC 6 analog outputs

Chain 2): MC17 used as source.
MC17 reading a 44.1 kHz FLAC/APE encoding of the same my own CD Audio => DSP Studio plug-in just to have 5.1 settings for ASIO 6 output enabling => Allocator plug-in with the same my XOver setup => MOTU ASIO Driver => MOTU 896 mk3 DAC 6 analog outputs

No other settings on MC17. All as "clean as possible".
Volume is set to be the same (carefully checking the DAC VU and peak meters).

I am also as Andrea an user of Allocator which delivers as DSP XOver fantastic results compared with passive XOvers solutions. The allocator setup is only usable by DSP! The JRiver workaround with 5.1 delivers a signal for a 3-way active XO. Allocator works as a VST-plugin in collaboration with ASIO which is also a superb solution as WASAPI. More infos: http://www.thuneau.com/allocator.htm

I expect as a to-be user of MC17 a better sounding DSP solution with MC17. If the setup of MC17 sounds inferior as Allocator with Foobar the DSP integration of MC17 has to be checked.

In my coming MC17 setup I will compare Allocator with Acourate, which is a next-generation active XOver with the opportunity of convolving and some other sound tuning points.
Logged

natehansen66

  • World Citizen
  • ***
  • Posts: 239
Re: Ripped sound is different from the original Audio CD
« Reply #28 on: May 12, 2012, 11:15:33 pm »

That isn't how "active systems" work in my world.  That is just one convoluted way to run "active speakers" without real crossovers.

There are many ways to skin the audio cat!  ;D

Your system is full of DSP - and while it's unfortunate that you cannot seem to get a normal 2 channel stereo feed going without adding in 5.1 plugins and Allocator and all that other stuff - but your question and concern cannot ever be properly discussed with all the audio going through all this "stuff".

Despite the DSP needed within J River to get Allocator to function properly, it "shouldn't" sound any different from the CD (going through the same "stuff" being Allocator) as long as there isn't any extra eq added in JR. All the 5.1 is doing is enabling the extra channels for Allocator to split the freqs. I do agree that the output isn't going to be bit-perfect, but then neither is the output from the CD after it's been run through Allocator. In my view that "not perfect" output should sound the same in either case.

MC doesn't add any channel delays when going from 2 channels to 6, does it? I'd hope that it doesn't, and I can't say I notice anything in my system, which is pretty similar to AndreaT's. I can't say I've done any comparisons like this though.


Logged

Vocalpoint

  • Citizen of the Universe
  • *****
  • Posts: 2007
Re: Ripped sound is different from the original Audio CD
« Reply #29 on: May 12, 2012, 11:56:23 pm »

All the 5.1 is doing is enabling the extra channels for Allocator to split the freqs.

Well - each to their own - but for me - if it's not "bit perfect" then I wouldn't even bother trying to compare the two streams. Splitting frequencies and so on? Who knows what kinda of changes are being applied here....

I am not disagreeing that it "should" sound the same if the same DSP is being applied to both CD and computer stream - but the OP is still not hearing it. If that's the case - there is something skewing the result. And if that's the case - it's kinda pointless to keep on about this if DSP is anywhere in the chain. It will never sound right.

VP


Logged

Scolex

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 1116
  • Cheers
Re: Ripped sound is different from the original Audio CD
« Reply #30 on: May 13, 2012, 12:48:21 am »

If I had to venture a guess I would say the difference is in the input chain from the CDP or the fact that when you enable 5.1 over SPDIF in MC it bumps the sample rate to 48K.

Just to cure my curiosity I did a test of my own and ripped a single song from 2 different CDs with MC and
Foobar using default FLAC encoder settings and using both normal and secure settings for both apps and discs.

CD1 perfect condition (bought ripped and put away) comparison of CD and all 4 rips were identical
CD2 some minor blemishes but does not skip during playback comparator showed differences between the CD and all 4 rips, the level of difference varied from rip to rip.

Nothing changed at any time other than changing apps and rip modes.

I had all of the reports spread out across my desktop and was going to screen cap them but must have had a
brain cramp cause I closed foobar which closed the reports and I don't feel like sitting through it again.
Logged
Sean

Alex B

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 10121
  • The Cosmic Bird
Re: Ripped sound is different from the original Audio CD
« Reply #31 on: May 13, 2012, 05:05:58 am »

I am surprised that no one has already said this, but the possible ripping errors never exhibit themselves as "changed audio quality". I.e., for example, the bass fidelity is always the same. If audible ripping errors exist in the files, they are short glitches, like clicks or skips. The possibly existing audio fidelity differences are caused by the differences in the playback chain.

Regarding the MC vs foobar2000 ripper comparisons, foobar2000 may have applied so called "offset correction" which slips the entire audio content on the timeline a bit and thus makes a direct comparison with ripped audio that has not been offsetted to fail, even though the actual audio content may be identical.

As Scolex explained, in addition to checking the secure rip log, a way to confirm that the ripped files are error free is to rip the same tracks more than once with MC. E.g. first in FLAC or APE format, then in WAV format (uncompressed is a bit faster), and then compare the resulting files with the foobar2000 bitcomparator. It is not likely that the possibly existing bad areas on the CD would produce the same ripped data twice.
Logged
The Cosmic Bird - a triple merger of galaxies: http://eso.org/public/news/eso0755

Alex B

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 10121
  • The Cosmic Bird
Re: Ripped sound is different from the original Audio CD
« Reply #32 on: May 13, 2012, 06:31:46 am »

My Audio CD in my CDP => CDP via SPDIF 44.1 kHz =>WHAT DEVICE IS HERE=> my PC => Allocator with my XOver setup => MOTU ASIO Driver => MOTU 896 mk3 DAC 6 analog outputs

You state "But I have just a Dell D830 with USB and Firewire" with that in mind there has to be some device used for the SPDIF input.

AndreaT,

I am not familiar with Allocator, but in addition to "what device is here", there must be some kind of host program that grabs the SPDIF input, hosts the Allocator VST plugin, and outputs the 6 filtered audio channels. Possibly the used host program comes with Allocator (the www.thuneau.com site mentions the included VSTShell component), but the playback chain is different, unless the host is MC in both cases. MC has the recently introduced ASIO line-in feature, but I don't know if it would work for this purpose.
Logged
The Cosmic Bird - a triple merger of galaxies: http://eso.org/public/news/eso0755

craigmcg

  • World Citizen
  • ***
  • Posts: 224
Re: Ripped sound is different from the original Audio CD
« Reply #33 on: May 13, 2012, 07:35:13 am »

Just to clarify something (I'm not an expert)- the Subject is that the ripped audio sounds different than the original CD but unless I'm missing something, different hardware is involved in the current comparison.

If the point is to compare the ripped audio (first ensuring that there are no errors in the rip) to the CD itself, I suggest playing the CD from the PC's cd drive using MC17 then comparing it to the sound of the ripped audio through MC17. In this way the same hardware is used and the comparison is between the files themselves.

If different hardware is used, I don't see how it is possible to know whether a difference (real or perceived) is caused by the different file types, the the hardware in use, or a combination of both.

I'm not surprised that you have a preference for one sounds vs. the other- I just don't think you can isolate the reason that you do,

Logged

InflatableMouse

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 3978
Re: Ripped sound is different from the original Audio CD
« Reply #34 on: May 13, 2012, 08:04:29 am »

If different hardware is used, I don't see how it is possible to know whether a difference (real or perceived) is caused by the different file types, the the hardware in use, or a combination of both.

Burn the APE back to an audio CD and play it in the hardware player.
Logged

NickF

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 860
Re: Ripped sound is different from the original Audio CD
« Reply #35 on: May 13, 2012, 06:05:09 pm »

The critical problem in the way Andrea is using MC is the conversion of a 2 channel audio stream into 5.1.  This involves some complex Dolby processing to create a simulated surround sound.  There is no way that this, when converted back to stereo by the other components by splitting into 6 frequency based channels, is going to sound the same as the CDP input.  Absolutely impossible.  I don't understand why Allocator needs 6 ASIO input channels.  This is what needs to be addressed.  SamuelMakei suggested wrapping the 2 channels in a 5.1 container.  This would probably resolve the problem.  This is done in Tools > Options > Audio > DSP & output format > Output Format.  Check Output Format and select 2 channels (inside 5.1 channel container) from the drop down list for Channels.  This will mean the original 2 channel stereo audio will remain unchanged.

Edit: Having read some of the Allocator user guide, it only requires a 2 channel input, as you would expect.  So why does MC need to be set to 5.1?  What audio device is MC seeing?  I would expect this to be Allocator.  Does Allocator present an ASIO interface to MC.  it doesn't say anything about this in the user guide.  Some clarification from Andrea would be helpful here.

Nick.
Logged
HTPC - Intel i5-760 CPU, Windows 7 64 bit, NVIDIA GTS450 Silent, RME 9632 with A04, BlackGold BGT3600; Video Processor - Lumagen Radiance XD Processor; Projector - ProjectionDesign Action Model 3 1080; Denon AVC-A1HD; 4 x Tannoy Berkley and Velodyne DD-10

NickF

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 860
Re: Ripped sound is different from the original Audio CD
« Reply #36 on: May 14, 2012, 04:50:31 am »

I apologise for my previous post.  I had not understood Andrea's configuration properly.

I believe Andrea is using Allocator as a plug-in within MC's DSP environment.  As Allocator preforms the frequency band analysis for the different loudspeaker drivers inside MC, a 6 channel output is required, 3 for each loudspeaker.  So MC's output is then routed directly to the external MOTU device using 6 channel ASIO over USB.

This pathway is using two DSP components.  The CDP playback path does not use any MC DSP components and presumably uses the native Allocator environment, not as a MC plug-in.

I think I understand the setup now but can't explain the difference in sound if the Allocator settings are really the same for the two different paths.

Nick.
Logged
HTPC - Intel i5-760 CPU, Windows 7 64 bit, NVIDIA GTS450 Silent, RME 9632 with A04, BlackGold BGT3600; Video Processor - Lumagen Radiance XD Processor; Projector - ProjectionDesign Action Model 3 1080; Denon AVC-A1HD; 4 x Tannoy Berkley and Velodyne DD-10

fooze

  • Junior Woodchuck
  • **
  • Posts: 72
  • TURN IT UP
Re: Ripped sound is different from the original Audio CD
« Reply #37 on: May 14, 2012, 08:34:55 am »

Try soloing all your channels one at a time in the Room Correction part of DSP in Media Center. Is each one playing correctly?

Logged

NickF

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 860
Re: Ripped sound is different from the original Audio CD
« Reply #38 on: May 14, 2012, 08:40:36 am »

Try soloing all your channels one at a time in the Room Correction part of DSP in Media Center. Is each one playing correctly?


Andrea isn't using the 5.1 channels in the conventional way for surround, hence room correction isn't relevant.    The six channels are for an active crossover.  Playing them solo will reveal nothing as each will only be a frequency band.

Nick.
Logged
HTPC - Intel i5-760 CPU, Windows 7 64 bit, NVIDIA GTS450 Silent, RME 9632 with A04, BlackGold BGT3600; Video Processor - Lumagen Radiance XD Processor; Projector - ProjectionDesign Action Model 3 1080; Denon AVC-A1HD; 4 x Tannoy Berkley and Velodyne DD-10

fooze

  • Junior Woodchuck
  • **
  • Posts: 72
  • TURN IT UP
Re: Ripped sound is different from the original Audio CD
« Reply #39 on: May 15, 2012, 12:48:05 am »

Nick, it will reveal if they are actually playing through the correct speaker driver. Or at all. I have a feeling MC isn't set up to route these frequencies in the same way the other setup is.
Logged

NickF

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 860
Re: Ripped sound is different from the original Audio CD
« Reply #40 on: May 15, 2012, 01:44:51 am »

Nick, it will reveal if they are actually playing through the correct speaker driver. Or at all. I have a feeling MC isn't set up to route these frequencies in the same way the other setup is.
I understand.  I'm not sure how or where the mapping of channels is done.  In the version of Allocator running outside MC, there is an option to select channels for the frequency bands but I couldn't find this in the MC Plug-in version.

Nick.
Logged
HTPC - Intel i5-760 CPU, Windows 7 64 bit, NVIDIA GTS450 Silent, RME 9632 with A04, BlackGold BGT3600; Video Processor - Lumagen Radiance XD Processor; Projector - ProjectionDesign Action Model 3 1080; Denon AVC-A1HD; 4 x Tannoy Berkley and Velodyne DD-10

jgreen

  • Citizen of the Universe
  • *****
  • Posts: 2419
Re: Ripped sound is different from the original Audio CD
« Reply #41 on: May 15, 2012, 04:24:30 pm »

Anyone heard from the OP lately? 
Logged

natehansen66

  • World Citizen
  • ***
  • Posts: 239
Re: Ripped sound is different from the original Audio CD
« Reply #42 on: May 15, 2012, 05:43:37 pm »

Nick, it will reveal if they are actually playing through the correct speaker driver. Or at all. I have a feeling MC isn't set up to route these frequencies in the same way the other setup is.

Nothing wrong with channel mapping in MC, I've used the Allocator demo in MC with no problems as far as channel output. It's pretty straight forward.

I can also say that the 5.1 and 7.1 output doesn't do anything funny to the signal as far as I can tell. I'm using the PEQ in MC to do the eq and crossovers for my active setup, and the eq needed isn't any different than what I would need for a similar flat response in miniDSP....
Logged
Pages: [1]   Go Up