It can all be explained in one word. "Jitter" (timing variations due to PSU etc.)
Yes, the checksums can still be identical if done correctly.
I get it, it's like some kind of digital audio homeopathy. Some essence that isn't conveyed in the actual bits is transferred, a kind of memory of the original music, the emotions of the musician, attach themselves to every copy of a recording, imperceptible to any instrument but clear as day to those with the ears to hear.
Come on sandyk, did you write this to provoke me? Should I report your post as disrespectful to members? Because it is beyond anything that any reasonable person should be asked to accept, that there is a difference between a file ripped to a USB stick and a file ripped to a HD and then copied to a USB stick.
You cannot be serious. I mean: you simply cannot be serious.
w
Computer Audiophile conducted a symposium some time back , where there were many industry figures present. The CA members also include the Chesky brothers.
There were carefully controlled tests at this symposium, and there was almost complete agreement among the attendees that SSD sounded better than HDD when playing identical files. As Ripley would say "Believe It Or Not "!
SandyK
P.S.
Chris Connaker, who is the Editor of Computer Audiophile , has stated that "Spinnig HDDs muddy the waters..."
There were carefully controlled tests at this symposium, and there was almost complete agreement among the attendees that SSD sounded better than HDD when playing identical files. As Ripley would say "Believe It Or Not "!
SandyK
P.S.
Chris Connaker, who is the Editor of Computer Audiophile , has stated that "Spinnig HDDs muddy the waters..."
Last edited:
Jitter of HDD
No, I do not believe that there is any difference between HDD and SS memory playback. The data from the drive, be it mechanical or solid state, is buffered in the computer ram, so this is a non issue. There are some manufacturers that would suggest otherwise, but I believe it is more of a selling point they are making, as opposed to reality.
No, I do not believe that there is any difference between HDD and SS memory playback. The data from the drive, be it mechanical or solid state, is buffered in the computer ram, so this is a non issue. There are some manufacturers that would suggest otherwise, but I believe it is more of a selling point they are making, as opposed to reality.
I'm curious over the details of what experiments he conducted to reach this conclusion. For example did he try a PC containing a spinning hard disk which wasn't being accessed for data? There might conceivably be power supply fluctuations from HDDs which aren't present when SSDs are used.
Richard
I have attached a link to the background behind the setting up of the Symposium.
I will need to serch much further to find the area I mentioned previously about the SSD and HDD comparisons.
Alex
A Look At The Inaugural Computer Audiophile Symposium | Computer Audiophile
Obviously, I am greatly upsetting the "bits are bits" brigade, so I will refrain from further posting in this thread.
Attached is a link to one of those reports. Chris Connaker does not permit whole posts to be copied.
http://www.computeraudiophile.com/content/SSD-SATA-interface-recommendation
Last edited:
So it was exactly as I said? We were really being asked to accept that there is a difference between a file ripped to a HD and one ripped to a SSD even though the checksums agreed?
Nobody is upset here, sandyk. I made a mistake is all. I should never have admitted to being sane.
w
What the DAC Chip Sees
What I was suggesting is that it is not the storage media that affects the sound, but what the dac actually sees. For example, a Logitech Squeezebox or Sonos buffers the data. If one uses a reclocker, such as the Pacecar or whatever, that buffers and reclocks the data, what difference would a hard drive make, especially after it is buffered in computer memory and yet again externally before hitting the dac? No doubt that the symposium members heard a difference, but the reason for that difference must be something other than the media storage in the computer. What it was, be it hardware or listener related, I do not know.
What I was suggesting is that it is not the storage media that affects the sound, but what the dac actually sees. For example, a Logitech Squeezebox or Sonos buffers the data. If one uses a reclocker, such as the Pacecar or whatever, that buffers and reclocks the data, what difference would a hard drive make, especially after it is buffered in computer memory and yet again externally before hitting the dac? No doubt that the symposium members heard a difference, but the reason for that difference must be something other than the media storage in the computer. What it was, be it hardware or listener related, I do not know.
The demonstrations at that Symposium were carefully arranged DBTs, that are so beloved of the engineering community.That would normally be sufficient to satisfy MOST "sane" people.
As I said previously, I will no longer participate in this thread. The thread has already gone way off the original topic.
Well, to try and get things back on track...
The reason for high end CD players in the first place was "better sound," addressed through a lot of attempts to mitigate the problems that can be encountered playing back CDs:
- Special transport mechanisms, or "hot-rodded" OEMs to get the digital data
- Sometimes special algorithms to process the data (oversampling etc, could be considered part of the DAC)
- Very high quality DACs
- Very high quality analog circuits
- And very robust and very isolated power supplies to power all that.
Back in the day, you simply couldn't do all that in a computer.
The next thing that came along was receivers with DACs inside, so then folks started asking "why can't I use a cheap player, bits are bits?!" Well, once they're recovered from the disc PROPERLY that's true, but that is not an easy task in streaming real-time. A cheap setup could conceivably run into more interpolation and therefore change the sound; there's another thread somewhere about that.
Then the media server concept came along, and now folks like the OP ask "why not just store the bits?" Why not indeed. If you can recover the data from the disc ROM-style, i.e. nonrealtime, you can be sure to get it 100% correct unless the disc is damaged. Then you can play it back reclocked, for possibly better sound than a realtime CD player ever could.
BUT
that still depends on the quality of the other stuff!
So unless the computer-based setup can achieve all that list of stuff the high end CD player has, it may not be better. If you have an optic output from the server running to a super high quality outboard DAC (or excellent DAC in a receiver), that could be superior to CD players. If you're just running iTunes in your cheapo computer, it probably sounds like dog vomit. Certainly the headphone output of my beloved iMac does.
P.S. I'm not a sound card expert, but I question: are their ANALOG stages really so wonderful? I know you can get 5.1 192/24 blah blah, but the focus always seems on the digital part. I question if something that size and for that market really addresses the post-digital side of things.
???
The reason for high end CD players in the first place was "better sound," addressed through a lot of attempts to mitigate the problems that can be encountered playing back CDs:
- Special transport mechanisms, or "hot-rodded" OEMs to get the digital data
- Sometimes special algorithms to process the data (oversampling etc, could be considered part of the DAC)
- Very high quality DACs
- Very high quality analog circuits
- And very robust and very isolated power supplies to power all that.
Back in the day, you simply couldn't do all that in a computer.
The next thing that came along was receivers with DACs inside, so then folks started asking "why can't I use a cheap player, bits are bits?!" Well, once they're recovered from the disc PROPERLY that's true, but that is not an easy task in streaming real-time. A cheap setup could conceivably run into more interpolation and therefore change the sound; there's another thread somewhere about that.
Then the media server concept came along, and now folks like the OP ask "why not just store the bits?" Why not indeed. If you can recover the data from the disc ROM-style, i.e. nonrealtime, you can be sure to get it 100% correct unless the disc is damaged. Then you can play it back reclocked, for possibly better sound than a realtime CD player ever could.
BUT
that still depends on the quality of the other stuff!
So unless the computer-based setup can achieve all that list of stuff the high end CD player has, it may not be better. If you have an optic output from the server running to a super high quality outboard DAC (or excellent DAC in a receiver), that could be superior to CD players. If you're just running iTunes in your cheapo computer, it probably sounds like dog vomit. Certainly the headphone output of my beloved iMac does.
P.S. I'm not a sound card expert, but I question: are their ANALOG stages really so wonderful? I know you can get 5.1 192/24 blah blah, but the focus always seems on the digital part. I question if something that size and for that market really addresses the post-digital side of things.
???
head_unit:
You're absolutely right about the sound card thing... I wouldn't touch the analog output of any PC with a 10ft pole. Cheap single rail op-amps, sloppy analog stages, electrolytic cap coupled outputs and all the PSU issues that come along with using a second rate OEM flyback supply that was never designed to power sensitive analog stages.
The only way to go is to run digital out of the PC and do the D/A conversion properly with an actual DAC outside the PC.
You're absolutely right about the sound card thing... I wouldn't touch the analog output of any PC with a 10ft pole. Cheap single rail op-amps, sloppy analog stages, electrolytic cap coupled outputs and all the PSU issues that come along with using a second rate OEM flyback supply that was never designed to power sensitive analog stages.
The only way to go is to run digital out of the PC and do the D/A conversion properly with an actual DAC outside the PC.
A question for you folks. I am using an M-Audio 2496 sound card in my modest home music PC server. Are the DACs in the M-Audio 2496 as good as most outboard DACs? Would I really gain a lot in sound quality if I got an outboard DAC?
Thanks for your help......
^^ Depending on which DAC you got, yes you could. I use a Buffalo DAC connected to the digital output of my E-mu 1212m, and the step up is pretty noticeable. In spite of the hobbyist implementation of the DAC, the sound is still capable of keeping up with a $2K CD player.
I find HDDs actually do affect the sound quality adversely, because of seek noise and the rumble they make while spinning. I set the player to buffer the files to memory and play them back. Buffer is set at 600MB, enough to buffer even very long Indian classical performances. even with extremely quiet Western Digital Green drives and soft mounting in the Antec NSK2480, the difference is noticeable in times of the day that are a little quieter - even though I sit about 8 feet from the PC.
I rip everything to wav, and use Foobar for playback. I don't think I'll ever go back to a standalone player. A good DAC and transport is all I need, the e-Mu is nice but apparently has a latency issue so is suboptimal. Plus it can't switch sampling rates on the fly, making it bothersome to use. The other card I have is a m-Audio Delta 66, which does not allow software to interface with windows volume control and is thus inconvenient to operate on the fly. And I need a really good DAC, good as the Buffalo is it's a bit tiring to use. The convenience of having over 2000 CDs at your fingertips is worth the hassle though
I find HDDs actually do affect the sound quality adversely, because of seek noise and the rumble they make while spinning. I set the player to buffer the files to memory and play them back. Buffer is set at 600MB, enough to buffer even very long Indian classical performances. even with extremely quiet Western Digital Green drives and soft mounting in the Antec NSK2480, the difference is noticeable in times of the day that are a little quieter - even though I sit about 8 feet from the PC.
I rip everything to wav, and use Foobar for playback. I don't think I'll ever go back to a standalone player. A good DAC and transport is all I need, the e-Mu is nice but apparently has a latency issue so is suboptimal. Plus it can't switch sampling rates on the fly, making it bothersome to use. The other card I have is a m-Audio Delta 66, which does not allow software to interface with windows volume control and is thus inconvenient to operate on the fly. And I need a really good DAC, good as the Buffalo is it's a bit tiring to use. The convenience of having over 2000 CDs at your fingertips is worth the hassle though
My rig......Reclocked CD-Pro using a reclocked I2S interface to 4 x TDA1541A-S1 chips. NOS.
The TDAs are a linear interpolation ( each dac delayed ). Tube output......soon to be a passive I/R.
All caps selected and bypassed.... and all resistors selected metal film or foil.....
The reclocking of the I2S and the CP-Pro eliminates tons of jitter....that can clearly be heard.
The linear interpolation is simply wow.....
Power supplies.... four transforms and more then enough capacitance for 250 wpc amp.....
Its all in the details of the implementation.... I plan to do the same for a high bit rate music server....
jk
The TDAs are a linear interpolation ( each dac delayed ). Tube output......soon to be a passive I/R.
All caps selected and bypassed.... and all resistors selected metal film or foil.....
The reclocking of the I2S and the CP-Pro eliminates tons of jitter....that can clearly be heard.
The linear interpolation is simply wow.....
Power supplies.... four transforms and more then enough capacitance for 250 wpc amp.....
Its all in the details of the implementation.... I plan to do the same for a high bit rate music server....
jk
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.