SD card transports

Hi Jimbo,
I have one, purchased from Chevron. Using EAC (Exact Audio Copy) to "burn" CDs (as wav files), I have not listened to a CD played on a conventional transport since. I am hearing much more detail, particularly noticeable in the bass where dynamics and impact are excellent. At just over 100GBP it is a bargain!
Regards,
Jim
 
Thanks jimsmy for your reply. Yes it does seem good value as i read on another site it compared well against a Shigaclone. How do you find what you want on the disk though, as the display seems rather small with basic info?
 
The display is not one of the players strong points but the remote control allows you to quickly navigate through the music on the card. I keep most of my favorite music on my computer (transferred by EAC) so it is a very easy and rapid process to transfer to a SD card. I think trying to build a library of music on SD cards would be obviously impractical. Overall the good points (sound quality) out way the less than ideal design etc.
Regards,
Jim
 
A problem that I can see with this approach is that SD cards have a limited life when they are used for frequent re-write operations. It is necessary to make frequent re-writes if your music collection is large because SD card do not have a big storage capacity.

The same sound benefit could probably be obtained by using a PC as the transport and having player software which transfers each album from the hard disk into a RAM drive when it is selected for playback. The cPlay software can do this and, I believe, also JRiver Media Center. Maybe also there are others, but I do not know them.
 
HvD,
I have not yet experienced a SD card failure (I use these in some of my digital cameras) but I appreciate your comment. I believe solid state memory (as opposed to "normal" hard drives) give a more consistent transfer of data; also computers utilise very noisy (RF) SM power supplies. In my personal experience the QLS player sounds significantly better than using a computer to store/play via a DAC. I am also using a SB Touch which although excellent streaming music from my PC, lacks a little something when compared to the QLS/SD card. However I need to experiment further! I think we can all agree that the old way of reading music directly via an optical drive has had its day, unless someone can convince me otherwise.
 
jimsmy said:
HvD,
I have not yet experienced a SD card failure (I use these in some of my digital cameras) but I appreciate your comment. I believe solid state memory (as opposed to "normal" hard drives) give a more consistent transfer of data; also computers utilise very noisy (RF) SM power supplies. In my personal experience the QLS player sounds significantly better than using a computer to store/play via a DAC. I am also using a SB Touch which although excellent streaming music from my PC, lacks a little something when compared to the QLS/SD card. However I need to experiment further! I think we can all agree that the old way of reading music directly via an optical drive has had its day, unless someone can convince me otherwise.
Click to expand...

Yes, you are correct when you talk about the noisy switch mode power supplies that are used in computers. It is possible, of course, to replace these with linear power supplies, but at the moment it is not so easy to do this unless you have the capability to design and assemble them by yourself. It is a surprise to me that none of the small specialist audio companies seem to have made the transition into offering nice power supplies for computer audio transports.

I made the change to using a computer instead of a CD player almost two years ago and there is no doubt in my mind that it provides greater resolution and listenability than anything I had previously. What is very good is that I was able to sell my CD player on eBay, build my computer with the proceeds, and still have some cash left over! At the moment I have to use a normal switch mode computer power supply because I do not have the skill to make my own linear supply. I did, though, make a small linear supply for the ESI Juli@ sound card that I use. I use only the digital section of this card and connect it to an external DAC so it does not have to do very much, but still I thought that the linear supply brought some benefit. A global linear power supply solution is something that I would certainly like to have.
 
The RF issue with SMPS can be countered or at least lessened by using the optical output of your sound card (if you have one); however optical while not being affected by RF is not good with respect to jitter. If your DAC is jitter "resistant" or buffers the data (cf Chord DAC 64) then optical is the way to go. However there are other considerations in obtaining a bit perfect signal into the DAC. CD transports must read the data in real time and correct on the fly which makes them prone to variation but once a bit perfect (or as near as possible) copy is made on to a hard disc or preferably solid state memory (using enhanced software such as EAC) then you have a much better starting point for your DAC to process.

Regards,
Jim
 
Reading data from a HDD also means you are reading data from 'solid state' memory. When a HDD reads data it dumps data from the media into read cache before issuing to the host.
 
lAmBoY said:
Reading data from a HDD also means you are reading data from 'solid state' memory. When a HDD reads data it dumps data from the media into read cache before issuing to the host.
Click to expand...

Do you mean the new solid state drives as conventional ones do use disks that spin, adding higher jitter?
 
lAmBoY said:
Reading data from a HDD also means you are reading data from 'solid state' memory. When a HDD reads data it dumps data from the media into read cache before issuing to the host.
Click to expand...

But it is just a very small cache and the system must constantly access the HDD for replenishment. Perhaps it is better to put all of the data into RAM so that no HDD access is needed during playback. To my ears it was better when I made the comparison.
 
small, not really. The data rates for audio are very small. Depending on the system cache use you could have a lot of audio data at any one time in the read cache. Probably easy to calculate if you knew the data rates and available read cache. (I dont know the split of say a 16Mb cache for read and write channels)
 
I think what we need is a media that is free from the computer as regarding playback as there are issues with noisy fans and power supplies. Also reading from an optical disk too. Using EAC as the copy software then recorded to the separate playing media. As SD disks are small and solid state they would be ideal but at the moment its early days.
 
I am somewhat reminded that some years ago drives were produced which were marketed as "suitable for AV use" (I forget the exact phrase). They didn't really catch on because the computer marketplace realised that the requirements of AV usage were quite undemanding and pretty much any drive was good enough for purpose (remember that much audio recording these days travels via run-of-the-mill laptops and PCs as it makes its way from microphone to purchaseable format).

I think it hurts the audiophile ego to realise this, but I suspect that on an objective/measurable basis off-the-shelf drives are more than adequate for our purposes. And certainly flash/SSD technology comes with its own drawbacks (as suggested earlier in the thread).

The SD player sounds like a fun gadget to fiddle with - but I'm dubious that it has legs.
 
lAmBoY said:
small, not really. The data rates for audio are very small. Depending on the system cache use you could have a lot of audio data at any one time in the read cache. Probably easy to calculate if you knew the data rates and available read cache. (I dont know the split of say a 16Mb cache for read and write channels)
Click to expand...

Let us assume a really good case where the read channel has the whole of a 32Mb cache for itself. This is still quite small for our purpose. An average pop track might be as little 30Mb (as an uncompressed file), which is ok to fit into the cache, but many jazz and classical tracks could be much, much larger than 30Mb. The system must therefore access the hard drive during playback to refresh the cache.

I think that the argument really is that it is good to reduce all extraneous computer activity to a minimum during playback to help to prevent disturbances which might cause increased jitter. Every unnecessary disk access and every unnecessary processor interrupt is best avoided.

I am aware that there are some who say that jitter does not matter and that already any current system can achieve results which are far below audibility. If this is true, then of course concepts such as memory playback are nothing more than a farce.
 
Dont forget that the cache would act as a hopper feed, the drive would repleneish the cache and keep it topped up. The feed across the interface (ie from read cache to host) would be uninterrupted. I dont see where a HDD would differ from an SD card at all (except for the HDD read cache is a much faster and reliable type of memory than crappy SD cards!). Thus, SD is a redundant idea really.
 
I think what we need is a media that is free from the computer as regarding playback as there are issues with noisy fans and power supplies.
Click to expand...

How much power does amplifier use (versus computer capable of playing audio files)? Building/buying passively cooled computer is not hard.

How does reading from HDD influence audio quality? Everything before DAC is digital and should not suffer from interference. There are some very quiet HDD-s out there. If you can still hear HDD spinning - you can soundproof PC case :)

Actually it's cheaper to have 2 computers for playback than to store everything on solid state discs. One server in another room and one passively cooled player in music room streaming data over network.
 
lAmBoY said:
Dont forget that the cache would act as a hopper feed, the drive would repleneish the cache and keep it topped up. The feed across the interface (ie from read cache to host) would be uninterrupted. I dont see where a HDD would differ from an SD card at all (except for the HDD read cache is a much faster and reliable type of memory than crappy SD cards!). Thus, SD is a redundant idea really.
Click to expand...

The issue is that the HDD is being accessed during playback. What you say is true. The cache acts like a hopper. This of course means that it is being constantly replenished from the HDD. This in turn means that the HDD is being constantly accessed.

The argument put forward by the memory player protagonists is that HDD access during playback, and any other unnecessary computer activity, should be avoided because it can increase jitter levels.

I, myself, have no way of making a proper investigation to see whether jitter levels really are increased and if any increase is above the threshold of audibility. What I can say is that my own audio computer has gone through a series of steps up to the present condition of being a memory player. I cannot claim to have made any careful A/B/X testing along the way, or to have made properly documented measurements, but I do think that one of the most notable improvements came when I began to put complete tracks into a RAM drive for playback instead of taking them from the HDD and its cache.
 
Your reply is most interesting, easy to follow even for an amateur like me....It seems some of the solid state advantages of this SD card player can cross over to computers.

I use a MAC, with a memory player. Since starting to use the media player in question I subjectively assess the improvement to be significant, so am very happy. I only currently have 2 gb of RAM - not an issue yet as I have nothing above redbook - so the music plays back from what is effectively a solid state drive.

In using a solid state hard disk, is the benefit supposed to be the removal of moving parts within the computer (I guess this is a source of galvanic interference), or the fact that SSDs are supposed to be faster & more efficient, therefore mitigating the cost (CPU load etc) of running the actual player?
 

Latest posts

Back
Top