Memory playback

There is a strong believe in the computer audiophile community that any electrical activity going on during play back has a negative impact on sound quality.

No access to the HD during playback by loading the song in memory first is one of the tweaks. This is called memory playback.

The advantage to memory play is reduction of activity in the computer box during playback. Normally, data is read off of the disk into memory and then copied to the device, so there are two separate activities going on. With memory playback only copying data from memory to the device is happening. Ideally, the hard drive could be spun down as well, so there wouldn't even been any electrical and acoustic noise while listening to music.

The assumption is that less activity means a cleaner signal out of the computer box to the DAC. Whether this is true or not, and whether or not this affects the output of the DAC will depend on the specific system.

The primary disadvantage to memory playback is the time spent up front loading memory. There is also an extra cost of large amounts of memory, which will be measured in gigabytes when playing long hi-res playlists. For example, I have one double album that would require 5 GB just to hold the WAV file for the complete program.

Tony Lauck

 

A spinning HD and its moving head will generate electrical activity.
You can minimize this by de-fragmenting your files.
This minimizes head movements and sure reduces the acoustical noise!

 

Replacing the HD by a SSD is another option, no motor, no head.
At the moment SSD is way too expensive to be used as a storage device for a substantial music collection. OS on the SSD, music on an external HD is often recommended.

 

A software solution is memory playback.
The media player allocates a buffer in memory and loads the song from the HD into memory first.
This won’t necessary eliminates all access to the HD during playback.
There might be other processes beside the media player accessing the HD.
But of course, a dedicated audio PC is tweaked to death so hardly any other process will be running.
Now Windows has a mind of its own. It might decide to swap this buffer to the page file.
Effectively you have the audio transferred from the HD to the page file on the HD!
Disabling the page file is probably needed too.

Perfect memory playback

No access of the HD during playback so no head movements.

This requires the entire track lo be loaded in memory before playback starts.

Electrical activity can be further reduced by spinning down the HD after loading the audio.


The audio format has to be converted to a format the DAC can understand (raw PCM).
This requires some CPU power.
A further reduction can be obtained by decoding the audio when it is loaded in memory.

Evidence

Are the reported improvements of memory playback the usual highly subjective badly tested typical esoteric audiophile claims?
One thing is sure; these claims are in general not backed up by an ABX test or any measurements.

Home Theater and High Fidelity tested the jitter on the SPDIF out of a blue-ray player in stop mode and running.
Indeed running the system doubles the jitter.
However, 5 or 10 ps are very low values.
The periodic jitter (the spikes) is probably more relevant.
One thing is obvious, the higher electrical activity of the system when running maps into a measurable difference in jitter at the SPDIF out.

It is possible that other components than the optical drive contributes to this raise in jitter.

 

In principle, the less disk activity, the better from the power-supplies point of view and hence the better for any device that is powered from the PC's power supply or is slaved to clocks in the PC. So having the whole track (or several) loaded into memory could help.

 

If you use an external asynchronous audio device that is self powered, there is a minimal effect (if any) from going to memory playback in J-River Media Center or setting the buffer in my other playback software to "ridiculously large".

 

Even without Async Devices, most modern hard disks have several MB of on-board cache. So there are several layers of buffering before the data is send out to the audio device. Modern operating systems also tend to pre-buffer as much data as they can when reading from disk anyway, my machine usually a GB or two for disk caching.

 

Further, what I find is that with my AV optimized (and de-fragmented - this part is crucial) hard drives tend to pre-buffer the data long before the playback software asks for it and many are very low power.

 

Hard drive access now, on my machine, is only blinking lights, no acoustic indication anything happened (e.g. non of the typical "disk head trashing" noise) and I cannot see any impact on the power supply with a fairly sensitive scope. It was not like that a few years ago, you could easily see the HDD modulating the supply during playback.

 

So the need for memory playback is IMNSHO declining rapidly, just as it has basically become ubiquitous.

Thorsten

One thing is for sure, if there is no access to the HD during playback, you won't hear the head rattling!

References
  1. Memory play on Foobar - Computer Audio Asylum
  2. OPPO BDP-83SE (Special Edition) and NuForce Edition Universal Blu-ray Players - Secrets of Home Theater and High Fidelity 25 January 2010