|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
66.42.1.213
In Reply to: Re: Thank you, but..... posted by Scrith on September 10, 2006 at 12:22:24:
http://homepage2.nifty.com/sunacchi/download/foo_playram.zip
Follow Ups:
Well, I couldn't get the plug-in to work (even with the old version of Foobar...luckily I keep that installed somewhere for tests like this).But I did install a RAM Disk program and manually converted one of my FLAC files that I often use for testing to WAV and copied it to the RAM Disk. I then started a blind test using my wife as the subject, and then a non-blind test using myself.
The FLAC file playing from the hard drive sounded EXACTLY the same as the WAV file playing from the RAM Disk to both of us.
I attribute this to the fact that I'm using a large buffer (64MB) in Foobar2000, which basically means it is playing back from RAM already anyway. Or that a hard drive works just fine, even without buffering.
I have compared a number of different products (DACs, amps, speakers, headphones, interconnect/speaker/power/digital/USB cables, USB to S/PDIF converters, power conditioners, plug-ins, file formats, Foobar2000 settings, sound driver settings, etc.) and have often heard differences. Not in this comparison. So I have concluded that this idea of using a RAM Disk in Foobar2000 (and WAV files vs. FLAC files, for that matter) is a complete waste of time. For me at least. I am equally certain that there will be people out there who insist a RAM Disk makes a difference, no matter how many other people say it doesn't. So please enjoy it if you really think it helps. :)
Test notes: The test was made using music ripped in Secure mode using EAC. The computer has been given a silencing treatment and the hard drives are regularly optimized to make sure all music files are unfragmented. The DAC is a Lavry DA10 fed by an M-Audio Audiophile USB using a Monster pro USB cable. The Lavry DA10 feeds a PS Audio GCC-250 amplifier using very high-end Cardas balanced cables with XLR connectors. The PS Audio GCC-250 amplifier has been upgraded using the new Critical Link fuses and is powered by a Black Sands Violet Z1 power cable (as is the Lavry DA10...the computer uses a PS Audio xStream Statement power cable). The speakers are Totem Acoustic Model 1 Signatures connected via a Cardas Golden Reference bi-wire 2.5m speaker cable. I also tested using Sony Qualia 010 headphones (with balanced Moon Audio Black Dragon cable) connected to a custom-built SFT Audio Dynamight aplifier (powered by PS Audio Plus power cables). I am confident that this headphone setup is as revealing of audio problems as any other used on this planet. :)
BUT changing the buffer size in Foobar affects the sound! Reduce it, and bass improves; increase it, and there is a mellowing effect.
Agree. Keep always the buffer size in Foobar2000 & ASIO as small as possible. This will lower the latency time in music playing. Also set "Direct Monitor Mode" in ASIO and "Real time priority" would help alot.
Reducing the buffer size in Foobar2000 creates a dependency on the hard drive to keep up with the datarate on a regular basis. This can't be a good thing. Unless you like the way this sounds (possibly someone who likes the sound of jitter?).The latency is definitely a drawback (your Foobar2000 Spectrum Analyzer will be slightly ahead of what you are hearing, and won't be as smooth). I don't really watch this very often though...I'd rather be certain that the data is there is memory, ready to be send to the sound device (and give my hard drive some rest, or time to work on getting data for other programs).
In order to prevent dependency on the hard drive. Put music file in RAM drive is a good direction. Also data rate of RAM drive much higher than hard drive. This will decrease the Foobar2000 latency induced by hard drive mechanical seek & rotation time.
But this plug-in is doing the same thing as Foobar's buffer...it's taking data off the hard drive in one large piece and storing it in RAM so that playback can continue without a dependency on the hard drive.The two should be functionally identical, though I'd be surprised if the Ramdisk plug-in's method was as simple as Foobar's buffer.
But how do we know what is right in Computer Audio when no ones seems able to pin down what exactly happens when one sets this and that. In Foobar, there are so many settings that it is just trial and error. No one seems to have made any measurements on jitter that are meaningful.
This post is made possible by the generous support of people like you and our sponsors: