CD and PC input To DirectStream


#1

Hello All,

Old Geezer here but a noob to forum discussion…Please bear with me.

Just acquired a DirectStream and connected it to a 2006 Rega Jupiter CD Player via coax. Certain Full orchestra and Heavy Rock songs sound like they are skipping or in fast forward. Ripped those CDs to my PC (Windows XP) using JRiver trial which encoded them to .ape files. Downloaded latest USB driver from PSA website and installed, connected to DS using USB and played those songs using JRiver Media 21. Same results until…I set the “Play From Memory” parameter on. Must the input stream to the DS meet a certain transfer threshold or do I have something set incorrectly? Will I have to rip all my CDs or just buy the upcoming PSA CD/SACD transport?!? (noise you hear is wife is expressing opinion on that).

Sincere apologies if this has already been covered, please point me to the thread, I could not find the answer.

PS After just a few hours the DS sounds glorious…thank you Paul, Ted and Arnie.


#2

It’s probably that you have the buffering in JRiver too small and you can’t read the data reliably from your disc in time to send it to the DAC. Despite JRiver’s suggestions for a shorter time I set the prebuffering option in the audio options to, say 20 seconds. And I also use the maximum buffer setting the audio driver supports - if you are using ASIO in the ASIO options there a Buffering setting that JRiver recommends to be 50ms and I change it to 500ms - if you are using WASAPI I change from the recommended 100ms to 500ms - if you are using Kernel Streaming I don’t know what the default is but I move the slider to about the middle which is 0.5 seconds - and the same for Direct Sound.


#3

Thank you Sir. Problem resolved. Much appreciated!


#4

Excellent news!

Please stick around and join in. 103_gif


#5

Thank you for the hospitality, Elk. These forums are such a great resource…so many smart people.

thought you might be interested in my follow up activities to Ted’s explanation. Per his advice, I could adjust the parameters in JRiver and cause failure or success. He seemed to imply the data stream from the source was being interrupted and needed more buffering, so I surmised error correction may be the culprit. I examined the offending CDs and sure enough there were scratches that appeared to be radiating from the center probably as the result of mishandling while loading into a input slot (car system). The other Cd was new but had an air bubble or blob near the edge and the problem appeared on the first song, then stopped.

Now my OCD flared and I called a couple of friends with better CD players than me. I visited the first and loaded the problem CDs into his universal player with an “O” on the front…same results. The next visit was a player with an “R” on the front and nuvistor thingys…again the same problem. My final stop was my local audio shop and a beast of a player with an “E” on the front and a price tag over $10K feeding a tube preamp and tube monoblocks. OMG, it sounded beautiful until that same spot in track 1 caused him to leap to the aid of the system, shut it down and explain valve overload.

So…Ted was right on track and this little soirée showed that if a transport could read the disk then play from memory allowing time for error correct/re-reads, that would be the answer. And listening to the different systems, if we could build components with the best of both tubes and solid state, that would be a dream. Ain’t it great to be ahead of the curve? The forthcoming preamp and transport should continue the trend.

Now that I have used up my allotted comments for this year, thanks for indulging me.


#6

Thanks for the update and information. This is both interesting and may well help the next guy in line.

I have had great success polishing CDs with minute and even heavy scratches and have got them to play fine. The underside is just plastic. The data is on the underside of the silver layer which is actually the top of the CD. Thus, polishing is not putting the data at risk.

And your allotment has not expired! We would enjoy continuing to hear from you. :slight_smile:


#7
Stevie B said I examined the offending CDs and sure enough there were scratches that appeared to be radiating from the center probably as the result of mishandling while loading into a input slot (car system). The other Cd was new but had an air bubble or blob near the edge and the problem appeared on the first song, then stopped.
The video rental stores in my area have disc polishing machines that can buff out scratches and make CDs/DVDs look like new. They charge between 2-3 dollars per disc.

CDs are read starting from the center and going out to the outside edge, so a bubble or blob near the outside edge should not effect track 1.


#8

test


#9
bstanwick said

CDs are read starting from the center and going out to the outside edge, so a bubble or blob near the outside edge should not effect track 1.


Excellent point.

Trivia: The TOC (table of contents) of a Redbook CD occupies the innermost 0.4 millimeters of the disc. I still find this sounds like magic.


#10

Elk and bstanwick, excellent advice, I will investigate the polishing options. And I should have said the inside edge near the hole. I was not aware of the TOC and guess I would have gotten an error if that could not be read.

In case of a read error, does the transport try to reconstruct the data, simply re-read X times or skip to the next readable sector, and while it is doing this, the DAC just waits? Is that a dropout?

Steve


#11

It is very cool how it works.

The player accesses redundant information on the CD to reconstruct the content lost when there is a read error - and there are always read errors.

When the data loss is too large to be handled by the above redundancy error correction scheme, the CD player employs a concealment circuit which interpolates the lost data from the surrounding data.

Short hand version: magic is invoked. :slight_smile: