Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Yulong Sabre D18 thread: reviews, impressions, discussion (full review added 2/5)
New Posts  All Forums:Forum Nav:

Yulong Sabre D18 thread: reviews, impressions, discussion (full review added 2/5) - Page 4

post #46 of 999
Thread Starter 
Quote:
Originally Posted by slackman View Post

Well.. I just figured out the Sabre ES9018 performs actual upsampling when it does its ASRC..

Sorry for shortly giving the expectation of big performance increase when using upsampling on the computer, when there is actually little to be had (the only difference will be the slightly higher quality of computer upsampling).

I should've waited with posting such things till I actually had my own D18 and tested it mysef.



It is possible to disable some of the internal processes from the ES9018. I can't recall who, but somebody does that with their designs. A really high end brand if I recall correctly. I don't know exactly how Yulong has chosen to implement the chip in the D18. I'll ask him (through Rachel at Grant Fidelity) when I get a chance.

 

BTW - thanks for the info over your last few posts. I'm not sure I completely agree with you about ASRC chips in DACs being completely bad though. Obviously there are trade-offs involved, but a good design (See Violectric V800) can really sound excellent. Without ASRC, you better have a super high-end source, because jitter is going to be a big issue. Async USB is a good option but not perfect, and not everyone wants to have their computer as source anyway.

post #47 of 999

 

Quote:
Originally Posted by Philharmoniker View Post

So my D18 just arrived. Interestingly, as I took it out of the box, the I heard a rattling sound, as if a piece of screw has come loose in the machine. Has any of you guys heard it? I will see if it turns on at all before opening it up...


 

Check the fuse holder, there is likely a spare fuse in there that makes noise when you shake the unit, just tested the Yulong and light sound like a small crew floating around. Same thing on our TubeDAC-11 which is built by Yulong as well. Nice to have a spare fuse eh!

 

Just now reading this thread, didn't even know it was happening, thanks for the heads up John.

 

Like the 'yu live for long time', as good as Rachel's Chi-Fi. how about "yu long" for great audio :)

 

Cheers,

Ian


Edited by Grant Fidelity - 12/21/11 at 5:16pm
post #48 of 999

I opened up the case with Sonic77's help and found a loose screw in there. After taking some pics of the board and personally sending it Xiaolong over IM, we agreed that the screw was just put in there carelessly during assembly and did not fall off from the board. He was very helpful with it. 

I did notice the manual mentioning a spare fuse. I hope I will never get to use it...biggrin.gif
 

Quote:
Originally Posted by Grant Fidelity View Post

 


 

Check the fuse holder, there is likely a spare fuse in there that makes noise when you shake the unit, just tested the Yulong and light sound like a small crew floating around. Same thing on our TubeDAC-11 which is built by Yulong as well. Nice to have a spare fuse eh!

 

Just now reading this thread, didn't even know it was happening, thanks for the heads up John.

 

Like the 'yu live for long time', as good as Rachel's Chi-Fi. how about "yu long" for great audio :)

 

Cheers,

Ian



 

post #49 of 999
Quote:
Originally Posted by slackman View Post

So far it appears to me nobody is using the D18 to anywhere near its full potential.

The D18 does not have an upsampler built in (and I'm not 100% sure but I think also no PLL jitter attenuation etc, other than what the Sabre DAC itself does)

The proper way to use the D18 is to upsample the music on the computer to 192Khz 24bit (or 32bit) and use a very high quality asynchronous USB to AES/EBU or spdif converter with low jitter like the Audiophileo2, Stello U3 or M2tech EVO.

 

The computer can upsample better than any included upsampler in any DAC, also in real time (SOX for foobar or cPlay for instance). Set it to linear phase with 99% bandwidth for correct upsampling (not nonsense minimal phase which has nothing to do with reconstructing the original wave according to the sampling theory upon which digital audio is built).

This upsampling will give you better than 175dB THD+noise, this is inaudible.

 

The difference between doing the above correctly and simply streaming 44.1 Khz audio through a mediocre USB-spdif converter (or directly from a cd transport) will be huge.

 

Btw, to make it lighter on the computer one can also upsample 44.1Khz audio to 176.4Khz instead of 192.

 

edit: here is a link to the SOX resampler for foobar2000: http://www.hydrogenaudio.org/forums/index.php?showtopic=67373 simply unpack the zip to your foobar2000 components folder and then in foobar load and configure the resampler under preferences - dsp manager.


Thanks for the link to the resampler - it does provide a very good detail retreival and performs much better than standard resamplers which IME can do terrible things to soundstage and imaging.

 

I do still find however that JPlay gives better soundstage size, and if I could get hibernate to work would probably be very close on detail also.  Would be nice if we could have both though, but IMO soundstage wins out.

 

Oh and don't forget the Jkenny HiFace mk3 bigsmile_face.gif.

 

post #50 of 999


Hi Ian,

Just wanted to say thank you for the easy transaction getting the Yulong D18 from Hong Kong, ordered last Monday and received it this Monday. I'm still enjoying the Tube Phone Preamp P-307 with Psvane Tubes, I bought from your company. The Love Yu Long Time, was in good humor, my wife is Vietnamese, so if anybody should be offended, it should be me.bigsmile_face.gif I hope to hear any additional insight you may have on this unit, I think you have a winner on your hands!
 

Quote:
Originally Posted by Grant Fidelity View Post

 


 

Check the fuse holder, there is likely a spare fuse in there that makes noise when you shake the unit, just tested the Yulong and light sound like a small crew floating around. Same thing on our TubeDAC-11 which is built by Yulong as well. Nice to have a spare fuse eh!

 

Just now reading this thread, didn't even know it was happening, thanks for the heads up John.

 

Like the 'yu live for long time', as good as Rachel's Chi-Fi. how about "yu long" for great audio :)

 

Cheers,

Ian



 

post #51 of 999
Quote:
Originally Posted by drez View Post


Thanks for the link to the resampler - it does provide a very good detail retreival and performs much better than standard resamplers which IME can do terrible things to soundstage and imaging.

 

I do still find however that JPlay gives better soundstage size, and if I could get hibernate to work would probably be very close on detail also.  Would be nice if we could have both though, but IMO soundstage wins out.

 

Oh and don't forget the Jkenny HiFace mk3 bigsmile_face.gif.

 


 

Yes you can have the best of both worlds and for free :)

It's called cMP^2

http://www.cicsmemoryplayer.com

 

The only inconvenience is that you have to have (or make) cuesheets for your albums (otherwise you can only select one song at a time)

 

post #52 of 999
Quote:
Originally Posted by project86 View Post



It is possible to disable some of the internal processes from the ES9018. I can't recall who, but somebody does that with their designs. A really high end brand if I recall correctly. I don't know exactly how Yulong has chosen to implement the chip in the D18. I'll ask him (through Rachel at Grant Fidelity) when I get a chance.

 

BTW - thanks for the info over your last few posts. I'm not sure I completely agree with you about ASRC chips in DACs being completely bad though. Obviously there are trade-offs involved, but a good design (See Violectric V800) can really sound excellent. Without ASRC, you better have a super high-end source, because jitter is going to be a big issue. Async USB is a good option but not perfect, and not everyone wants to have their computer as source anyway.


 

Thanks for asking Yulong about the DAC implementation! But if you're asking it for me personally you don't have to do it :)

It would make an interesting read but while I have some insight into digital audio I seriously lack knowledge about DAC chips and it wouldn't really influence my buying decision which I've pretty much decided on already anyhow, I'm going to take the plunge with the D18. Just hope it does soundstaging / depth well (one of my 3 big soundquality complaints of the Lavry DA10 I used up till now)

 

About the ASRC being bad, I didn't mean that in a strong sense.

They're a little bit bad in theory :) But correct upsampling in general is very very good (also when done by ASRC) so overall they're good :)

I do believe the V800 is a good DAC, but it reminds me a bit too much of the Lavry DA10 in how it functions (without ever having heard the V800) and there's no way I'm willing to risk the same errors as the DA10 popping up again.. I know them too well by now and my new speakers lay them too bare for what they are and they don't match.

It's probably mostly an unreasonable fear though.. Maybe I'll get a chance to listen to the V800 one day as it's made close by in Germany.

 

 

post #53 of 999
Quote:
Originally Posted by slackman View Post


 

Yes you can have the best of both worlds and for free :)

It's called cMP^2

http://www.cicsmemoryplayer.com

 

The only inconvenience is that you have to have (or make) cuesheets for your albums (otherwise you can only select one song at a time)

 


BUT cMP^2 pretty much requires a dedicated music server to sound it's best (or any better than foobar etc) and requires ASIO to be set up.  Its a bit of a pain to use tbh also with cue sheets etc and I never got it to work.

 

On my upcoming music server build I may try a cMP^2 install as well as realtime Linux but IMO either of those takes a considerable amount of IT knowledge and fiddling to get the most out of.

 

I can always live in hope that JPlay adds an upsamplertriportsad.gif

 

In an attempt to look on-topic though - this DAC looks tasty and if it is fatigue free with a smooth treble and good soundstage should be awesome.  Price is right too.

 


Edited by drez - 12/21/11 at 6:40pm
post #54 of 999

I was unaware that computers (is it really this generally accepted??) will do a better job of upsampling than any dedicated piece in any DAC.

 

Just clarifying, is this true?  I am naturally skeptical, do you have an article or source?

 

Thanks, however!!

 

EDIT: Quoting post #30


Edited by R-Audiohead - 12/21/11 at 6:46pm
post #55 of 999

Yes, but it's all doable.

For normal use while I'm doing other things at the same time I'm using cPlay, which doesn't take over the whole computer.

I'm still hearing the benefits of reduced jitter compared to foobar, and haven't heard an improvement over cPlay when using full cMP (but it's also impossible to do direct AB switching to compare, as is easy to do with fooar and cPlay)

Btw I see that you have the JKenny transport as your picture?

This is an asynchronous USB transport, if this is what you're using it shouldn't matter at all which player you use as long as it's bit perfect, computer jitter is not an issue anymore then, so I'd go with foobar?

If you're indeed hearing a difference between JPlay and foobar while using the JKenny transport then perhaps this is due to increased physical computer noise while using foobar as it reads from the HD etc while playing while JPlay does not (this seems silly but this can actually make a fairly big difference, our brain automatically filters out these noises but it's far from perfect and in doing so also filters out music)

Only other thing I can think of is that perhaps you're using the computer as volume control as well and haven't set foobar to dithering? (only gives a small difference if you're really in the low end of the volume control for 24bit)

post #56 of 999
Quote:
Originally Posted by R-Audiohead View Post

I was unaware that computers (is it really this generally accepted??) will do a better job of upsampling than any dedicated piece in any DAC.

 

Just clarifying, is this true?  I am naturally skeptical, do you have an article or source?

 

Thanks, however!!

 

EDIT: Quoting post #30



Yes it's really true :)

You can see specs of computer resampling here: http://src.infinitewave.ca/

(for some really impressive resampling see Izotope 64bit SRC steep)

I don't have specs at hand for ASRC chips but I've come across them from time to time and they're considerably lower. A google search may turn these up.

Infact, you can see the specs for one of the best (if not THE best) hardware sample rate converters in the world, the Weiss SFC2, at the link above.

It's a dedicated hardware sample rate converter costing thousands I belief, used in mastering studios to link different digital hardware. Even that monster doesn't do as good as the good software ones. (edit: though it does so well I'm sure I couldn't tell any difference by far no matter how hard I'd try :) but this isn't the case for regular ASRC chips)


Edited by slackman - 12/21/11 at 7:12pm
post #57 of 999

So does jplay make a big enough difference to buy?

 

Never mind my anti virus wouldn't go for this program.


Edited by Sonic77 - 12/21/11 at 7:44pm
post #58 of 999
Quote:
Originally Posted by Sonic77 View Post

So does jplay make a big enough difference to buy?

 

Never mind my anti virus wouldn't go for this program.



I personally haven't tried JPlay, but cPlay should sound the same for jitter sensitive transports and is free.

But JPlay doesn't allow upsampling (like cPlay and Foobar do) and in my personal experience upsampling makes a bigger difference (potentially MUCH bigger difference depending on your DAC) than the improvements of computer player related jitter on a jitter sensitive source (cheap non asynchonous USB to spdif).

So I'd say for best soundquality of these 3 cPlay / cMP comes first, then foobar2000, then jPlay (based on specs/theory, have not used jPlay). And the fun is that the best 2 are free :)

Probably better spend that $99 towards something that really matters.

post #59 of 999
Quote:
Originally Posted by slackman View Post



Yes it's really true :)

You can see specs of computer resampling here: http://src.infinitewave.ca/

(for some really impressive resampling see Izotope 64bit SRC steep)

I don't have specs at hand for ASRC chips but I've come across them from time to time and they're considerably lower. A google search may turn these up.

Infact, you can see the specs for one of the best (if not THE best) hardware sample rate converters in the world, the Weiss SFC2, at the link above.

It's a dedicated hardware sample rate converter costing thousands I belief, used in mastering studios to link different digital hardware. Even that monster doesn't do as good as the good software ones. (edit: though it does so well I'm sure I couldn't tell any difference by far no matter how hard I'd try :) but this isn't the case for regular ASRC chips)


Not to move too far off topic...

 

I am using a D100, which utilizes a ASRC unit and I believe converts everything to 110kHz sampling rate anyways.  Are you suggesting upsampling via the foobar component you recommended to say 176.4 will (or can theoretically) generate improvements at the end of it all?  Would you recommend resampling to a different frequency? (Like 88.2?)

 

Thanks for the advice in advance

 

post #60 of 999
Quote:
Originally Posted by slackman View Post

Yes, but it's all doable.

For normal use while I'm doing other things at the same time I'm using cPlay, which doesn't take over the whole computer.

I'm still hearing the benefits of reduced jitter compared to foobar, and haven't heard an improvement over cPlay when using full cMP (but it's also impossible to do direct AB switching to compare, as is easy to do with fooar and cPlay)

Btw I see that you have the JKenny transport as your picture?

This is an asynchronous USB transport, if this is what you're using it shouldn't matter at all which player you use as long as it's bit perfect, computer jitter is not an issue anymore then, so I'd go with foobar?

If you're indeed hearing a difference between JPlay and foobar while using the JKenny transport then perhaps this is due to increased physical computer noise while using foobar as it reads from the HD etc while playing while JPlay does not (this seems silly but this can actually make a fairly big difference, our brain automatically filters out these noises but it's far from perfect and in doing so also filters out music)

Only other thing I can think of is that perhaps you're using the computer as volume control as well and haven't set foobar to dithering? (only gives a small difference if you're really in the low end of the volume control for 24bit)


Btw, I forgot to add how to make cuesheets of music you already have on your HD.

You make a playlist you want in for instance foobar2000, you can load songs from any folder you want (as long as it's either flac or wav), and then save this playlist as a .m3u playlist.

Then you use m3u2cue (http://christophe.peignot.perso.sfr.fr/dev/#m2c) to convert this playlist to a cuesheet (click "use absolute path" when converting to make it work).

Takes a bit of time to convert your playlists, but once done it works and cPlay becomes pleasant to use.

 

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Yulong Sabre D18 thread: reviews, impressions, discussion (full review added 2/5)