hi all..recently i made a USB DAC based on CS8414 and TDA1545A*3 which use X-Fi Surround 5.1 Pro as the USB to SPDIF
i actually liked how the TDA1545A sounds.but i started to look at ASIO and DSD/DXD files which my X-Fi and 1545 couldnt play.foobar2000 always said "this hardware didnt support 44.1/88.2/192/384kHz output" everytime i want to play DSD/DXD files. (depends on what i play)
i heard XMOS and PCM5102 could do it.but how they sounds?.are they good?
and i also use ECC84 SPDIF buffer currently.if i upgrade, is it still useable or just throw it to my bin?
last, how much XMOS and PCM5102 costs?..
or if you have another suggestion, i appreciate it! 🙂
thanks in advance 🙂
-add-
can i use tube output for 384kHz output?.i wonder is tube can reproduce all spectrum or limited? (i'm a noob 😛 )
----
i still impressed how ASIO can makes my cheapo DAC sounds better than before :shock:
i should know this since the start!.i thought ASIO didnt supported on windows 8.1!
i actually liked how the TDA1545A sounds.but i started to look at ASIO and DSD/DXD files which my X-Fi and 1545 couldnt play.foobar2000 always said "this hardware didnt support 44.1/88.2/192/384kHz output" everytime i want to play DSD/DXD files. (depends on what i play)
i heard XMOS and PCM5102 could do it.but how they sounds?.are they good?
and i also use ECC84 SPDIF buffer currently.if i upgrade, is it still useable or just throw it to my bin?
last, how much XMOS and PCM5102 costs?..
or if you have another suggestion, i appreciate it! 🙂
thanks in advance 🙂
-add-
can i use tube output for 384kHz output?.i wonder is tube can reproduce all spectrum or limited? (i'm a noob 😛 )
----
i still impressed how ASIO can makes my cheapo DAC sounds better than before :shock:
i should know this since the start!.i thought ASIO didnt supported on windows 8.1!
Last edited:
You *don't* want to reproduce the sample rate frequency - you have to low pass filter away anything above half the sample rate.
You can start filtering from 20 kHz, the only advantage of the high sample rate is that you can use a softer, gentler filter and still make sure you are removing enough of the HF.
Thus the output stage doesn't need to go above 20 KHz (from a FR point of view - phase behaviour etc. is another matter).
yes.i know samplerate was 2x higher than maximum frequency range being produced like 44.1 was 22.05kHz max 🙂
and i forgot that 384k means 192k out which human can't hear 😀
how's the filtering did?.is it just using a coupling caps just like usual?
i still don't know much in such filtering things
and i forgot that 384k means 192k out which human can't hear 😀
how's the filtering did?.is it just using a coupling caps just like usual?
i still don't know much in such filtering things
Many designers consider the filtering to be the "secret sauce" that makes their designs unique. It is not a completely trivial subject 🙂
TDA1545A can run at 8X OS so if I were you I'd use that in preference to PCM5102. However I've found it doesn't sound as good when run faster (more glitching is my best guess) so I suggest you don't run more than 2X OS. If you have many TDA1545 (say 19 but the number is somewhat flexible, they are cheap DACs) you could implement my digital filterless oversampling concept called 'LAID' 😀
2x means 88.2kHz?..i runs it at 96k currently.it's more than 2x but not more than 4x 😕
and how is glitches sounds like?..running on 96k sounds more sparkling but i liked how the soundstage being wider 😕
and how is glitches sounds like?..running on 96k sounds more sparkling but i liked how the soundstage being wider 😕
now i becomes more curious
can you give an example? 😀
Well, the simplest one is really a simple RC filter. But this is a huge pandora's box... A very complex subject.
If you like deep soundstage, I've found that comes from passive filtering after passive I/V conversion (a resistor). Glitches sound like less 'refinement', greyer tonal colours (on piano for example), reduced 'delicacy' of HF sounds like the plucking of a classical guitar.
Yes 2X OS means 88.2kHz. To get 96k do you have an ASRC?
'Sparkling' to me sounds like what I call 'false detail' - if you love that sound then run the DAC as fast as you possibly can and use opamp I/V (LM4562 should deliver) 😀
Yes 2X OS means 88.2kHz. To get 96k do you have an ASRC?
'Sparkling' to me sounds like what I call 'false detail' - if you love that sound then run the DAC as fast as you possibly can and use opamp I/V (LM4562 should deliver) 😀
hmm..so this RC runs to ground.am i right? 😀
in piano tones, is glitch sounds like unnatural?.as guitar the plucking sound wasnt detailed or ? 😕
what is ASRC?
oops...i listened to 128k MP3 when i say "sparkling"
.when i tried to listen a 24/192k classical it sounds good 😀There are many ways to implement a RC filter - but a capacitor to ground is indeed the most common.
Asynchronous sample rate converter.
aah..i see..
i'll take a look on google for LPF filters 🙂
no.i don't have any async sample rate converter (hardware based?)
i used ASIO and set my foobar resampler at 96k (my x-fi didnt support 88.2/192/384k.even 44.1 didnt supported! 😡 )
i'll take a look on google for LPF filters 🙂
no.i don't have any async sample rate converter (hardware based?)
i used ASIO and set my foobar resampler at 96k (my x-fi didnt support 88.2/192/384k.even 44.1 didnt supported! 😡 )
The problem with words like 'detailed' is different people attach different meanings to it. To me 'detail' is something to avoid, its distortion, unnaturally accentuating some HF. No, I tend to only notice glitch when its attenuated, it doesn't sound noticeably 'unnatural'. Just when its removed the SQ is obviously better but its not perceived as 'bad' before.
ASRC means asynchronous sample rate converter.
128k mp3 is hardly reference grade material 😉
<edit> If you can get hold of TDK's 7mm range of SMT inductors I can suggest a fairly straightforward low pass filter for you.
ASRC means asynchronous sample rate converter.
128k mp3 is hardly reference grade material 😉
<edit> If you can get hold of TDK's 7mm range of SMT inductors I can suggest a fairly straightforward low pass filter for you.
so.."details" is way too wide to determined 😀
when i moved to ASIO, it sounds much better imo.every instruments sounds clearer and no fatigue at all.since my X-Fi didnt supported 44.1 or 88.2, i set my sampler at 48k at the first.when i move to 96k it sounds more spacious but didn't compress any details
as 128k, that's old me when i have no big storage as now 😛 .i prefer FLAC/WAV nowadays.but my TOTO collections was ripped to 128k and the FLAC was gone w/ my previous HDD which died by badsectors 🙁
when i moved to ASIO, it sounds much better imo.every instruments sounds clearer and no fatigue at all.since my X-Fi didnt supported 44.1 or 88.2, i set my sampler at 48k at the first.when i move to 96k it sounds more spacious but didn't compress any details
as 128k, that's old me when i have no big storage as now 😛 .i prefer FLAC/WAV nowadays.but my TOTO collections was ripped to 128k and the FLAC was gone w/ my previous HDD which died by badsectors 🙁
- Status
- Not open for further replies.