Faust3D said:
So long as the result is ok why worry. There is a PCM63 dac modified in a similar fashion somewhere in the forum that is also said to produce stellar results. One problem though. Repeatability. It is hard, or so it seems, to consistently achieve the same results.
Can someone please, hang, draw and quarter the next person to whine about SPDIF. Within its limitations it does what it is supposed to. If you want something better, do something about it.
analog_sa said:
Not sure about the caps. We did test it without the transformer; DAC sounded harsher, less analog. We also tried it with and without output filters and surprisingly I liked it better without filters. This is the first DAC I am happy with, that why I wanted to see what people think about it. I had a lot of other commercials DACs: Museatex, Audio Note etc and like this one so much more.
🙂
Re: Re: DAC - Opinion needed
Thanks a lot man! I am not really worried about it at all.
I just wanted to see what people have to say about it, since I was totally floored when this DAC was finished and I heard it in my system. I was very skeptical about my friend's claims. He had made a lot of good amps that I like and his NOS DAC was something he wanted me to try. 😀
PS: I lurked on this forum for some time, and was impressed by a lot of great designs I found here.
rfbrw said:
Thanks a lot man! I am not really worried about it at all.
I just wanted to see what people have to say about it, since I was totally floored when this DAC was finished and I heard it in my system. I was very skeptical about my friend's claims. He had made a lot of good amps that I like and his NOS DAC was something he wanted me to try. 😀
PS: I lurked on this forum for some time, and was impressed by a lot of great designs I found here.
Faust3D said:
Use Elso's I2S Direct or build the DAC into the CD player.
http://www.diyaudio.com/forums/showthread.php?postid=882816#post882816
(nothing new, already used by Audio Alchemy and North Star)
Re: Re: DAC - Opinion needed
Sorry blowing your bubble rfbrw. SPDIF is poor when implemented as in the CS8412 datasheet and those transformers in the Philips CD80 and the like ......🙁
With a Wildmonkeysects loop filter and low noise supply for that part of the DIR and a few other tricks things improve but when comparing with I2S Direct I subjectively miss bass slam with SPDIF!
rfbrw said:
Sorry blowing your bubble rfbrw. SPDIF is poor when implemented as in the CS8412 datasheet and those transformers in the Philips CD80 and the like ......🙁
With a Wildmonkeysects loop filter and low noise supply for that part of the DIR and a few other tricks things improve but when comparing with I2S Direct I subjectively miss bass slam with SPDIF!
Pretty polly . Pieces of eight . SPDIF is bad
Save the parrot phrases for someone else. I do not use the CS8412.
I attach little value to the subjective opinions of others.
QSerraTico_Tico said:
Save the parrot phrases for someone else. I do not use the CS8412.
I attach little value to the subjective opinions of others.
Re: Pretty polly . Pieces of eight . SPDIF is bad
CS8414 is even worse!
So what are you using as DIR?
rfbrw said:
CS8414 is even worse!
So what are you using as DIR?
Re: Re: Re: DAC - Opinion needed
There is nothing wrong with S/PDIF that can't be overcome with the application of a little intelligence. My DAC uses an oscillator and a synchronous counter to provide all the clocks needed for D/A conversion and export a clock of the appropriate frequency the slave a CDP or a word clock to sync a PC sound card or other digital source.
Along with the sample data, S/PDIF indicates whether or not each sample was correctly read from the source storage device (CD, HD, etc) and received without error. In case of error, I can choose to repeat the previous good sample or interpolate through the error. I2S doesn’t tell you squat.
S/PDIF also tells me if the samples were mastered with pre-emphasis. I know most of you don’t regard de-emphasis as a required feature of a DAC but I’ll bet many of you have ‘emphasized’ CDs and don’t even know it. For example, the Brilliant Bach set, published in 2006, includes more than a dozen CDs that were mastered with emphasis.
QSerraTico_Tico said:
There is nothing wrong with S/PDIF that can't be overcome with the application of a little intelligence. My DAC uses an oscillator and a synchronous counter to provide all the clocks needed for D/A conversion and export a clock of the appropriate frequency the slave a CDP or a word clock to sync a PC sound card or other digital source.
Along with the sample data, S/PDIF indicates whether or not each sample was correctly read from the source storage device (CD, HD, etc) and received without error. In case of error, I can choose to repeat the previous good sample or interpolate through the error. I2S doesn’t tell you squat.
S/PDIF also tells me if the samples were mastered with pre-emphasis. I know most of you don’t regard de-emphasis as a required feature of a DAC but I’ll bet many of you have ‘emphasized’ CDs and don’t even know it. For example, the Brilliant Bach set, published in 2006, includes more than a dozen CDs that were mastered with emphasis.
Re: Re: Pretty polly . Pieces of eight . SPDIF is bad
Two to be getting on with.
"Opinion is not fact" and "I have more to learn about digital audio"
This may come as a surprise to you but there where DIR's before the CS8412.
Perhaps you should learn some new phrases.QSerraTico_Tico said:
Two to be getting on with.
"Opinion is not fact" and "I have more to learn about digital audio"
This may come as a surprise to you but there where DIR's before the CS8412.
Re: Re: Re: Pretty polly . Pieces of eight . SPDIF is bad
Yeah good old YM3623 was better but you did not answer my question!
😡
rfbrw said:
Yeah good old YM3623 was better but you did not answer my question!
😡
Re: Re: Re: Re: Pretty polly . Pieces of eight . SPDIF is bad
If your history of SPDIF devices is somewhat lacking, it is not for me fill in the gaps.
QSerraTico_Tico said:
Yeah good old YM3623 was better but you did not answer my question!
😡
If your history of SPDIF devices is somewhat lacking, it is not for me fill in the gaps.
Re: Re: Re: Re: Re: Pretty polly . Pieces of eight . SPDIF is bad
What a great help.
🙄
rfbrw said:
What a great help.
🙄
Re: S/PDIF sucks
Mr. Tico_Tico, you say S/PDIF sucks and so does the CS8412/14 but you don’t say why. For the benefit of an ignoramus like me, would you please explain why? As far as I am concerned, S/PDIF only has to move the data bits from here to there and it does a pretty good job. The source of the sample clock belongs in the DAC, not anywhere else. It’s not as if the timing of the D/A conversion has to be synchronized with some global event.
Mr. Analog_sa seems to agree with you, regarding S/PDIF sucking, but he fears modifying a CDP and drilling holes in order to slave it to a DAC. I wonder why is he participating in a DIY forum where modifying commercial products is almost a way of life? Besides, I am sure anyone with a little intelligence and creativity can find a way to slave a CDP without having to drill holes and cut traces. Here’s how I did it.
I identified the components in the CDP that were involved with the oscillator and S/PDIF generation: A 74HCU04 and a 74HC74. I unsoldered those chips and installed DIP sockets in their place. I made a PCB daughter board, which plugged into the sockets, and included the ‘U04 and ‘74 and a full-duplex, differential transceiver. The PCB also has jumpers to allow the selection of either the internal or external oscillator source. On the back of the CDP I removed the optical output jack and replaced it with an RJ-11. I used Cat5 twisted pairs to connect the transceiver to the RJ-11. No holes drilled and no traces cut. The CDP can be restored to its original condition by removing the daughter board and RJ-11 and replacing the original chips and optical jack.
The modified CDP can use an external clock source or not and it outputs S/PDIF via both 110 ohm differential twisted pair (RJ-11) and 75 ohm coax (RCA). My DAC receives S/PDIF via 75 ohm coax (RCA or BNC) or 110 ohm twisted pair (RJ-11 or XLR) and transmits a 44.1 KHz word clock via 75 ohm coax (BNC) and a multiple there of (8.4672, 11.2896, 16.9344, or 33.8688 MHz) via 110 ohm twisted pair (RJ-11). The DAC also has the option of using its internal oscillator or the recovered S/PDIF clock so the DAC can be used with an un-synched digital source.
Now, what were you saying about S/PDIF? Oh yeah, it sucks. What part of digital audio doesn’t suck?
Mr. Tico_Tico, you say S/PDIF sucks and so does the CS8412/14 but you don’t say why. For the benefit of an ignoramus like me, would you please explain why? As far as I am concerned, S/PDIF only has to move the data bits from here to there and it does a pretty good job. The source of the sample clock belongs in the DAC, not anywhere else. It’s not as if the timing of the D/A conversion has to be synchronized with some global event.
Mr. Analog_sa seems to agree with you, regarding S/PDIF sucking, but he fears modifying a CDP and drilling holes in order to slave it to a DAC. I wonder why is he participating in a DIY forum where modifying commercial products is almost a way of life? Besides, I am sure anyone with a little intelligence and creativity can find a way to slave a CDP without having to drill holes and cut traces. Here’s how I did it.
I identified the components in the CDP that were involved with the oscillator and S/PDIF generation: A 74HCU04 and a 74HC74. I unsoldered those chips and installed DIP sockets in their place. I made a PCB daughter board, which plugged into the sockets, and included the ‘U04 and ‘74 and a full-duplex, differential transceiver. The PCB also has jumpers to allow the selection of either the internal or external oscillator source. On the back of the CDP I removed the optical output jack and replaced it with an RJ-11. I used Cat5 twisted pairs to connect the transceiver to the RJ-11. No holes drilled and no traces cut. The CDP can be restored to its original condition by removing the daughter board and RJ-11 and replacing the original chips and optical jack.
The modified CDP can use an external clock source or not and it outputs S/PDIF via both 110 ohm differential twisted pair (RJ-11) and 75 ohm coax (RCA). My DAC receives S/PDIF via 75 ohm coax (RCA or BNC) or 110 ohm twisted pair (RJ-11 or XLR) and transmits a 44.1 KHz word clock via 75 ohm coax (BNC) and a multiple there of (8.4672, 11.2896, 16.9344, or 33.8688 MHz) via 110 ohm twisted pair (RJ-11). The DAC also has the option of using its internal oscillator or the recovered S/PDIF clock so the DAC can be used with an un-synched digital source.
Now, what were you saying about S/PDIF? Oh yeah, it sucks. What part of digital audio doesn’t suck?
Re: Re: S/PDIF sucks
OK why it sucks:
1)The SPDIF is sent through a 75-Ohm transmission line but by standard it is wrongly terminated with RCAs / Cinch.
These are NOT 75 Ohm so reflections result. Often NON 75-Ohm cables are also used worsening the situation.
2) The clock is generated from the SPDIF stream with a PLL, as the datasheet of the CS8412 indicates lower frequency jitter is not suppressed. Anyway the clock is far less good than in the player.
3) Many Philips players use a very poor SPDIF driver especially the transformer.
All this trouble to get all audio through one coax cable.
Also imagine one has 1st to encode SPDIF in the player then decode in the DAC.
I2S Direct is far more simple as this encoding/decoding process is omitted. Yes one has to drill holes. Even four if you have a DAC with digital filter. [DATA, BCK, WS, CLOCK]
whatsup said:
OK why it sucks:
1)The SPDIF is sent through a 75-Ohm transmission line but by standard it is wrongly terminated with RCAs / Cinch.
These are NOT 75 Ohm so reflections result. Often NON 75-Ohm cables are also used worsening the situation.
2) The clock is generated from the SPDIF stream with a PLL, as the datasheet of the CS8412 indicates lower frequency jitter is not suppressed. Anyway the clock is far less good than in the player.
3) Many Philips players use a very poor SPDIF driver especially the transformer.
All this trouble to get all audio through one coax cable.
Also imagine one has 1st to encode SPDIF in the player then decode in the DAC.
I2S Direct is far more simple as this encoding/decoding process is omitted. Yes one has to drill holes. Even four if you have a DAC with digital filter. [DATA, BCK, WS, CLOCK]
Attachments
I have buffered the Spdif from my CD304mk2 with a chip DS8922. Removed transformer and coaxial jack, mounted Svhs socket (differential output) All Ecdesigns idea. Spdif output waveform on scope looks sharper now (less jitter for sure) Will try a dac soon.
Interesting I2S solution.
Interesting I2S solution.
- Status
- Not open for further replies.