Hello! There are no changes, I just needed a new set of WaveIO boards 
External PSU: this card will accept external supply connected to the screw-type power connector. The voltage was set to 5V DC or at least up to 6V because there are few Vregs which can't withstand to more than 6V Vin.
Happy New Year to all of you,
L
External PSU: this card will accept external supply connected to the screw-type power connector. The voltage was set to 5V DC or at least up to 6V because there are few Vregs which can't withstand to more than 6V Vin.
Happy New Year to all of you,
L
Last edited:
Hi Lorien,
I'm a complete know-nothing on the subject of electronics but I am interested in the potential of I2S. I do have a good relationship with a very competent professional audio engineer and I am interested in exploring the application of I2S with his help. Can you advise me of the cost of the WaveIO boards as and when you have them available.
TIA,
Dave.
I'm a complete know-nothing on the subject of electronics but I am interested in the potential of I2S. I do have a good relationship with a very competent professional audio engineer and I am interested in exploring the application of I2S with his help. Can you advise me of the cost of the WaveIO boards as and when you have them available.
TIA,
Dave.
I hope your friend is a professional audio electronics engineer, because most audio engineers won't have any experience with I2S. You'd be better off with a digital electronics engineer who knows nothing about audio than an audio engineer who knows nothing about digital circuit design.
My apologies if you were already aware of the distinction and merely used different terms. I'm only picking nits because the world of audio engineering is huge and full of many competent professionals, but nearly all of them have never connected an I2S bus between some sort of digital stream and a raw DAC digital input pin.
@ DaveK: I think I know who's the 'very competent professional audio engineer' you're talking about... so you're in good hands You have PM.
@ audiodesign: The cards are in customs now and I hope I'll have in my hands in few days. After that I'll need (other) few days to fulfill some delayed orders.
Kind regards,
L
You have PM.@ audiodesign: The cards are in customs now and I hope I'll have in my hands in few days. After that I'll need (other) few days to fulfill some delayed orders.
Kind regards,
L
A little i2s history
Frankly, I am thankful for the latency in shipping so I could digest my 'wire' sandwiches and JC-80 space heater. The digital side opus izReady to gobble WaveIO bitz when it arrives.
i2s has been around nearly forever. Audio Alchemy pushed it for interconnect (in fact 'Audio Al* i2s' will turn up surplus cables on eBay) years ago in their now classic little boxes. I have attached what I believe to be a not copyrighted schematic illustrating 5v logic for 16/44.1k i2s input. This may be helpful for reference but likely needs adaptation for 3.3volts.
VERY few audio engineerz will have ever encountered i2s IMHO. It was designed for short-run IC interconnect and excels at jitter-free transfers.
Frankly, I am thankful for the latency in shipping so I could digest my 'wire' sandwiches and JC-80 space heater. The digital side opus izReady to gobble WaveIO bitz when it arrives.
i2s has been around nearly forever. Audio Alchemy pushed it for interconnect (in fact 'Audio Al* i2s' will turn up surplus cables on eBay) years ago in their now classic little boxes. I have attached what I believe to be a not copyrighted schematic illustrating 5v logic for 16/44.1k i2s input. This may be helpful for reference but likely needs adaptation for 3.3volts.
VERY few audio engineerz will have ever encountered i2s IMHO. It was designed for short-run IC interconnect and excels at jitter-free transfers.
I do not see the attachment. Can you just upload it to a site somewhere and provide a URL? ... or would it be better to attach as a file hosted at diyAudio? Either way, I would definitely be interested in taking a look at that schematic.
Oh!
Lemme try again. DDE3.pdf: Your file of 253.9 KB bytes exceeds the forum's limit of 195.3 KB for this filetype. That 'splainzIt.
If this part of the file izInadequate for your needs, you can download original DDE3.pdf
http://www.audioasylum.com/images/DDE3.pdf
Lemme try again. DDE3.pdf: Your file of 253.9 KB bytes exceeds the forum's limit of 195.3 KB for this filetype. That 'splainzIt.
If this part of the file izInadequate for your needs, you can download original DDE3.pdf
http://www.audioasylum.com/images/DDE3.pdf
Last edited:
The serial alternative requires clock extraction from the signal and is without a doubt MUCH worse. If you plan to move bitz electronicly your options are limited and parallel bus technology has advanced but at a price. Lorian knows better than I what is available.
Compared with internal computer bus speeds or even modern scsi, i2s is horse and buggy stuff. It is the native language of virtually all audio ICs today so unless you want to go thru more levels of translation, get used to the idea or get used to even worse jitter.
Single ended...
I2S is excellent for short distance transmission (under 10 cm), assuming it is well implemented. For longer distances, LVDS balanced I2S as used by PS Audio, Wyred 4 Sound, and with Twisted Pear Audios' new Bit Teleporter can be used for much longer distances with no compromise.
I2S is excellent for short distance transmission (under 10 cm), assuming it is well implemented. For longer distances, LVDS balanced I2S as used by PS Audio, Wyred 4 Sound, and with Twisted Pear Audios' new Bit Teleporter can be used for much longer distances with no compromise.
In addition to what Wolfsin has said, I think i2s is perhaps a little more implementation sensitive than spdif. It is important that the signal connections are made appropriately to ensure signal integrity is maintained and maximises the benefits of the parallel data transfer.
This is why I believe terminating to u.fl connectors for i2s between pcbs as available on Lorien's WaveIO board (this thread) is a good idea. I have no practical or measured data to support this though, yet. The topic of i2s signal wire terminations came up briefly some time back in this thread though from memory.
edit: barrows can type faster and less verbosely than I.
This is why I believe terminating to u.fl connectors for i2s between pcbs as available on Lorien's WaveIO board (this thread) is a good idea. I have no practical or measured data to support this though, yet. The topic of i2s signal wire terminations came up briefly some time back in this thread though from memory.
edit: barrows can type faster and less verbosely than I.
Ha Ha...
not really! I like u.fl as well, but still do not really "know" if it is superior. Lorien did make a remark that he has seen clock signals get degraded by pin headers... Most u.fl connectors I am aware of use a right angle fitting, which may be a compromise for clock signals (just as sharp right angle PCB traces are not ideal for clock signals).
In my current DAC, I am using pin headers with silver plated copper/FEP ribbon cable for I2S, and the masterclock goes over a separate u.fl coax cable and connectors. I do not know if the I2S signal is degraded by the pin headers/ribbon, but it sure is sounding fantastic, way better than the same DAC by SPDIF... I have seen pin headers and ribbon cable used in quite a few very high end DACs for carrying I2S signals: while this may not be the "best" way, it surely works well enough for Ayre, dCS, AMR, etc.
I would love to see the results of some testing on this by someone with a really good high speed 'scope and some 80 MHz square waves...
not really! I like u.fl as well, but still do not really "know" if it is superior. Lorien did make a remark that he has seen clock signals get degraded by pin headers... Most u.fl connectors I am aware of use a right angle fitting, which may be a compromise for clock signals (just as sharp right angle PCB traces are not ideal for clock signals).
In my current DAC, I am using pin headers with silver plated copper/FEP ribbon cable for I2S, and the masterclock goes over a separate u.fl coax cable and connectors. I do not know if the I2S signal is degraded by the pin headers/ribbon, but it sure is sounding fantastic, way better than the same DAC by SPDIF... I have seen pin headers and ribbon cable used in quite a few very high end DACs for carrying I2S signals: while this may not be the "best" way, it surely works well enough for Ayre, dCS, AMR, etc.
I would love to see the results of some testing on this by someone with a really good high speed 'scope and some 80 MHz square waves...
Excellent summary.
It is important to keep in mind the design constraints of any existing solution. I2S was never designed to connect two separate systems which do not share an identical ground reference, so I would say that not only does I2S work best over short distances, but it almost absolutely needs both ends running on the same power supply. As distances get longer, ground loops appear or ground potentials increase, and when the distance spans two enclosures you almost surely have a separate power supply, making the problem even worse.
If you use a standard outside of its design constraints, then you're very likely to fail to get the original benefits. Whenever you connect two separate enclosures with different power supplies, you need some kind of interface that is designed to deal with the different references. I2S was not designed for that. But it makes sense that "LVDS balanced I2S" could enhance plain I2S with the balanced additions handling the potential differences in reference levels (ground).
There's no need to get into the drawbacks of SPDIF and AES3, but the one thing those ancient standards have going for them is that they were absolutely designed to connect separate pieces of gear. While it's obvious that we need to look for modern replacements so that we can finally say goodbye to the failures of SPDIF & AES3, that doesn't make it possible to shoehorn something that was not designed to overcome the same obstacles into a solution without considering the additional requirements.
I think it's great that PS Audio, Wyred 4 Sound, and Twisted Pear Audio have pioneered a better solution. Are there any convenient links to the technical details of these interconnects? Sorry if the question has been answered in this thread already, but I recall suggesting LVDS interconnects on diyAudio and someone suggested that it would be a bad idea due to increased jitter. If the above three companies are having success with LVDS I2S, then I'd like to learn more.
Agree, it is far from a sure thing but giving some consideration to the connection between circuit boards seems prudent when dealing with a clock signal regardless of preferred method.
I second that idea! I expect it'll be a few years before I can afford the sort of scope that will give any meaningful answer to these questions. In the mean time I'll enjoy the music instead of chasing the last 0.1% of performance from i2s termination.
I've not had much chance to read up on LVDS balanced i2s, yet. Though I have no need to ever traverse more than a few cm between source and DAC in any of my current plans it is certainly an interesting topic!
I second that idea! I expect it'll be a few years before I can afford the sort of scope that will give any meaningful answer to these questions. In the mean time I'll enjoy the music instead of chasing the last 0.1% of performance from i2s termination.
I've not had much chance to read up on LVDS balanced i2s, yet. Though I have no need to ever traverse more than a few cm between source and DAC in any of my current plans it is certainly an interesting topic!
qusp...
as noted, I use u.fl. But there are coax alternatives which do not make the right angle turn. I am curious what experienced RF design engineers might have to say about this. I have heard that pcb board traces are often purposely designed with gentle curves rather than hard right angles for high speed signals. How much difference these choices might make are fun to speculate about, but without some serious measurements, it really is just all speculation.
as noted, I use u.fl. But there are coax alternatives which do not make the right angle turn. I am curious what experienced RF design engineers might have to say about this. I have heard that pcb board traces are often purposely designed with gentle curves rather than hard right angles for high speed signals. How much difference these choices might make are fun to speculate about, but without some serious measurements, it really is just all speculation.
Though it is not a technical detail, a schematic for transmitter side of PS-Audio is available from here.
http://www.fidelix.jp/img/PS-HDMI.jpg
A Japanese DAC, CAPRICE made by Fidelix provides an option of receiver side compatible with the PS-Audio standard and Chiaki's SDTrans384 has a built-in interface of transmitter side. I have never perceived any serious degradation in resulting sounds through this connection even for 352.8 kHz/24 bit PCM or DSD128 play. However, when I use I2S between board kits, I use wires of one inch length.
Bunpei
