|
|
Post by krzysztof on Jun 22, 2015 5:46:27 GMT
Anyone interested should email me. I am not good with remembering requests on forums for some reason. Sure and thanks. I've sent you PM. |
|
|
|
Post by ChrisB on Jun 22, 2015 6:23:18 GMT
Krzysztof, I think you may have misunderstood Stan. I think he would prefer an email to a PM. You will find his address in his forum profile - theaudiostandard.net/user/64 |
|
|
|
Post by krzysztof on Jun 22, 2015 6:28:34 GMT
Krzysztof, I think you may have misunderstood Stan. I think he would prefer an email to a PM. You will find his address in his forum profile - theaudiostandard.net/user/64Ouch, I've sent the email to him, but was not precise on the forum. By PM I meant an email ;-) |
|
|
|
Post by ChrisB on Jun 22, 2015 6:29:27 GMT
In that case, all is good!
|
|
|
|
Post by krzysztof on Jun 22, 2015 8:29:52 GMT
For the sake of my own curiosity, reading the initial feedback about better imaging, soundstaging etc this could be done by something similar to using asynchronous techniques, when input phase and the output phase is matched by referencing the frequency and if all that is matched up the timing will be correct. Bass is improved, better soundstage/imaging , pacing is more accurate according to some sources. According to an AES paper (chapter 5 also I think) www.audiocraftersguild.com/AandE/npt.on.jitter2.htm:Popular S/PDIF receiver chips like the Yamaha YM3623B and Crystal CS8412 are NOT crystal controlled but rather recover the necessary clock from internal Phase Locked Loops (PLL) locked onto the incoming data stream. The simple two pin can crystals often seen directly attached to '3623's and '8412's are optional. The 3623 uses the crystal clock to quickly lock onto the S/PDIF signal. The 8412 uses the crystal clock to determine and display the sample rate and jitter level of the S/PDIF signal. Both parts ignore the local crystal clock once locked onto the S/PDIF signal. Higher end parts will use a sample rate converter to further clean the jitter. Both TI and Cirrus have spdif receivers with built in ASRCs It is noted that the use of ASRC is controversial in high end audio (there are camps on both sides). So, if I understood properly, PLL recovery based on the sample rate is rather limited to high end devices, but could be problematic when the source is not top quality and could cause locking problems. Bravo for Stan if he has managed that stable. Lots of interesting stuff here also: www.jitter.de/english/how.html |
|
|
|
Post by krzysztof on Jun 23, 2015 9:03:55 GMT
My SFW3 chip is on the way. I will put the feedback here after some listening tests for sure. BTW, anyone who already have it could chime in to say about the perceived differences in the processing modes and what you're preferring? IMHO, this is quite interesting how to change the sound signature using the PIC programming only, isn't it?
|
|
|
|
Post by MartinT on Jun 23, 2015 10:24:38 GMT
I think it's fascinating what can be achieved, Krzysztof. I've been listening to two early versions of the SFW firmware and am also awaiting the latest version with the different modes to arrive. What Stan has done with the Caiman-II has taken it to serious competition for any DAC I've heard. As you say, amazing what programming and use of the on-chip facilities can achieve.
|
|
|
|
Post by stanleyb on Jun 23, 2015 11:08:46 GMT
What is also a pleasant surprise to me is that the differences are also clear to hear with the standard power supply. I was initially worried that I would only pick up the improvements when the Caiman was powered by batteries.
I use a very cheap DVD player for many of my tests, including TOSLINK and coax differences. I know that the TOSLINK on the player is nowhere near the quality of the coax. It is a 10 year old player after all. The coax now sounds even better on this old player.
|
|
|
|
Post by krzysztof on Jun 23, 2015 11:19:22 GMT
I think it's fascinating what can be achieved, Krzysztof. I've been listening to two early versions of the SFW firmware and am also awaiting the latest version with the different modes to arrive. What Stan has done with the Caiman-II has taken it to serious competition for any DAC I've heard. As you say, amazing what programming and use of the on-chip facilities can achieve. How many DAC you've heard? I am very interested in the PLL enabled mode which could be programmed with controlling active bits in the interrupt registers I am guessing, then locking PLL and choosing N-factor according to the input sample rate could almost remove any jitter attenuation in the output signal - this is I guess Mode-4. This however could be hard to make it stable and could not lock properly to the input signal. Don't know if Stan is using voltage control oscillators as well as the another possible option for improvements. Savvy stuff! |
|
|
|
Post by MartinT on Jun 23, 2015 11:46:43 GMT
Quite a few, including models from the Absolute Sounds stable and the top dCS kit.
|
|
|
|
Post by krzysztof on Jun 23, 2015 11:47:40 GMT
Quite a few, including models from the Absolute Sounds stable and the top dCS kit. Oh, interesting to compare CMII to dCS kit ;-) |
|
|
|
Post by MartinT on Jun 23, 2015 11:51:45 GMT
I wonder if Minimum Phase is possible in firmware? When my Ayre was upgraded to MP level, it was a big jump in SQ. |
|
|
|
Post by MartinT on Jun 23, 2015 11:54:06 GMT
Oh, interesting to compare CMII to dCS kit ;-) I'm NOT saying they're equivalent, but the C-II is starting to reach that level of detail retrieval. As for SQ, that's an even more interesting comparison. |
|
|
|
Post by krzysztof on Jun 23, 2015 12:23:09 GMT
I wonder if Minimum Phase is possible in firmware? When my Ayre was upgraded to MP level, it was a big jump in SQ. Depend on the surrounding DAC implementation I guess. Not sure if Stan's improvements are touching FIRs as well. Guessing that he payed the attention to the reference clock and its frequency adaptation using PLL or VCXO if CMII already has it. Just my guessing… |
|
|
|
Post by MartinT on Jun 24, 2015 17:25:27 GMT
Listening to Stan's SFW3 with four data processing options. To my ears, options 3/4 sound much better than 1/2. The differences between 3 and 4 seem to be dependent on the format:
1. Standard XTAL retrieval: sounds like the Caiman-II of old.
2. Optimised XTAL retrieval: sounds much the same as far as I can hear.
3. DATA recovered clock: this sounds excellent with many of my CDs, spacious, detailed and with good flow.
4. PLL clock: this sounds especially good with SACD sent as 1-bit, improving the incisiveness while still preserving the soundstage. Overall the most obviously detailed of the four, but 3 sounds better with CDs.
I need to listen to a lot more music so the above are just impressions after playing a few discs.
|
|
|
|
Post by krzysztof on Jun 24, 2015 19:44:10 GMT
Listening to Stan's SFW3 with four data processing options. To my ears, options 3/4 sound much better than 1/2. The differences between 3 and 4 seem to be dependent on the format: 1. Standard XTAL retrieval: sounds like the Caiman-II of old. 2. Optimised XTAL retrieval: sounds much the same as far as I can hear. 3. DATA recovered clock: this sounds excellent with many of my CDs, spacious, detailed and with good flow. 4. PLL clock: this sounds especially good with SACD sent as 1-bit, improving the incisiveness while still preserving the soundstage. Overall the most obviously detailed of the four, but 3 sounds better with CDs. I need to listen to a lot more music so the above are just impressions after playing a few discs. Nice. As I was suspected, that 1 and 2 will be similar to v6/100. I will concentrate mostly on 3 and 4 since I already have v6/100. Try to listen to deep acoustic double bass passages alone and with some jazzy touch shimmers of brushes with that as well. Also good recording with good grand piano, like Fazioli or Boesendorfer - Paul Bley Solo in Mondsee for instance. In this kind of detail I think we should compare the edges of the particular instrument tone and timbre. Soundstage is mostly subject to another stage, but could be better if S/N decreased - could be deeper I think. The good piano recording could reveal some differences between the SPMS and battery power as I heard on my own. Very subtle of course, but more smooth and not so edgy on the top of the register. |
|
|
|
Post by Greg on Jun 25, 2015 18:38:45 GMT
Listening to Stan's SFW3 with four data processing options. To my ears, options 3/4 sound much better than 1/2. The differences between 3 and 4 seem to be dependent on the format: 1. Standard XTAL retrieval: sounds like the Caiman-II of old. 2. Optimised XTAL retrieval: sounds much the same as far as I can hear. 3. DATA recovered clock: this sounds excellent with many of my CDs, spacious, detailed and with good flow. 4. PLL clock: this sounds especially good with SACD sent as 1-bit, improving the incisiveness while still preserving the soundstage. Overall the most obviously detailed of the four, but 3 sounds better with CDs. I need to listen to a lot more music so the above are just impressions after playing a few discs. Martin, I've been following these latest developments with interest and have ordered the latest firmware from Stan. I have an RPi2 with Hi-Fi Berry which I use for listening to streamed flac and high resolution flac services. Are you able to comment on how such services sound with the new firmware installed? Thanks, Greg |
|
|
|
Post by MartinT on Jun 25, 2015 20:37:31 GMT
Greg, the RPi2 with HiFiBerry Digi+ with C-II / SFW3 are a superb combination. Definitely excellent music making and the DAC is left mainly in mode 4. I have very little on FLAC but what I have tested so far sounds indistinguishable from WAV to me, and very good indeed. Streams like Spotify hi-bitrate really benefit from the new-found clocking stability.
|
|
|
|
Post by Greg on Jun 25, 2015 21:51:51 GMT
Thanks Martin for your rapid response. That is very encouraging. I look forward to installing SFW3.  Oh, and as ever, once again, well done Mr Beresford! |
|
|
|
Post by MartinT on Jun 26, 2015 6:22:29 GMT
Oh, and as ever, once again, well done Mr Beresford! Agreed! |
|
