Thank you for saying so! Nice to know there are interested people, not just the critics.
Thanks to the others that have been supportive and or in agreement, as well!
I try to be helpful and informative, but it tends to get tiring with the trolls and hecklers showing up in every thread.
🙂
Last edited:
Here’s my take on that plasnu,
If your looking for correctness in reproduction you must reproduce at the proper level.
Example, a folk singer and acoustic guitar should not be reproduced much louder than what it was in the first place......things start to get over embellished.
Now, if we’re talking about most any rock based bands, it’s meant to be heard loud......in fact the music doesn’t even come across at lower levels (unless the recording is over compressed)
I’ve listened to live/loud music my entire life and it’s not been detrimental to my hearing at all, except for the normal age decline of hf (can only get up to 14k +/- nowadays (54) my hearing is actually better than it’s ever been. Hopefully it’s not a fleeting thing, like when a small block Chevy runs best right b4 it blows!
If your looking for correctness in reproduction you must reproduce at the proper level.
Example, a folk singer and acoustic guitar should not be reproduced much louder than what it was in the first place......things start to get over embellished.
Now, if we’re talking about most any rock based bands, it’s meant to be heard loud......in fact the music doesn’t even come across at lower levels (unless the recording is over compressed)
I’ve listened to live/loud music my entire life and it’s not been detrimental to my hearing at all, except for the normal age decline of hf (can only get up to 14k +/- nowadays (54) my hearing is actually better than it’s ever been. Hopefully it’s not a fleeting thing, like when a small block Chevy runs best right b4 it blows!
I wouldn't say I2SoverUSB has a sota i2s layout and ground design for high speed dac chips mentionned here. It was made more in the spirit to use pcm chips at the firt beginning.... Imho Markw4 missed the point here. Such sensible fast chips deserves or more exactly NEED the best designs which are hard to do with multiple boards. At least I will not put less than a Iancanada board with such chips with reliable interlinks.
With no offense but in a personal thought, I am a little worried, often, by the advices Markw4 is giving...Looks like tottaly subjective, shortcuted and limited.... this dac is the best, this dac chip is the best now, this ldo...big Wima caps gave me the best and so on. I'm not sure he has big experience of music diversity and unplugged one. That's just a feeling but seems to me he is looking at dacs for not so much long time. Not a personal attack, just think lake of experience, need longer experience curve which will be reached sooner or later, just to say I will not class him in Dac authorities and most of the time pass away when he talks about powersupplies for dacs. For the moment I would take his advices with a grain of salt while enlight he is sharing his experience and sharing is always helpfull... but just a little to confident with himself for the moment... And yes measurements alone mean not all or it will be easy to do dacs at the outputt of EE school.
Imho and YMMV of course. Just to relativise and say not all is bad to take...
Also I'm not sure why the word Science is always involved in all our threads? Are we dealing with sorcery here ? The words knowledge and experience (humility too)seem imho glue more... especially when it comes to advice a standalone device to someone without knowing too much about...the rest of parameters.
With no offense but in a personal thought, I am a little worried, often, by the advices Markw4 is giving...Looks like tottaly subjective, shortcuted and limited.... this dac is the best, this dac chip is the best now, this ldo...big Wima caps gave me the best and so on. I'm not sure he has big experience of music diversity and unplugged one. That's just a feeling but seems to me he is looking at dacs for not so much long time. Not a personal attack, just think lake of experience, need longer experience curve which will be reached sooner or later, just to say I will not class him in Dac authorities and most of the time pass away when he talks about powersupplies for dacs. For the moment I would take his advices with a grain of salt while enlight he is sharing his experience and sharing is always helpfull... but just a little to confident with himself for the moment... And yes measurements alone mean not all or it will be easy to do dacs at the outputt of EE school.
Imho and YMMV of course. Just to relativise and say not all is bad to take...
Also I'm not sure why the word Science is always involved in all our threads? Are we dealing with sorcery here ? The words knowledge and experience (humility too)seem imho glue more... especially when it comes to advice a standalone device to someone without knowing too much about...the rest of parameters.
Actually, I bought and listened to Ian's FIFO_Pi and the other FIFO with a separate clock board (bought that too).
Also tried u.fl and SMA connectors, along with pin headers. uf.l are not very reliable with repeated use. The construction is poor with a flimsy crimped ground for the shield, no solder. SMA are the best, assuming a good brand and that they are properly torqued and if good quality coax is used. Even SMA with cheap coax is audible bad.
Pin headers are workable if gold plated and wires are kept very short.
FIFO_Pi is okay, the other FIFO did not work so well. Plug in clocks using dip sockets is unreliable and adds inductance. Same for TP, not satisfactory in my comparisons, excessive audible jitter.
I suggested to Ian that he put a MCLK divide by 2 feather on FIFO Pi so it could be used with AKM dacs, but he had no interest.
Also tried diyinhk, and several Chinese boards.
I2SoverUSB had the lowest audible jitter of all of them. It also provides the correct MCLK frequencies for AK4499.
Don't think just because I don't tell you everything I know or have done doesn't mean you should assume I don't have good reasons for recommending what I do.
Last edited:
Have you asked Markw4 how he knows about the sound quality that he claimed above? Sharing such knowledge would be what really counts because that's where the "rubber meets the road" in audio term. If he doesn't (he didn't so far), then you should be skeptical of his claims.
Sure Markw4,
I also tested these stuffs and not only for Pi hats. IanCanada stuffs need also some work on the side of'power supplies especially as all the splitted designs btw.
My point is we should stay always open minded that means not definitive in our words.... You know all is about trade off in hifi stuffs.
I also tested these stuffs and not only for Pi hats. IanCanada stuffs need also some work on the side of'power supplies especially as all the splitted designs btw.
My point is we should stay always open minded that means not definitive in our words.... You know all is about trade off in hifi stuffs.
No offense nor troll, but I would like to hear Markw4's subjective reviews of several different off the shelf DACs. They should be very different each other, and I'm just curious what he thinks about pros and cons of them at different sampling rates and different filter settings.
Last edited:
Not too many dacs come through here. There is the Benchmark DAC-3 of course. Jam brought over an Auralic Vega and a Chord Qutest. There a Lynx II sound card, a Focusrite Scarlett 6i6, and RMI ADI-2. There are a couple of RPi dac hats: Allo Katana, and Iancanade Dual ES9038Q2M. There have been some others, don't remember the names offhand.
Don't know that I can describe them all in detail, but ranking the top few:
1. AK4499 as powered and clocked at this point
2. My 2nd modded ES9038Q2M, AC power filtered by Monster HTPS 7000 MkII.
3. Benchmark DAC-3
4. Qutest (playing DSD256)
5. Vega
Note: not sure if #4 and #5 might be swapped.
Note 2: #2 based on votes by other listeners and enjoyment of listening to music played on it. Not so much of that Sabre sound that some people find hard to listen to for very long since it has/had hardware upsampling to DSD256. Not better than #3 in certain respects.
....
It gets kind of hard after that to rank them exactly. Would have to get them all back here lined up in a row and start doing a bubble sort on the ones that sound best.
AK4499 is the cleanest, most detailed, and most musical, best with very high quality DSD512. Power supplies and clock/USB_board have a lot of influence on the sound. Some (common mode?) RF leakage though XLR audio outputs. XLR cables can be used to attenuate RF before it gets into HPA, depending on cable type and length. AC grounding through Monster HTPS 7000 MkII.
Comparison of AK4499 to other, more expensive dacs done by Jam based on his memory of the high end dacs he has heard. Neither of us has heard Chord DAVE. We would also like to give a listen to 'Holo Spring 2 Level 2' dac, just to try out a good ladder dac for a change.
Don't know that I can describe them all in detail, but ranking the top few:
1. AK4499 as powered and clocked at this point
2. My 2nd modded ES9038Q2M, AC power filtered by Monster HTPS 7000 MkII.
3. Benchmark DAC-3
4. Qutest (playing DSD256)
5. Vega
Note: not sure if #4 and #5 might be swapped.
Note 2: #2 based on votes by other listeners and enjoyment of listening to music played on it. Not so much of that Sabre sound that some people find hard to listen to for very long since it has/had hardware upsampling to DSD256. Not better than #3 in certain respects.
....
It gets kind of hard after that to rank them exactly. Would have to get them all back here lined up in a row and start doing a bubble sort on the ones that sound best.
AK4499 is the cleanest, most detailed, and most musical, best with very high quality DSD512. Power supplies and clock/USB_board have a lot of influence on the sound. Some (common mode?) RF leakage though XLR audio outputs. XLR cables can be used to attenuate RF before it gets into HPA, depending on cable type and length. AC grounding through Monster HTPS 7000 MkII.
Comparison of AK4499 to other, more expensive dacs done by Jam based on his memory of the high end dacs he has heard. Neither of us has heard Chord DAVE. We would also like to give a listen to 'Holo Spring 2 Level 2' dac, just to try out a good ladder dac for a change.
Last edited:
Thank you, Markw4. 2 questions.
1. Your reference sampling rate is pre-upsampled DSD only?
2. As a current user of Sabre, I would like to hear a little more about 1 and 2 (hopefully in PCM mode.)
1. Your reference sampling rate is pre-upsampled DSD only?
2. As a current user of Sabre, I would like to hear a little more about 1 and 2 (hopefully in PCM mode.)
Plasnu,
1. No. I always try at least 16/44 and upsampled DSD (to the extent possible per dac).
2. I think 1 sounds more technically correct and more musical. Just a better dac chip to start with I think. Main problem with 2 was a little bit weak or less forward in the lower midrange, where the chest resonance of vocals tends to be (DAC-3 is very well balanced there due to the FPGA external PCM interpolation filter, IMHO). My guess would be that more work on the output stage might have improved that for 2. If upsampling to DSD had been done in FPGA rather than AK4137 then fixing that might have been done in the upsampling interpolation filter.
1. No. I always try at least 16/44 and upsampled DSD (to the extent possible per dac).
2. I think 1 sounds more technically correct and more musical. Just a better dac chip to start with I think. Main problem with 2 was a little bit weak or less forward in the lower midrange, where the chest resonance of vocals tends to be (DAC-3 is very well balanced there due to the FPGA external PCM interpolation filter, IMHO). My guess would be that more work on the output stage might have improved that for 2. If upsampling to DSD had been done in FPGA rather than AK4137 then fixing that might have been done in the upsampling interpolation filter.
Ok, thank you. I'm looking forward to have a chance to hear 4499 in the future. I'm not finding any problem with lower midrange of my Sabre DAC so far, but I know I cannot clearly see the problem UNTIL I hear the better one. (as always 🙂)
BTW, what do you think about Sabre and AKM internal FIR filter at 44.1K? I'm sure that people would have very different experiences depends on their filter setting of Sabre... Recommended or your preferred setting?
PS: Well, I also use software upsampling interpolation filter, so I will test my DAC with 44.1K native to make sure if I find the problem with the lower midrange.
BTW, what do you think about Sabre and AKM internal FIR filter at 44.1K? I'm sure that people would have very different experiences depends on their filter setting of Sabre... Recommended or your preferred setting?
PS: Well, I also use software upsampling interpolation filter, so I will test my DAC with 44.1K native to make sure if I find the problem with the lower midrange.
Last edited:
Surely the big problem here is that the request is for a 'good to great' DAC and one mans great is another mans poor? Withput any starting point it seems similar to asking what a great car is and wondering why you get 50 different conflicting answers?
For myself I have been told on good authority by several people that my system is clearly rubbish which is why I have such low standards 😛
For myself I have been told on good authority by several people that my system is clearly rubbish which is why I have such low standards 😛
No, you merely opined that my system was possibly not resolving enough. I've had proper dissings from the usual suspects 🙂
Plasnu,
Filters tend to be an issue for me since they all involve various engineering trade offs. With dacs I usually find that the default filter sounds best if the implementation around the dac chip is not problematic in some way that probably should be fixed.
One example of a trade off, slow filters tend to cut into the top of the audio band noticeably altering high frequency response. That's not good.
Most experience with particular attention to filter preference was as I noticed during the development of modded dac #2 that which filter sounded best changed a number of times as various implementation shortcomings were found and fixed.
Same thing happened to a lesser extent as I observed Allo fixing the original Katana dac harware filtering. I started out with a pre-release Katana and found it sounded better with more hardware filtering. So did other test reviewers. Allo was able to confirm the problem with measurements. They made substantial improvements to hardware filtering before final release. The final version sounded best to me with the default dac chip interpolation filter.
Filters tend to be an issue for me since they all involve various engineering trade offs. With dacs I usually find that the default filter sounds best if the implementation around the dac chip is not problematic in some way that probably should be fixed.
One example of a trade off, slow filters tend to cut into the top of the audio band noticeably altering high frequency response. That's not good.
Most experience with particular attention to filter preference was as I noticed during the development of modded dac #2 that which filter sounded best changed a number of times as various implementation shortcomings were found and fixed.
Same thing happened to a lesser extent as I observed Allo fixing the original Katana dac harware filtering. I started out with a pre-release Katana and found it sounded better with more hardware filtering. So did other test reviewers. Allo was able to confirm the problem with measurements. They made substantial improvements to hardware filtering before final release. The final version sounded best to me with the default dac chip interpolation filter.
Yes, possibly.
Don't know. More of an open question.
Hard to explain online. If you ever visit it will be much easier just to show you what I am thinking about. Hopefully, it would be a fun thing for you to do.
One warning though: You will be eaten alive here if you ever admit to some of what you heard. 🙂
Last edited:
Some of what could be heard could be provided without people having to pop round to yours from all over the world. 🙂
ScottJ
Maybe after we have a few visitors. Probably should be apparent why later if not obvious now.
Sorry, right now it is my word against a bunch of people that will shoot first and ask questions later. That would be true even for how recordings are made and what they are actually recordings of.
Maybe after we have a few visitors. Probably should be apparent why later if not obvious now.
Sorry, right now it is my word against a bunch of people that will shoot first and ask questions later. That would be true even for how recordings are made and what they are actually recordings of.
Last edited:
Markw4, I personally prefer gentle filter at 44.1k, but I think I understand what you say. It makes whole spectrum a little more midrangy (someone would say analogue or musical), and little less hi-fi ultimately. I think it’s all depends on the taste, music genre or system synergy, though. I currently use gentle minimal phase filter for my software upsampling (you probably not.). Whenever I try sharp filter, I always come back to gentle one later. Maybe my EQ setting and equipments are a bit more bright than yours.
Anyway, it’s good to know that I probably can hear at least a part of what you hear. I have not tested 44.1K native yet.
Anyway, it’s good to know that I probably can hear at least a part of what you hear. I have not tested 44.1K native yet.