I must say again you have an SOTA setup for both listening and measurement 
Hopefully, the output side can be improved further…as you reckon
Adding this ‘no-dac’ filter in my opinion does no justice. I have seen the excited merriment at this no-dac camp before as to how natural sounding and unbridled sound stage, only to realise later noise has also come to the party. And then all the plumbing with tubes and transformers…
Hopefully, the output side can be improved further…as you reckon
Adding this ‘no-dac’ filter in my opinion does no justice. I have seen the excited merriment at this no-dac camp before as to how natural sounding and unbridled sound stage, only to realise later noise has also come to the party. And then all the plumbing with tubes and transformers…
Last edited:
Acko,
Didn't I already say it doesn't sound like that? It sounds pretty fabulous, actually. No noise problem like you are thinking of.
So, let's just wait and let some other people try it. They won't exactly have the same filtering effects I have here, but maybe they will come up with something that works pretty well on their systems.
Didn't I already say it doesn't sound like that? It sounds pretty fabulous, actually. No noise problem like you are thinking of.
So, let's just wait and let some other people try it. They won't exactly have the same filtering effects I have here, but maybe they will come up with something that works pretty well on their systems.
@Markw4:
Hi Mark ... you may already know, or it may not be feasible in your context, but just FYI I am able to play back the EC7v2 algorithms at DSD128 on my laptop - with only a very rare glitch. It is a Dell Latitude 7490 with an I7-8650u processor and 16 GB of RAM + SSD HD. Benchmark for the CPU is (only) 6295. Bought it for USD260 one year ago second hand.
Jesper
Hi Mark ... you may already know, or it may not be feasible in your context, but just FYI I am able to play back the EC7v2 algorithms at DSD128 on my laptop - with only a very rare glitch. It is a Dell Latitude 7490 with an I7-8650u processor and 16 GB of RAM + SSD HD. Benchmark for the CPU is (only) 6295. Bought it for USD260 one year ago second hand.
Jesper
Last edited:
As alternative output arrangements have made it onto the agenda, perhaps it will be worth revisiting the 'CD Enhancer' arrangement that i originally mentioned back in post #26. I did put one together, using various parts I had to hand, when I had the loan of Marcel's prototype RTZ but didn't persevere to get it working. I still have the transformers to hand so could try again but not until I've assembled another RTZ board for myself.
http://www.itsonlyaudio.com/ultranalog/
http://www.itsonlyaudio.com/ultranalog/
As an update, we have done some more listening here. Find there are some tradeoffs between using the filter board as versus the passive filter. Therefore, some experiments are being lined up to see if certain changes the output board will help or hurt. Don't know, really could go either way. Also, as Hans has been suggesting, the filter board will get its own clean voltage regulators. That in itself may help some. Also, some more parts will arrive at some point to see if we can further improve clocking. Aim is to learn as much as we reasonably can from having the dac on loan here. However, could be be less frequent between reports going forward as some of these things may take some time. Have to see how it goes.
Present status of dac looks like this:
One end or the dac is tipped up for the u.fl cables to reach (which will be fixed later when the clocking is revised). Change this time was to run the filter board on clean +-15v power. As usual the only shared grounding is at the load (at the filter board in this case). This modification did change the sound in a way I like better. Vocal sibilance in some performers who tend to have more of it now sounds, for example, less like a lisp and more like normal 's' sound. Similarly cymbals now sound less like bursts of noise and more like real cymbals. Maybe some more comments on the change later after more people have listened.
EDIT: Now running analog output to line amp in SE mode, which sounds more clear. Balanced sounded more muffled and blurred in comparison. Would not call the difference subtle IMHO.
One end or the dac is tipped up for the u.fl cables to reach (which will be fixed later when the clocking is revised). Change this time was to run the filter board on clean +-15v power. As usual the only shared grounding is at the load (at the filter board in this case). This modification did change the sound in a way I like better. Vocal sibilance in some performers who tend to have more of it now sounds, for example, less like a lisp and more like normal 's' sound. Similarly cymbals now sound less like bursts of noise and more like real cymbals. Maybe some more comments on the change later after more people have listened.
EDIT: Now running analog output to line amp in SE mode, which sounds more clear. Balanced sounded more muffled and blurred in comparison. Would not call the difference subtle IMHO.
Last edited:
When this could help to get some more insight: in the attachment Left, Right and Clock are visible as generated by the Amanero at DSD512.
Data comes with some delay after the negative slope, and Marcels Firdac takes the data over with the upgoing slope.
Whether the data signal can disturb the upgoing slope of the clock signal at this point seems a bit unlikely.
Hans
Data comes with some delay after the negative slope, and Marcels Firdac takes the data over with the upgoing slope.
Whether the data signal can disturb the upgoing slope of the clock signal at this point seems a bit unlikely.
Hans
Turns out Andrea makes some accessory boards that can be used with his FIFO buffer board.
One accessory board that will soon be on its way here to try with Marcel's dac is an isolator/reclocker that will be located very close to Marcel'S dac so the cables can be very short and the signals very clean:
There is another board that can generate non-inverted and inverted phase signals to drive the dac shift registers directly, if desired.
Anyway, maybe in a week or so there will be some things we can try in the way of improved clocking.
One accessory board that will soon be on its way here to try with Marcel's dac is an isolator/reclocker that will be located very close to Marcel'S dac so the cables can be very short and the signals very clean:
There is another board that can generate non-inverted and inverted phase signals to drive the dac shift registers directly, if desired.
Anyway, maybe in a week or so there will be some things we can try in the way of improved clocking.
This was measured directly on the Firdac board where the Amanero’s signals are entering.
But Marcel makes valid point that because of capacitive coupling the data can push the clock up and down.
That could explain why a 010101.. pattern gives the same S/N independent of DSD rate, but that a real DSD512 signal with rather chaotic switching from between zero and one is so much worse as at lower rates.
Hans