Perhaps someone will set up a controlled blind test once this covid stuff is over. Should be fun, especially if users that are given time to become familiar with each component for a few weeks before they participate in the blind test. Would be interesting to see if people can pick out the differences they highlight during their subjective evaluation period.
The DAC3 came out in late 2016 / early 2017 as far as I can tell FWIW. I realize they have been using that part since the DAC2.
I think you are right. Nonetheless, it was very good at the time it was introduced. D90 changed things though.
I guess its okay to say in an AK4499 thread that D90 seems a bargain at its price point. CD audio is better than DAC-3, but not perfect. DSD256 seems a strong point, CD audio near its best, assuming use of software processing.
Last edited:
Can you explain to me how the SRC4392 is supposed to sound bad when not powered correctly? I'd imagine you'd have to do something monumentally stupid with the power supply to have any impact on its performance.
Yeah, I don't really have too much of a problem with the high prices. I think given the amount of work put in, the prices that Bruno asks for the Mola Mola gear isn't crazy. Whether it's worth that is another question.
Honestly, it is pretty hard to compete with Topping given their pricing and what you get.
Hmm, perhaps we should give D90 a listen before tweaking the SU-9, at least to be comforted in what direction to go and how much ground has still to be covered?
I am afraid this D90 DAC stands sadly in no relation pricewise with what my friend wants to spend to remain consistant with the rest of his system, but at least to give it an ear...
Thanks for all this
Claude
I am afraid this D90 DAC stands sadly in no relation pricewise with what my friend wants to spend to remain consistant with the rest of his system, but at least to give it an ear...
Thanks for all this
Claude
You are obviously a stone-deaf person that relies entirely on sighted engineering, listening to music on the logic analyzer screen.
I listened to D90 for a few days to see if the sound will grow on me. During that time I compared it to DDDAC (NOS, no digital filters, analog-out straight off the I/V resisitros). I also have very revealing 2-gain stages Aleph J.
My impression of D90 was not positive. It sounded cold, processed, unpleasant. The positives were a lot of details (but details that are being kind of forced onto you...). It also had good separation of instruments, but again... it just seemed over-processed and pushed hard out of its analog-out. DDDAC actually had more details that were presented in a very natural, relaxed manner for you to enjoy. D90 seemed like it was pushing everything towards the listener; that was all too hard to cope for longer than 10-15min. On quick A-B it wasn't that bad, but when I tried to do something else (write/read/talk to my family), D90 was too tiring... I had to lover the volume and/or just turn it OFF.
I am just wondering if you did anything to D90 to make it sound more natural? Maybe get rid off OP amps' analog stage and do something less harmful?
Hello i had similar experiênce here. Had a d90 from a friend and compared to miros360 ad1862 nos dac and i prefered the ad1862. The sound of d90 lacks some emotional content....
But has always this is subjective.
D90 takes about 3-days of continuous power-on time for the clocks to fully warm up, so a bit of patience may be needed. It does have very good bass, and good L/R imaging. Where it starts to sound less good is unfortunately at vocal frequencies and above that. At those frequencies there is bit of graininess (presumably partly from clock jitter - the Accusilicon clocks are divided by 2 in a CPLD for most playback modes). At enough SPL that bit of graininess tends to be more well masked. Also, the best clocking is only with USB. There is no ASRC or FIFO for SPDIF/TOSLINK, only an AK4118 recovered MCLK.
For people that like 'high end' audio sound, its not that type of dac. For people that listen for and or are sensitive to low level IMD, it is pretty good. In any case, its best at DSD256. Also it is designed to appeal to people who like what they do over at ASR, and right now that is important to have for mass-marketed devices.
For people that like 'high end' audio sound, its not that type of dac. For people that listen for and or are sensitive to low level IMD, it is pretty good. In any case, its best at DSD256. Also it is designed to appeal to people who like what they do over at ASR, and right now that is important to have for mass-marketed devices.
Last edited:
Thanks, Mark. I think you summed it up really nicely. It does have its audience that's for sure.
It is possible that after 3 days, the internal air (inside D90) has warmed sufficiently enough so that there are no air currents any more, and the crystal is evenly heated from that moment onwards. The temperature equilibrium, in turn, would result in a reduction of phase noise below 0.1Hz which is critical for the perception of what we humans call natural and unfatiguing sound.
According to Andrea Mori who has the equipment to measure clock jitter, it can take SOA clocks weeks of power-on time to to fully settle. It isn't just Accusilion clocks in a closed box. IME Crystek 957 clocks in either open air or pretty shielded also take about 3-days to audibly settle. That said, my observations have mostly been made using AK4499 which may make a difference since different dac architectures are known to have different sensitivities to jitter.
Last edited:
I do not want to shatter your self confidence, but Santa and Bigfoot do not exist.
Would be good if someone could send one to purrin:
DAC Blind Test Series | Super Best Audio Friends
Purrin’s setup could even be used to check how valid some of the subjective claims in this thread are.
Andrea is obsessed with a solution to a problem that doesn't need further solving, so what else would you expect? It's funny, I just checked the manual for the 8 GHz Tek spectrum analyzer next to me and it doesn't require you to warm up the unit for 3 days to achieve spec. Same for the 40 GS/s scope with jitter analysis built-in.
Last edited:
Guys...
It is a well understood fact that high precision clocks require settling time to reach their best phase noise performance, this is NOT debatable, it is an established fact proven by Phase Noise measurements.
We can debate all we want about to what degree close in phase noise may be audible, or not, and such audibility will be system dependent, but there is no debate regarding the fact that clock performance does improve with (lower close in phase noise) with a certain amount of settling time.
It is a well understood fact that high precision clocks require settling time to reach their best phase noise performance, this is NOT debatable, it is an established fact proven by Phase Noise measurements.
We can debate all we want about to what degree close in phase noise may be audible, or not, and such audibility will be system dependent, but there is no debate regarding the fact that clock performance does improve with (lower close in phase noise) with a certain amount of settling time.