Yes, ridicule is one deterrent to investigation but also lack of curiosity seems to be based on the assurance/reliance on stock measurements which generally use steady state test signals.
My premise is that this lack of curiosity is also the result of being on a forum - it pushes people towards wanting to be proven/be seen to be correct rather than admit they don't really know that answer. Hence simplistic tests such as measurements of steady state signals & ABX blind testing are relied upon because they deliver results which are "easy to digest" soundbites & good for winning debates rather than be involved in analysis of the complex area of what aspects of audio reproduction are important & at what level in our auditory perception.
The whole area of power supply & its interactions in the many devices that interconnect in the audio playback chain is a complex & interesting but widely ignored area, IMO
Right, some form of pack mentality maybe, whetever...
I have a WM8741 board from a few years back that is a clone of an Arcam dac 63 I believe, but without the lm4562 op amps, and some values in the output filter changed. I’m using Burson op amps now, and an ad744 for the servo.
After messing with mostly the power supply, and filter circuits for some four years on and off, with daily listening, I can say it’s sounds like I want it to. It feeds a headphone amp that shares the enclosure.
Here is a link to another, more simple take on the same board.
another EBay DAC-05 WM8741 - audio-talk
I have a WM8741 board from a few years back that is a clone of an Arcam dac 63 I believe, but without the lm4562 op amps, and some values in the output filter changed. I’m using Burson op amps now, and an ad744 for the servo.
After messing with mostly the power supply, and filter circuits for some four years on and off, with daily listening, I can say it’s sounds like I want it to. It feeds a headphone amp that shares the enclosure.
Here is a link to another, more simple take on the same board.
another EBay DAC-05 WM8741 - audio-talk
Over ten years old, original company bought by another company that probably had different interests.
I wouldn’t call it current.
To try and evaluate a dac chip using a cheap product isn’t going to reveal very much you know, as the power supply and output filter are going to be way more of the determining factors.
I wouldn’t call it current.
To try and evaluate a dac chip using a cheap product isn’t going to reveal very much you know, as the power supply and output filter are going to be way more of the determining factors.
because the majority of people are listening to the parameters of DAC, in their opinion the most modern 32bit overpriced DAC must sounds better, it just must
Oh..
That's sad :|
Right.
But on the other hand it's ready made, so who can resist..
But getting back to this again:
Maybe I should buy several pieces of the WM8741 and keep the aside,
so on the day I wish to do something with it, I wouldn't have a difficulty in getting them...
Yesbecause the majority of people are listening to the parameters of DAC, in their opinion the most modern 32bit overpriced DAC must sounds better, it just must
Many people listen to numbers, instead of the actual quality attributes
It's a very common problem indeed: all digital signals by definition have images. On top of that, sigma-delta modulators produce lots of ultrasonic noise.
An analogue filter requires capacitors or inductors or both. You can integrate inductors on a chip, but only with far too small values to be of any use to audio circuits. You can also integrate capacitors on a chip, but only with values up to a few hundreds of picofarads and those already cost a lot of chip area. Using a very limited amount of capacitance results in a limited dynamic range. For example, for a given cut-off frequency, a simple RC low-pass with very small C would need a very high R, which produces a large thermal noise voltage density.
All in all, there are DAC chips that have the complete filter built in, but it is often more practical to keep at least a part of it outside, especially for very high performance DACs.
I see.
This returns back to the point where it's better to keep many parts outside the chip,
to get better sound, rather than al all-integrated chip..
It always amazes me that ICs are no longer just Logic Gates inside,
but various kinds of electric and electronic components.
Regarding this again:
WM8741GEDS/V Cirrus Logic | Mouser
21$ a piece..
When searching in Google, it seems that only mouser holds them..
WM8741GEDS/V Cirrus Logic | Mouser
21$ a piece..
Last edited:
Because of this:
Or else, what will happen with them, is what happened with the TDA1541A
Or else, what will happen with them, is what happened with the TDA1541A
The WM8741 based DAC arrived.
I opened it to see its PCB,
and saw something weird on the "VIA VT1796A" USB chip on the left:
You see there something that looks like 2 soldering problems?
I then used a Macro len (1$ from eBay, extremely useful),
and took a much better picture of this chip and its legs,
and see this:
It really looks like a really scary soldering error..
that happened twice.
But when looking on the PCB traces that lead to these legs of the chip,
you can see that each trace like this connects to 2 legs, on both of these places that look like errors..
So I conclude that it's not a PCB soldering error,
and that the solder tin simply "climbed up", because that's how hot solder behaves,
like liquid, that attracts to any material that it can attract to.
(in this case the liquid is more picky: it's metal attracting to metal)
I will now test the DAC and see how it sounds..
I opened it to see its PCB,
and saw something weird on the "VIA VT1796A" USB chip on the left:
You see there something that looks like 2 soldering problems?
I then used a Macro len (1$ from eBay, extremely useful),
and took a much better picture of this chip and its legs,
and see this:
It really looks like a really scary soldering error..
that happened twice.
But when looking on the PCB traces that lead to these legs of the chip,
you can see that each trace like this connects to 2 legs, on both of these places that look like errors..
So I conclude that it's not a PCB soldering error,
and that the solder tin simply "climbed up", because that's how hot solder behaves,
like liquid, that attracts to any material that it can attract to.
(in this case the liquid is more picky: it's metal attracting to metal)
I will now test the DAC and see how it sounds..
Last edited:
OK, there are updates.
After listening to it for several hours,
here it is:
The WM8741 sounds amazing,
yet there is a problem here that also happened with several other DACs that I tried.
The problem is that this DAC emphasizes the edges of the Frequency Range,
meaning Bass and Treble.
The emphasis is too strong, it makes listening to music quite stressful,
and you need to keep the volume low due to this.
For comparison,
the SMSL M3 doesn't do this, you get a flat frequency range.
Also Lee's TDA1387 dac gives you a balanced frequency range, without any changes.
I assume that this problem is mostlikely due to the implementation of this DAC Product,
and not due to the DAC chip used.
1 nice thing regarding this implementation, is that the PCB hosts the TL072 and 2x NE5532P via a chip socket,
so anyone who would like to use other op-amps (which are leg compatible), can do it here easily, without any need for de-soldering.
I don't know If I could solve the Low+High frequencies emphasis via using other op-amps,
or if it's more like something caused by values of other components on the PCB (resistors, capacitors, etc, which are all soldered)
I am now back to my SMSL M3.
After listening to it for several hours,
here it is:
The WM8741 sounds amazing,
yet there is a problem here that also happened with several other DACs that I tried.
The problem is that this DAC emphasizes the edges of the Frequency Range,
meaning Bass and Treble.
The emphasis is too strong, it makes listening to music quite stressful,
and you need to keep the volume low due to this.
For comparison,
the SMSL M3 doesn't do this, you get a flat frequency range.
Also Lee's TDA1387 dac gives you a balanced frequency range, without any changes.
I assume that this problem is mostlikely due to the implementation of this DAC Product,
and not due to the DAC chip used.
1 nice thing regarding this implementation, is that the PCB hosts the TL072 and 2x NE5532P via a chip socket,
so anyone who would like to use other op-amps (which are leg compatible), can do it here easily, without any need for de-soldering.
I don't know If I could solve the Low+High frequencies emphasis via using other op-amps,
or if it's more like something caused by values of other components on the PCB (resistors, capacitors, etc, which are all soldered)
I am now back to my SMSL M3.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.