With this thread being about DAC recommendations and with DIY options giving choices of chips that do different bit depth processing and linearity from maybe 13 to possibly 21bits (?), Im keen to find out if my ears can benefit from better bit depths
Thanks for the tips for hearing the limits of a 16bit DAC. When you say "edge" is that like a harsh or hard timbral quality in music? If so would it be easier to hear it on acoustic recordings eg female vocals or with more complex music?
I guess its good to have an A/B test set up. Can you suggest a good comparator for exposing the hearing limits of a TDA1541a 16-bit DAC please?
Thanks for the tips for hearing the limits of a 16bit DAC. When you say "edge" is that like a harsh or hard timbral quality in music? If so would it be easier to hear it on acoustic recordings eg female vocals or with more complex music?
I guess its good to have an A/B test set up. Can you suggest a good comparator for exposing the hearing limits of a TDA1541a 16-bit DAC please?
Thanks for your help and pointing out the over simplifications in my guesstimates. Thanks for the links. Ive got lots to learn.
I hope your three wifes didn't leave you because of that like mine did ..i have lost a thousands of money searching the perfect dac during the last 20 years , ( Weiss , Wadia , Pass D1 , EAD , holo , Berkeley , Burl , T+A DSD , emprical , Lessloss , altman , TotalDAc , Metrum , EC design , JADIS , Chords , etc....) running in circle during years to finally accept that the sound i like is the sound of the 90' (Cello reference Dac , Wadia 9 , PAss D1 , EAD DSP 9000 ) so i end up and will die with an Old French PCM63 based Dac at 500$ and can"t be more happy . when it's good it"s good , to get better is an illusion costing time and money same for speakers , amps and women
Is this old French dac Audiomat by a chance ? I always wanted to hear it . I liked Weiss Dac quite a bit although it's a dac which will appeal to a " cake with a hole in the middle" type of people , Berkely left me cold , naim , Audio research , arcam , and many others which probably during controlled blind test I wouldn't be able to tell apart BUT after long and perilous process I restored AudioMeca Mephisto / Enkianthus combo , yeah all that oversampling crap I supposed to dislike and I think I don't need a better dac. Stumbled upon Meridian 508.24 delta sigma NO-No and I like it quite a bit too . I guess I'm approaching the age when basic Sony player from 90's would sound perfectly acceptable , well maybe not so fast because I tried Nakamichi MB1s and the focker was bright as hell.A couple of years ago, a blind ABX listening test for various kinds of dither and for additive noise was done on this forum:
https://www.diyaudio.com/community/threads/high-order-dither-listening-test.313257/post-5207912
@Mooly and @PMA were the test persons. The questions we tried to answer were:
1. Is there an audible difference between quantization with triangular dither and to one bit more with 11th order dither? (The RMS value of the error is the same in both cases, the probability distribution is not.)
2. Is there an audible difference between dithered quantization with triangular dither and pure additive noise with the same probability distribution as the total error of a triangularly dithered quantizer?
The quantization was fairly rough to make the differences more audible, 7 to 12 bit.
Mooly and PMA heard differences in both cases, even though the literature on dither usually implies that humans shouldn't hear any difference. They did not have a preference for additive noise over dithered coarse quantization, they disliked both equally much.
Unfortunately I made a stupid methodological error that made it impossible to draw hard conclusions: I failed to exclude the possibility that Mooly and PMA would also have heard a difference between two realizations of the same noise process.
Anyway, if you have or borrow a high resolution DAC, you could use a similar set-up as used in the thread I linked to to see if you can hear any difference between 24 bit, 16 bit without dither and 16 bit with triangular dither: process music files with an audio editor and use an ABX tool to compare them.
Last edited:
no my wife doesn't care , in fact i think she prefer me staying hours in the cave listening music or buildings amps than outside of her control areaI hope your three wifes didn't leave you because of that like mine did ..
yes Audiomat it is : Tempo 2 Ev with PCM63K add a T+A DAC8 DSD before , far better dac with volume control , discret DSD dac 512DSD upsampling with Hq player . spectacular 3D sounding ..i have sold it to friend who love it too , and i will finish my audiofool life with the old 16/44 machine because it is the sound that s make me happy nothing more ..
Besides the custom AK4499, the other dac I have setup here is a 'The Well Audio' DAC_Lite with FIFO_Lite and The Well Audio 11/12MHz clocks. Don't have a TDA1541a, but some people have compared the DAC_Lite to TDA1541a and DAC_Lite always seems to be preferred. Like TDA1541a, DAC_Lite is basically a non-oversampling R2R design with no output filter (so the output looks like stairsteps on a scope). https://www.thewellaudio.com/lite-series/
What I find is that without an output filter DAC_Lite is a little bright sounding. If I run it though a custom buffer which also smooths the stairsteps, it sounds a lot like the AK4499 dac except maybe a little less refined at reproducing higher frequency low level details (HF articulation, one might call it). Thing about DAC_Lite is that its two circuit boards (plus two clocks) and the rest is up to the dac builder. Power supplies, layout, shielding, etc., all has to be figured out. And all that stuff potentially affects the final sound, especially so for Vref power. For people who like that type of dac, there are some commercial dacs made by Holo in China that seem to be the hot thing to have. Haven't heard one myself yet, so no personal opinion.
Last edited:
Thanks. Nice DACs you have but Ill have to some up with some other DAC to test bit depth audibility for myself.
Actually I think testing 16bit versus 24bit using different DACs is not valid. Heaps of confounding variables with different DA architectures, IV's, outputs etc... eg when you heard edgy on a 16bit DAC it might not have been 16bits or dithering that you heard but eg bad power filtering on that DAC compared to your state of the art DAC. Those old TDA Phillips chips are renowned for a non-edgy sound so it may well be something else was going on for you? So for a valid test I think it needs to be done on just one 24bit DAC but fed 16bit depth files vs 24bit depth files. So just one experimental independent variable. Right?
I think what I will do to test 16bit versus 24 bit is:
- use one DAC that processes 24bits and has been measured multiple times, independently with high correlation
- get a 24bit test track confirmed to have 24bit depth
- DSP to 8bits / 16bits / 24bits +/- dithering
- then save them wrapped within 24bit WAV files
- use the Foobar ABX to do a blind trial
Any advice on methodological issues?
Seems like it should work. There are some freeware VST plugin bit meters you can use to verify bit-depth is as specified. For example: https://www.stillwellaudio.com/plugins/bitter/
Before doing all that however, there is more than one dither algorithm used for mastering CDs. You might find it interesting to see if you can tell a difference between any of them sighted first, just in case you decide based on that to use one or another for your ABX test. https://www.google.com/search?q=best+audio+dither+vst&rlz=1C1CHBF_enUS771US771&sxsrf=ALiCzsZmZtRnxaE7qFsz-zTKZkUQ3a6TBQ:1662912619254&ei=awgeY7WUD6P19AOOooeACA&ved=0ahUKEwj19dmfkI36AhWjOn0KHQ7RAYAQ4dUDCA4&uact=5&oq=best+audio+dither+vst&gs_lcp=Cgdnd3Mtd2l6EAMyBQghEKsCOgoIABBHENYEELADOgYIABAeEBY6BQgAEIYDOgUIIRCgAUoECEEYAEoECEYYAFDkLVibNmCMOGgEcAF4AIABaogB-QKSAQMzLjGYAQCgAQHIAQjAAQE&sclient=gws-wiz
Looks like there is some freeware that includes the following dither choices:
Dithers: Monitoring2, Monitoring, Dark, PaulWide, PaulDither, TPDFWide, TPDFDither, NotJustAnotherDither, Beam, TapeDither, SpatializeDither, VinylDither, DoublePaul, Ditherbox, BuildATPDF, NodeDither, StudioTan, DitherMeTimbers, RawTimbers, NaturalizeDither, HighGlossDither, DitherFloat
https://www.airwindows.com/vsts/
Last edited:
You know, I did try it with a coax output from my blue ray player. Some reviews, folks were getting no sound. I got sound, in both channels. I noticed that one woofer was doing a hula dance. I put in for a return and they sent another. Same issue. The USB input works fine.
I replaced the khadas temporarily with a 20 dollar Fiio dac, no issue. So it isn't the blue ray player. One channel is unstable, oscillating.
Russellc
Interested to know others experience with the Khadas coax input? I've read a few reviews where people said they got no sound, only noise, etc. My experience was music played but one channel was unstable, causing woofer to dance. A replacement was sent that behaves the same way. The USB input works great on both units.
Russellc
Russellc
I have and have used the KTB. I had no issues with it at all.
It was a stepping stone to me seeing if I could tell any difference from standard stuff and like yo say, great value for money.
I used the coax with no issues at all. It did what it was meant to do.
I am still very much a noob mind.
I have replaced it currently but I am sure it will have a place somewhere.
If nothing else, just to show people the benefit of removing one's DAC from the PC.
It was a stepping stone to me seeing if I could tell any difference from standard stuff and like yo say, great value for money.
I used the coax with no issues at all. It did what it was meant to do.
I am still very much a noob mind.
I have replaced it currently but I am sure it will have a place somewhere.
If nothing else, just to show people the benefit of removing one's DAC from the PC.
A couple of years ago I built a streamer using the Khadas Tone 1, driven by a Raspberry Pi 3B+ running piCorePlayer. I am pretty happy with the sound - I have been using it with my DIY "imitation" F7. So I can recommend it as a good low-cost DIY DAC option.
The display is a 3.5" DSI touch screen and the case was a Galaxy 2U chassis from Modushop, who machined the front panel for me. The power supply is linear and the PCB was based on DIYer Andy's (@iamwhoiam) D-Noizator over in the power supply forum. The front panel encoder is used to cycle through the music library and make selections. The back shows a WiFi dongle with an antenna and it has a wireless remote as well.
The display is a 3.5" DSI touch screen and the case was a Galaxy 2U chassis from Modushop, who machined the front panel for me. The power supply is linear and the PCB was based on DIYer Andy's (@iamwhoiam) D-Noizator over in the power supply forum. The front panel encoder is used to cycle through the music library and make selections. The back shows a WiFi dongle with an antenna and it has a wireless remote as well.
There are probably multiple reasons why one dac sounds different from another. So trying to answer to 'in what way' can be rather complicated. Maybe there are multiple ways which could be described.
How to express how something sounds by way of text over the internet continues to be be an ongoing communication problem.
How to express how something sounds by way of text over the internet continues to be be an ongoing communication problem.
Last edited:
In 2020, I think the Tone 1 was the best bang for the buck for a bare board DAC, which is why I used it. Today there may be better-sounding DACs at that price point in the ESS/USB category, but I doubt you can do a lot better. If you care about measurements, check out Amir's review in ASR here.
I can say, while it hard to call "winner" I find for esb, my Toneboard is getting way more time than Audioquest Red, which uses ESS chip if I recall.
For a C note, Toneboard better than it should be, but at the end of the day you cant expect everything from a 100 dollar DAC.
I searched info for Mira, found an older closed group buy, is it available somewhere else?
Russellc
For a C note, Toneboard better than it should be, but at the end of the day you cant expect everything from a 100 dollar DAC.
I searched info for Mira, found an older closed group buy, is it available somewhere else?
Russellc