iBasso D10. .UPDATES 1st page, with Current Opamp Choices by HiFlight . . . images page 1, 12,... - Page 51

But what I don't understand is: if the player can send out digital signal without be converted (skipping the internal DAC) and can read the file 24bit192Khz shouldn't the signal exit as it is read?

That question seems to be moot if RockBox downsamples to 16/44 before passing it on to anything else. I guess RockBox has decided on this approach as there are very few players that can send digital data out and I can't think of any player that has a buildin dac that handles better than 16/44. An even if the digital data can be send out I suppose as iRiverdude states we don't know if the optical output on the iRiver players can handle it (I have no technical understanding of what goes on in the output stream) and also we don't know what the dac in the iBassos are capable of accepting as input. I just got carried away by the fact that the 24/192 files played and didn't think further

So it's in part a problem of the software, but which is the purpose (or the benefit) of the software to be able to read 24bit192khz if then it's all downsampled to 16bit44khz?

The benefit is you can play it back at CD quality.

And if/when sometimes in the future a device comes that can play at full resolution you already have the software ready

Just worth noting: A 16/44 FLAC file will always sound better than a 320kb AAC file of the same recording. These are two different codecs, one of which is lossy.

A better test would be to take a 24/96 FLAC file, transcode a copy to 16/44 (still in FLAC, just lower bitrate and sampling rate), then compare both files on the same player to see which sounds better. Foobar2000 is good for transcoding. It's also worth comparing these results using a different digital source (ie: a MacBook Pro or something similar). I don't know if the D10 can handle 24/96 through USB, but I DO know that the D1 is limited to 16/48 on USB, but can reach 24/192 via SPDIF. You would need something like Foobar2000 and ASIO4ALL on Windows to be able to reproduce this. Don't know how to set that on OSX, and I don't think Linux has much capability for it yet.

It's worth noting as well that I've tried some 24/192 files through TOSLINK to my D1 using Foobar2000 and ASIO4ALL, and the difference was subtle at best. And for my case, I prefer the sound of my Monica 3 Non-oversampling 16-bit DAC, though that's dedicated and more expensive. *shrug*.

I'm curious to hear how the D10 stacks up against the D1, though.

I ordered the D10 last week, it left Hong Kong Tuesday, and arrived yesterday. Amazing.

I really appreciate all the little extras they supply with it. The 120v to USB charger plug will come in handy for my iPod.

Even without any break-in, the DAC section is flat out incredible. I've never heard my E500s sound so spacious. The HD650 sound great, though they could use a bit more current. I get a little bit of noise via USB, but that could be the computer, not the D10.

I'm hearing more detail than I do from my Benchmark DAC-1! How is that possible? I just don't get it. In any case, I'm very impressed.

The comparison that to me seems to be the most important is between the 24/192 downsampled by RockBox and the same recording in the 16/44 version tha Linn also offers to decide whether the RockBox downsampler actually degrades the sound below a native 16/44 version but I have my doubts as to whether I have good enough ears to detect that.

As it is now, the sound quality my 320kb MP3 files fed via iRiver optical output thru my D10 optical input which in turns feeds my Stax006t via the D10 Aux output exceeds the ability of my ears to resolve anything further.

My equipment keeps improving while my ears don't! I wonder if more burn-in would help them! :>)

How does downsampling work?

Is it an "down" re-encoding via software?

Now I don't know this so it is pure conjecture. RockBox seems to work on 32 bit sampling rate and at 192 Khz internally so my guess would be that since this procuces many more samples than what is needed in 16/44 som sort of average of a number of samples is taken to produce enough samples for the 16/44 soundstream and thereby loosing some resolution.

I don't think that my ears would feel the difference anyway I'm curious to try listening of 24bit192khz files!

In a nutshell, its like doing an analog to digital conversion (ADC) except its a digital to digital conversion. Rockbox can take up to a 32 bit sample and bring it down to 16 bits. So it is throwing away the least significant 8 bits in a 24 bit sample at 44kHz rate. And this is being done by software in real-time, meaning as you are listening to the music it is doing this conversion.

Understood, thanks a lot!

You might need Rockbox to handle 24/96 (mine has it).

I know when the Macbook down-samples the 24/96 to 16/44 that they sound almost identical, and it takes a lot of time and concentration to pin down the difference. I haven't bothered to see how well the H140 does, and I haven't tried 24/192 since all my optical DACs are limited to 24/96 while their coax can do 24/192. I only have one readily accessible 24/192 coax source, my Akai portable DVD player, and it doesn't get much use since I don't have many DVD-Audio.

I'm pretty sure the D10 coax can do 24/192, just don't know about the optical. I can set my Macbook to output 24/96 optical and the DAC accepts it fine, so we know at least 24/96 optical. The D1 however would choke on any optical more than 24/48, and just make noise.

320kb MP3 wouldn't be my optimum test data rate, because with the right combination of gear I can hear the difference in 320K and lossless, although 320K sounds fantastic and is what I often use as well for entertainment purposes. But, when I'm testing gear and looking to see just how far I can hear into the music, lossless is the way to go for me.