iBasso DX300 Qualcomm Snapdragon 660 Octa-core 6GB RAM ******NEW Firmware 2.00 Android 11******
Dec 25, 2020 at 9:13 PM Post #856 of 14,596
Well, everything matter, especially timing and phase noises. If timing errors occurs too much, and noises would results in too much errors, then the FPGA and it algorithms can only do guessing to reconstruct the info received. That means, it is guessing based on whoever coded the program......it isn’t original anymore.

In a simpler example, if FPGA can add 10% accuracy, then if the clocks is 89% accurate, we will have 99% total. If the clocks are some cheap toys, and is only 29% accurate, then we have 39% total.

It isn’t that simple, but I tried to explain it in a simpler way lol
 
Dec 25, 2020 at 9:16 PM Post #857 of 14,596
But what difference is it going to make if they use Xilinx (soon to be acquired by AMD) or Altera (already acquired by Intel) or any other FPGA brand? :D It is just the idea of them implementing FPGA to off-load SoC. FPGA is just a programmable digital logic blank chip you have to code with a specific functionality, and we don't even know if it is done in house or outsourced per iBasso spec. Those are not off the shelf chips with built-in audio processing functionality.

Nice stock tip to know about the M&A of well known chips for my stock portfolios. Lol

Ibasso is marketing them as “FPGA Master” as if it’s somehow better. They do after all do all the audio functionality before it hits the analog board. Only what transistors they have and how fast they processor but no biggie.



ell, everything matter, especially timing and phase noises. If timing errors occurs too much, and noises would results in too much errors, then the FPGA and it algorithms can only do guessing to reconstruct the info received. That means, it is guessing based on whoever coded the program......it isn’t original anymore.

In a simpler example, if FPGA can add 10% accuracy, then if the clocks is 89% accurate, we will have 99% total. If the clocks are some cheap toys, and is only 29% accurate, then we have 39% total.

It isn’t that simple, but I tried to explain it in a simpler way lol

Yep! The LDO that powers them also contributes to noise as well. Why would they market that?
 
Last edited:
Dec 25, 2020 at 9:54 PM Post #858 of 14,596
If the clocks are some cheap toys, and is only 29% accurate, then we have 39% total.

😛 clock that are cheap toys
D47CC0B9-A53F-4375-A668-A0B23F5B33CD.jpeg
 
Dec 25, 2020 at 10:15 PM Post #859 of 14,596
OK, I just got back from the sales rep.

The distributors in the US should receive the first batch of items in mid-Jan.

No update for EU market yet, but maybe people can consider buying from US distributors and ship it to EU?
 
Last edited:
Dec 25, 2020 at 10:25 PM Post #860 of 14,596
The sales rep Jannie just gave me two more pics of DX300 to share. :smirk:
 

Attachments

  • D05A67C2-5C4D-45B9-A010-5B2A5FDE9E0A.jpeg
    D05A67C2-5C4D-45B9-A010-5B2A5FDE9E0A.jpeg
    3.1 MB · Views: 0
  • A78996E2-9465-436F-9F54-57C70D0A62DA.jpeg
    A78996E2-9465-436F-9F54-57C70D0A62DA.jpeg
    3.2 MB · Views: 0
Dec 25, 2020 at 10:52 PM Post #862 of 14,596
The sales rep Jannie just gave me two more pics of DX300 to share. :smirk:


Jannie is being nice thank you :) and shes nice too :stuck_out_tongue: :ksc75smile: :heart_eyes: That Dx300 definitely is on a bigger side of daps
 
Dec 26, 2020 at 12:41 AM Post #863 of 14,596
It is the LED status indicator. During playback an audio file, it will change depending on the file type (i.e. mp3, flac, dsd, etc.). Not only that, during charging battery and the remaining battery life status as well. How does this work? It will be explained in the DX300 manual. :slight_smile:

Tonight @ 2100 Japan time zone, further discussions
Musin / DX300

Feed for queries
Query form google docs
 
Dec 26, 2020 at 12:45 AM Post #864 of 14,596
Size feels like a 20,000mah Power bank. But that is alright in my bag.

Quad DAC for more power or resolution?

The quad dac is simply because Cirrus Logic chips don't handle as many channels as the competitors do in a single chip. Solution is to use more CS chips.

So not a more is better meme, but needed. Though I am sure the PR department is having a blast.
 
Last edited:
Dec 26, 2020 at 2:48 AM Post #865 of 14,596
OK, let me explain my understanding of how "FPGA Master" makes the difference.

Since DX80, the first native DSD-capable iBasso DAP, an XMOS chip was used to stream sound data to a DAC, but only for PCM higher than 192kHz and DSD. As I understand, "FPGA Master" performs the same task (probably among some others), but does it always.

One note: when such an iBasso DAP works as USB DAC, that XMOS chip handles all formats. Actually, in USB DAC mode, the SoC is not involved at all. So, having any iBasso DAP, you can test the difference right now: just reproduce the same file in PCM format, not higher than 192kHz sample rate, via Mango player and from a PC using the DAP as USB DAC. This is the difference.

Another way to hear the same material with and without XMOS is to upscale a PCM file with sample rate not over 192kHz into 384kHz (352kHz for 44.1kHz source) and compare them. But USB DAC way is better: it reduces loaded SoC impact on the power supply.

Why iBasso didn't use XMOS always? Because it consumes the power. Probably, with FPGA tailored for their needs, power consumption became reasonable enough to have it turned on always.

My firmware (and add-ons) for full Android DAPs have USB Audio app for years, which exactly allows to use XMOS chip as the internally connected USB DAC. There are people who used (and still are using) it as such, tolerating faster battery discharge for the sound difference. This is one more way to try what "FPGA Master" may bring.

Also, while I'm writing, I'll tell you guys my opinion on what DX300 is:

Since May 2014, I was working hard, trying to squeeze the best from iBasso DAPs hardware, mostly by minimizing impact on power supply, and then by utilizing XMOS for all formats. With DX300, I see nothing how sound can be affected by firmware! Separate power supply for analog path (I hope it includes DAC analog power inputs as well!) and FPGA Master for all formats make it independent on what the SoC does.

Of course, there are conspiracy theory believers and psychoacoustics ignorers who would "hear the difference" anyway, as always, even when there is no a difference. For all the normal music and high quality sound lovers, DX300 is the device which you can like, or don't like, but there is no way to make it sound better or worse by a firmware, neither official nor custom.
 
Dec 26, 2020 at 3:04 AM Post #867 of 14,596
Nice!!! So @Lurker0 is here!!! Can’t wait to hear the “modified firmware” with the possibilities to “improve the sound performances” :)

The Dx300 is large ? Or Jennie is small ? Which one ? Lol
If you read his post, he says you will not be able to improve the DX300 with firmware.
 
Dec 26, 2020 at 3:28 AM Post #868 of 14,596
Size feels like a 20,000mah Power bank. But that is alright in my bag.



The quad dac is simply because Cirrus Logic chips don't handle as many channels as the competitors do in a single chip. Solution is to use more CS chips.

So not a more is better meme, but needed. Though I am sure the PR department is having a blast.
Sounds like an old LG phone to me. Definitely will listen with my ears before purchasing....
 
Dec 26, 2020 at 3:52 AM Post #869 of 14,596
If you read his post, he says you will not be able to improve the DX300 with firmware.

But getting the PlayStore and removing Bloatware, should still be a thing.
 
Dec 26, 2020 at 4:04 AM Post #870 of 14,596
If you read his post, he says you will not be able to improve the DX300 with firmware.
I read it, he just has an “iconic move” that states “possibly improving” :wink: and that usually come with when he mentioned “psychoacoustics arrogant” who believes firmwares improve sound. He always put out that description “possibly improves” LOL.

Regarding FPGA and error corrections, clock reconstructions, in R2R devices, firmware can alter the sound performances. That is based on the parameters of the corrections and guessing of the algorithms. Regardless of what you do, in digital worlds, as long as there are “errors”, there are ways to slightly altering the result, take picture and digital filter for an example. Simply put , errors is everywhere in the universe, be it digital or analog, and as long as there are errors, there are ways to change the outcome.
In analog paths, the tuning with different values, components, amplifications feedbacks can totally alternate the sound. But then again, there also are people who believe all DAC and Amp sound the same :wink:

So, you see why I am one of those guy that “everything can slightly altering or improving or degrading” the sound ?

Honestly, regardless of the outcome, Ibasso devices and the modifications group of people is super fun to be around :)!!!

How much differences does Ape DNA differ from Human DNA ? And how does that distinct between us ? 😂
 
Last edited:

Users who are viewing this thread

Back
Top