[immortalsoul]

I don't get it why people need powerful cpu, we are not buying a smartphone, we are buying a music player. The focus needs to be on sound quality, as for price I believe a 4000$ dap would scare a lot of customers since a lot of good daps are now on the market for less

 

[LuckyPantsu]

I don't get it why people need powerful cpu, we are not buying a smartphone, we are buying a music player. The focus needs to be on sound quality, as for price I believe a 4000$ dap would scare a lot of customers since a lot of good daps are now on the market for less

Indeed. I don’t understand why it should be monster in performance? Maybe only to make android work normally without lags.
That is why the conception of N8, LPGT, Sony wm1z with not the fastest CPU but with good stable software is much closer to me. In that case - money goes to the sound, I want to believe in that .Why should we have the second device with a high performance in a pocket if we have a smartphone for this purpose.

 

[Shawnb]

Sound should always come first but you also want or need the performance to switch between music/folders quickly. Basically you need the performance for everything besides the music.

So I'll take as much performance they can cram in there as long as the sound comes first

 

[brams]

I consider myself a "sound above all" audio purist, but had to accept that times have changed. Like it or not the world has moved to streaming so regardless of the reasons for using Android based MCUs any TOTL dap that does not support streaming will be a very tiny niche within an already small market and will likely never sell well enough to justify its existence. The ability to explore new music while also listening to local or network stored favorites is the best of both worlds. Even better if I can download those favorites directly to my DAP.

For example, I really like the audio performance of my Questyle QPM. However, although it has the ability to stream by a wired connection from a phone, I would gladly give up some of its amazing performance for two things: Onboard wifi streaming and system Parametric EQ. Although it sounds good on paper, for reasons that I don't fully understand, the need to connect a phone for streaming is not that appealing in practice. As far as EQ is concerned, all iems and headphones have flaws that can be minimized by a little tweaking if done correctly. Given this, I have always wondered why DAP manufacturers provide Graphic EQ which is usually useful only for rough corrections and only if the pre-chosen filter points just happen to largely match up with the frequencies that require adjustment.

I think the days when Android was a real limitation to ultimate performance are long gone. I'm not saying there is no performance hit at all, but instead that any audio impact is so minimal that it's largely obscured by the DAP "tuning" and other factors in the playback chain.

 

[vkenz]

conclusion, no android no life.

 

[davidmolliere]

I don't get it why people need powerful cpu, we are not buying a smartphone, we are buying a music player. The focus needs to be on sound quality, as for price I believe a 4000$ dap would scare a lot of customers since a lot of good daps are now on the market for less

True but on the flipside a decently fast SoC like Snapdragon 660 is a big leap in user experience compared to 430, lag time, freezes are a thing of the past now and I can't really go back. But sure a DAP doesn't need the latest chip for smartphone. Relatively 660 is an "old" chip but sufficient.

 

[Whitigir]

idk, it would be great if the MCU is powerful for a high end DAPs, which may stand the test of time over 6-7 years right ? Because we are talking about streaming and android ..... or, just be like Sony Walkman, using it own OS, no streamings, but that will never be out of dates LOL....

 

[Andykong]

Allow me to crunch some number Andy, Snapdragon 850 is merely $299-$399 each (One plus 7T). Now, considering a niche DAP is $3,999 which is 10x higher. So, if you are saying that 100k quantity minimum, that can easily be done with that price for 10K devices sold isn’t it ? Now, you only need to sell 2,000 units of N8II each year over 5 years period. However, I am sure you can sell more than that , be the first of it kind to be so advanced, that others looking jealousy at it

Hello Andy

the point you are making is very interesting and I'm curious to elaborate on it.

My question related to the "bolded" statement above fundamentally is "why" ?

Is it an economic issue? Hope not.
I would understand the approach when assuming to start a project for a lowend / lowbudget device: outdated hw is a monumental opportunity to compress costs and therefor prices and that's key on a project like that.
On the opposite end a highend device is targeted to the higher tier of the market, both technically and economically speaking.
How much more does state of the art hw cost to you, compared to outdated hw? Just tossing fake numbers here: 150$ vs 30$ ? So a net aggravation of 120$ per device.
Add you industrial markup, let's say 400%. That would mean that a "NX" device based on state of the art hw would need to be priced in the high 3K$ instead of the low 3K$. I doubt this would make much or even any difference on that device's target population. Am I wrong?

Is it maybe a technical issue? Big hw manufacturers are unwilling to support non-mainstream projects on state of the art models? (you would be getting too few pieces, and they would not dedicate tech support people to help you from the backend on modifications etc?)


Cheers,
Alberto

SD 850 would not be a good choice. Better go for a more recent mid-tier 700 or 600 series. In a dap you don't need the peak performance you need in a smartphone. Instead, you need an efficient cpu for long use. Think of marathon vs sprint. In a phone you need peak preformance to react to user input. In a dap you need to decode and process audio for several hours. A SD850 would overheat and throttle back considerably.
Also, $299-399 is a lot, even if the dap costs 4000. You are simply looking at the bills of materials, but the manufacturer has to recover the RND costs, not just materials.
Besides, if they price this at 4000 they are not going to sell many, particularly considering the price of SP2000T.

I hope we can stop here because:

• Choice of MCU is finalised at the very early stage of project development. Given the N8ii takes 2 years to complete R&D, this is a done deal at 12-18 month ago, We can't afford to change the MCU three month before product launch, or even 6 month before launch. Nobody can afford that.
• Please don't suggest as if we have freedom of choice but we denied because we want to cut cost or minimize work load. If you have experienced to approach Snapdragon or Samsung as a "new" customer and convince them DAP is an option that should consider, you'll know what I mean. The semi-conductor industry has a very tight entry barrier, we can only conduct business through sub-vendors, and availability of MCU to sub-vendors is a down-stream business model.
• Between sub-vendor and manufacturer such as Cayin, we have another layer of "solution provider". This is how we deliver important customisation such as DTA (Direct Transport Audi that allow us to bypass Android SRC with all third party applications), or to perform necessary DSP/DDA between MCU and DAC chipset, to enhance digital audio quality before decoding. The size of DAP industry is not attractive to Snapdragon and Samsung, they won't provide a solution to us. In fact, they don't even want to provide support to our solution provider directly, definitely not on the current models at the time of our development.
• Semiconductor business didn't look at the unit price of your product as a consider for business opportunity. They prefer lowend / lowbudget devices with a large quantity than high-end device with a low quantity. If they want to pick a client between a $99 budget mobile device with 10 million units over complete product life-time vs $1000 "high-end" device with 50K units over complete product life time, their preference is very straight forward.
• I believe Whitigir was making a joke with me only. For example, the One Plus 7T was announced all the way back in 2019 and the price was $599, but he use EOL discounted price of One Plus 7T to compare with SRP at time of launch of N8. He also 100K minimum quantity as a comparison benchmark, not the actual sales volume. Are we sure One Plus will commit to such a project if they can only sell 100K units, at $399 each? We have been teasing each other in many posts previously and had a lot of fun, so please don't take his conclusion seriously and develop more theory over that.

So lets leave this topic and continue to enjoy our music.

 

[Skullar]

My god! he unintentionally disclosed N8II price! Now they cannot dare to change it!!

$1000 "high-end" device with 50K units over complete product life time, their preference is very straight forward.

 

[Andykong]

My god! he unintentionally disclosed N8II price! Now they cannot dare to change it!!

Glad you have a short moment of happiness from my illustration. By the way, our "last" Android DAP was happen to be in the $1000 price range.

 

[MarkParity]

Glad you have a short moment of happiness from my illustration. By the way, our "last" Android DAP was happen to be in the $1000 price range.

And a truly outstanding effort it was.

 

[kleimart]

Is there a chance that Cayin will use a R2R architecture in the N8ii?

 

[blotmouse]

Is there a chance that Cayin will use a R2R architecture in the N8ii?

No chance. Dac already specified on 1st page.

 

[ThanatosVI]

Is there a chance that Cayin will use a R2R architecture in the N8ii?

Doesn't mean that they aren't working on another DAP with R2R and tubes for a later release date.
The N8ii project initiated even before the AKM incident, so it's very well possible that there is something that will utilize R2D in development already

 

[LuckyPantsu]

Doesn't mean that they aren't working on another DAP with R2R and tubes for a later release date.
The N8ii project initiated even before the AKM incident, so it's very well possible that there is something that will utilize R2D in development already

So we need a separate topic for n8 mk3 with R2R+ tubes