UT Dream hardware
-
@3arn0wl iMX make some good products but there are a lot of problems too.
I'm working with iMX right now and we have many issues with PCIe server as well as real time issues...
But sure you can find many hardware to fit our geek needs, but to create a phone is more than just specs and software.
The work done by Pine64 is great and integrating all those parts is not an easy task cf. the problems encountered by Librem.While picking a component off the shelf seems a good way to go it can be problematic while integrating the hardware and sometimes a total waste when trying to port the software.
[Edit] About NXP iMX6 we also had trouble with the different RAMs used and eMMCs, the components are seen as 100% compatible but really are not there are slight differences which made on batch OK while a newer purchase didn't work with our product...
I'm not an expert, but it appears that I'm working with NXP's product for few months now and I see the limitations
I also see the pros of using an integrated board so I appreciate even more the work done for the Librem5 and the PinePhone. -
Thanks for that, @AppLee -
@AppLee hello
i'm asking to learn ,
isn't the eMMC much slower than UFS ? and using faster storage result in better performance ? -
Hi @stan
You're probably right, but in my project the OS is fully loaded in RAM so no performance issue.I don't know about UT but I guess it's something similar you load the system in RAM and try to limit disk access.
UFS was designed for cameras so it need to write a 40Mpx picture fast enough and it should also be able to write video stream...Another aspect is the price tag. Which I don't know about...
-
So maybe let's flip the question around. How much would we be willing to pay for a dream UT device?
Taking into consideration that high-end specs will bring with them high-end prices, but will not automagically mean that the software running on it will run smoothly, cf all the work needed to get the software optimized for PinePhone and Librem 5.
Also, a higher price point would mean a much bigger investment upfront for the company producing the hardware, if they were to partner with the community to produce the software like Pine64 did. Multiple devkits and prototypes were sent to developers.
If the company decided not to be so generous, then they would be forced to hire devs to work on the software for them, because the price point would be a barrier for many in the community to get involved early on.
-
@arubislander
Well... What I'm really desperate to do is retire my 2007 laptop before it gives up the ghost entirely!
So yes... A truly converging phone would be my laptop replacement, and yes, I would be prepared to pay good laptop prices for the use of that. -
@arubislander I agree with you on that. My Xx shows how smooth an OS can run on nearly 4yrs old hardware. Therefore, pine64 may be is not that for away from a consumer model with fits the need bigger user group.
-
The Reason i asked my question was to learn more about what makes the hardware more compatible, and what we need from the manufactures.
like a clear specs if handed to manufacturer can build and hand it to us. so we can can easily use UT on it -
@stan The answer to that can be pretty simple, we need FOSS drivers for their parts.
When you plug a new piece of hardware to your computer you need drivers to access it (Printer, mouse, keyboard, scanner, etc.).
FOSS drivers allow us to talk to the hardware while closed source drivers makes it really difficult. -
@AppLee so we need high spec hardware with foss drivers.
it's not easy i guess other wise it would have been done many times -
@stan it is easy enough. After all, the drivers get written. So the software exists. What is needed is for the source code to be open sourced and made accessible to study and use.
-
@arubislander i have a
lot to lean
thank you for the help
