Snapdragon SoC 410E and 600E


It could, but what does this have to do with the Pyra?
 
It's 100% pyra related if ED can make a CPU board for it?

Don't trash this thread already.
 
It's 100% pyra related if ED can make a CPU board for it?

Don't trash this thread already.
Well that's the thing, there won't be any plans for a new CPU board if the current units don't sell. Right now he will need to sell just slightly more than twice what he has already sold to break even. Don't expect an upgrade for a few more years.
 
Let's hear that from the dragons mouth.... A forum moderations word doesn't really count :D

He'll likely agree with your post, but you know... I'll throw money for the upgrade, if he can order small batches
 
Let's hear that from the dragons mouth.... A forum moderations word doesn't really count :D

He'll likely agree with your post, but you know... I'll throw money for the upgrade, if he can order small batches
Even if MOQ would be a non-issue, you still need to spend a lot of time and money to design a new CPU board I believe...
 
That's more like a downgrade, Cortex-A53 is close to a Cortex-A7 in performance and it's only 1.2 GHz, so it'd be going back to pandora level of single core performance.

The open GPU drivers is a nice thing to have though...
 
Well I guess I'll move it back to the Pyra General Chat as now it seems Pyra specific in discussion.
 
Ah, an A53, not an A57...okay.

Hm, might still be interesting. It's four times a Pandora, needs less power and has LTE already included... so it would bring down costs of an LTE unit by 110 EUR approximately. Well, the design would still cost something...
 
I'd go for this as an upgrade. How will a mobile version Pyra handle two LTE thingies though?
 
Don't get too much excited by more cores/powerful cpu.
As an example, the odroid-xu4 ships a 4xA-15 2GHz + 4xA-7 but you can't use full power because even the active cooler + heatsink can't handle the temperature, then the system crash.
So there's no need having a high number of cores and/or more MHz if you can't use them.
 
and has LTE already included... so it would bring down costs of an LTE unit by 110 EUR approximately. Well, the design would still cost something...
Not only the design. The real challenge is the full certification of the radio system which must include the firmware. And every time you change the firmware it must be re-certified. And, FSF would complain about another binary blob because we are not allowed to publish the firmware source. Expect several million EUR per design...
This is why pre-certified modules are cheaper in total because we only have to certify the integration into the whole system and not the module or its built-in firmware.
So this LTE feature of Qualcomm chips is fundamentally unavailable to us.
 
Well.. for those of us whom don't care about LTE, hopefully another chip can be used without those hassles :) but nice to know.

If you decide to remove the hardware screen rotator.. you'll need to recode the 'software' rotate code for every SOC?

Back to watching the clock for somebody to release an updated friendly SOC :)
 
Back
Top