Discussion in 'Pyra News' started by EvilDragon, Sep 15, 2018.
okay so let hope for arm-64bit with more cores an more quickness per core
Yeah, that sounds nice! But it will probably be a few years.
1. Attainability. Can it be had in project quantities at project prices?
2. Features. Will it support all of the mainboard?
3. Performance per Watt doing features/tasks.
Architecture agnostic. Any SoC that can pass 1 & 2 should be in the rank order for 3.
If it needs too much power it won't work because we don't uave active cooling though.
That's why you check the perf/W at the end there. Filter out anything that can't be supplied to us, or can't be used with the current mainboard without too much faff, then anything that scores more on the perf/watt scale than our current solution should be under consideration, because you can always tune them to run at a specific wattage that our cooling solution can handle, and still get more performance out of the end of it.
Not an expert by far but in my own experience playing with LEDs on a Raspberry pi, adding resistors wouldn't change the brightness. They use what is called PWM (Pulse Width Management i think?) to make the light rapidly blink at different frequency emulating different brightness. In my case, using LEDs designed for 12v Alarm panels on the 5v Rasperberry Pi, they ended up the exact same brightness.
And PWM can be either be emulated by software although no very accurately or controlled by a dedicated chip which is probably the case here.
I'm glad to see an update, it looks like we're entering the long stretch.
With the lack of people able to debug some issues is there some way we can provide an incentive? Like chip in money for a bug bounty or something?
My knee-jerk thought was that you could have some people help debug by giving them ssh access to a Pyra, and maybe rigging up that setup I seem to remember you had with the webcam the ability to do remote reboots. The problem is, I suspect we're still at the stage where many problems are most easily solved by having physical access to the device so that you can do some tests against the hardware. It's also possible that the remaining issues are few enough that the time and effort set up for remote debugging would be wasted.
Additionally, even with remote debugging, there is still a dependency on finding people who have the skill-set and the time to look into things. Which loops around to the bug bounty as a way to possibly encourage the members of our community with the skills but lack of time.
Hopefully there aren't too many remaining issues that would prevent the the pyra from shipping, which would then give more members of the community the ability to find and solve bugs.
One of the current gotchas being debugged seems to indicate that the units don't currently reboot cleanly; presumably if you shut them down then power them up that's okay though. But fixing that via remote access is likely to be rather difficult.
Hns has already asked for people just to sit and read code so they've got a chance of being able to figure out which bit does what, and therefore be able to provide pointers when bits of it break. I'll admit I myself have been too lazy even to figure out if I've got the skills to be able to help on that yet.
Hopefully, with ED producing prototype units by the day, more will get sent out to developers who can look into this kind of thing, and do other stuff like getting 3d working again and demonstrable software running on it. I'm slightly disappointed that we didn't get a video of a unit running Hydra castle labyrinth, since there were at least two units with the code to run, people to play it and other people there to hold the camera from the recent expo event, but I guess that wouldn't have proved much and we did get photos of that to prove they work.
Just realised I haven't said 'thanks' to @ED for the update - it's great to hear that the tests are going well. To be honest, I've been longing for the time that it's not just up to EvilDragon to take time from his busy schedule to deliver news, and am hoping for more tidbits from other developers and early prototype users soon.
At least it will then start to feel like we're in the transition from a horde of ravenous zombies to a community of Pyra users & developers.
Oh, and thanks @ED!
Somehow I missed this. I'm unlikely to find any bugs that hns may have missed, but if I can find time I would like to see what the code looks like. Is there an easily accessed text file repository for the code or stepwise instructions for a novice on how to find it?
Well some of the direct community projects are here, not sure how much is being worked on here: https://dev.pyra-handheld.com/explore/projects
Kernel development: https://projects.goldelico.com/p/gta04-kernel/
I think aTc builds the Pyra kernel based on hns Letux kernel sources and needs to enable certain modules.
That's a very polite way to say "Get some damn prototypes into the hands of some damn developers already, dammit!"
I've found the message I was thinking of; it's slightly older than I was anticipating, but I guess most of the questions if not all from the last link in the message are still good:
The code is all in the kernel git on goldelico.com, although I haven't quite got my head around the branch structure, but I guess we should be looking at the most upstream branch to be able to answer the questions.
A resistor will work with PWM but will also reduce the maximum brightness. So "100%" brightness would be less bright, but the lower levels might be closer to what you want.
I was talking about an external battery charger, to charge INTERNAL BATTERY ALONE, outside of Pyra: i.e. you charge your Pyra internal battery outside Pyra. In this way you can have 2 (or more) internal batteries and you can swap them yet charged.
You can multiply Pyra autonomy or you can use Pyra without wires, while you charge other internal battery outside Pyra.
This is an example of that device for a LG phone battery:
You insert phone internal battery on that external charger and you charge the phone internal battery outside phone.
Actually we don't know if Pyra would need something like this, because we don't know how many hours it can get with our use running real soft (games, emulators, utilities).
There are standard chargers for internal batteries, but they are more uncomfortable and less safe:
I think a solar powered battery charger is a bad idea because:
1/ Heat is very bad for battery, so it is better to take battery and charger outside direct Sun rays reach.
2/ If you want solar power you can use small panels already on market with USB outlet, so you can connect that panels to Pyra, to battery charger or to other USB devices. And in this way you can have battery/device away from solar panel (connecting between them by wire), avoiding heat from Sun on them.
3/ A portable solar panel cost (in money and energy-pollution) isn't easily amortized: you need to use it a lot of hours/years so you begin to recover the energy consumed and pollution generated in its fabrication (and the money you paid), so if it is more universal (for example with USB or even other connections plus USB) you have more chances to use it and make it profitable (for your pocket and for nature).
I do like the idea of a stand-alone charging case with interchangeable batteries. Why stop there? It will need some sort of processor to manage the charge. It may as well also have a USB port to charge other devices too. And a display showing currently loaded battery capacity. There are a lot of these types of power packs on the market, but I'm not aware of any that a person can swap the battery in.
I saw Lidl a few months ago selling a battery charging station that included along with slots for round batteries (AAs and so on) and PP3 (9v blocks, if I remembered the code correctly) an adjustable section to charging li-ion batteries.
To be honest, I've never found the need for devices as small as phones and handhelds to have an external charger. I did have two batteries for my last phone, and I'd just periodically shut the phone down and swap batteries to keep them both moderately charged. My DSLR camera has an external charger, and can't actually charge the batteries internally, and while plugging a whole DSLR and lens into an adaptor might be a bit of a pain, I find having to yank the battery and wait hours for the little light to change colour a bit of a faff.
My previous round of phones for the family were four Galaxy Note 3 units. When we had them ~2 years and the batteries started to fade, I purchased 4 new batteries and two wall-plug stand alone battery chargers. It was very handy to always have a fully charged battery that could be swapped to if someone forgot to put their phone on a charger or had been playing games most of a day and needed to go out. It is the primary feature I miss on our new phones (Google Fi LG V35thinQ). The answer has been to get a piggyback rechargable battery cover/case that wraps the back of the phone and interfaces at the USB C port.
Having an independent 'battery pack' based around a Pyra battery would be great, but to get there the Pyra (niche item) would first have to sell in quantities to where a derivative device (niche of niche) could make sense.
I would really like an external charger for Pyra batteries.
It would be awesome to just swap the batteries if you need a new one and be done with it instead of leaving it at your desk or whatever to wait for the recharge.
So @EvilDragon: shut up and take my money!
I was planning to build an external charger myself, but II would prefer to buy one from ed.
Separate names with a comma.