Finalizing the PCBs


Getting back to the subject, I think most of us Pyra pre-orderers want to be able use the device with each CPU @1.5GHz at full load during a long period.
Only the unreasonable ones who don't understand how modern mobile CPU/SoC systems work (optimized for peak/valley load processing not always pegged).

I really don't wan't to hear that "even at 1.00GHz it will be more powerful than the Pandora".
How is the truth not relevant?

Also, like it or not, newer CPUs will still produce heat, because better 3D, more cores etc.
All CPUs generate heat. CPU manufacturers take the exact same CPU and over/under clock them to fit specific thermal envelopes all the time. The Pyra, and pretty much every other device designed to be able to operate while pocketed, has a pretty low thermal envelope. However, newer CPUs with smaller processes (14nm instead of 35nm) use less energy per unit of computing. Future SoC options for the Pyra will likely have the same or potentially less power requirements and thus produce less heat.

However, this doesn't preclude using a higher max W and max TDP processor - so long as everyone understands that it will only be able to sustain those rates for short bursts.

So it's the right time to include that inside->outside heat spreader that should have been engineered from the beginning. And also add to this case modification the possibility of a micro-fan for future CPUs.
I don't want to have to do it myself for such a pricey device.
I agree that some consideration should be given toward a radiator - i.e. a heat shedding component. But, that should be a consideration for a future product revision or add-on and not the current Pyra.

I have shared many ideas around this over the last several years. The shortest path to being able to really shed heat is to replace the lower plastics (battery level surrounding piece) with a milled aluminum piece. This would radiate and conduct to the surrounding air AND the user's hands - making it liquid cooled. The single piece of milled aluminum to replace that piece of plastics would cost $000s to develop and would need to add charges on the order of $75-$150 per unit.

Short of that would be to run a sheet of aluminum through between the SoC board and main board, under the SDXC slots and fill in the right half of the case front with folds/ribs for added surface area to act as a a convection radiator. Since this could be made out of stamped & folded sheet aluminum, it would not be nearly as expensive. It would still require a case mold revision - and as such should not be considered at this time.

ED's placement of a heat capacitor/sink in the case void under the USB ports intrigues me for possible future user modifications. I could see potential to cut out a section of the case plastics, create a mold negative, create a clay mold then pour a solid silver heat sink with radiator ribs that fits into the same space plus gets exposed to the air. (If you're going to dream... it would take ~$150 in silver metal. Why silver? It has the highest thermal conductivity short of truly exotic and expensive stuff.) It could also be done using aluminum which would be nearly as good or copper if you have access to an oxygen free mold pour environment or lead if you don't care about health & safety. The point here is that from the placement of EDs heat sinks/capacitors, there is nothing but a thin plastic wall between that spot and the outside. If the chunk of metal can fill that void entirely, it wouldn't take much to cut some windows to expose that metal to the outside world. But - that is a customization for the modifications community for now and doesn't need to happen for Pyras to happen.
 
as a last resort, there is always thermal epoxy : https://www.masterbond.com/tds/ep30tc

no modern mobile cpu is meant to run full power all the time even desktop cpus are supposed to throttle and sleep a lot and only spike for short times (because that is more energy efficient, not because of insufficient cooling)

the one place I could image an extra fan (thats big enough to make a difference) would be inside the battery compartment, ... Pyra extreme edition with smaller battery, extra fan and some holes in the bottom case for air intake
 
And that cooling works perfectly fine without a fan and has heatsinks small enough to fit into a case (and don't forget that inside a case the hot air naturally can't move away, so put your devboard into a plastic box as well)?
If there was an easy way to handle this, how come even multibillion dollar smartphones company can't handle this?

The point is to make out of the constraints the most efficient way to dissipate heat.
There's not that much choice. While the system shown on the video is a step forward, this is not enough.
 
If there was an easy way to handle this, how come even multibillion dollar smartphones company can't handle this?
Because they're in a race to be the thinnest for some glob forsaken reason. They pack everything so tight there's no room for a heatsink or even air, so they design chips with fast-finish in mind so they can throttle back down, then market them at their top speed. That IS their "handling" it.
As far as I'm concerned, adding the heat sink already puts you three steps ahead of the major companies, even if it still needs to throttle after a while.
 
The point is to make out of the constraints the most efficient way to dissipate heat.
There's not that much choice. While the system shown on the video is a step forward, this is not enough.

How exactly do you come to that conclusion? Most of us want a small form factor mobile device. If you want to run your handheld at full speed for hours on end get a device with a fan and proper heat sink ( I suggest a small linux laptop to you).

I for one want a small device with NO fan and not a metal brick weighing 2 and a half pounds. So I rather use the pyra in the way a mobile SoC is intended and not like a gaming laptop and get way better battery life along the way. Because if you are enabling the SoC to run at full TDP you will drain the battery even faster.
 
You are misinterpreting what I said.
We have the opportunity to dissipate the heat better, we should take it to get the most out of the device.
This is the way the Pandora have been designed : to have the best device.
 
  • Like
Reactions: ___
We have the opportunity to dissipate the heat better, we should take it to get the most out of the device.
How? Yes, I saw the mockup of one solution, but I don't think it solved the big issue: Extruding the spreader outside the plastic case would expose the mechanical strains of the outside world to the whole CPU-board and from there to the entire motherboard?

Even mild pressure on anything connected to the spreader would have and effect inside the case. It's a problem hard enough to avoid without a heatspreader extruding out of the case (no matter how thin or flexible).
Way back when an aluminium Pandora case was discussed there was talk of an aluminium Pyra option once production runs smoothly. That would greatly alleviate this issue (though it would cause a sweaty palms issue...).
Toying with untested heatspreader designs which very likely risk strain on the PCB:s and internal connections can be done later when the basics (i.e. electronics) operate properly.

Don't misunderstand: I'm also in the "camp" that wants maximum performance, since I'll be spending time waiting for programs to compile no matter how fast the Pyra is! I often have to leave the Pandora on overnight just to compile software so I can see next morning what the next issue to fix is... and after the fix it sits in the corner redoing most of the work it just did last night :-(. This is why there's software I don't even try to upgrade unless I'm having a holiday.
But I really think solving this heat issue witht he Pyra without either causing mechanical issues or simply using a fan is a very, very hard problem.

(I had already paid for my order for one aluminium Pandora case when they got cancelled... really if one was offered now I'd take it in a heartbeat even today no matter how close the Pyra is! But that is not due to cooling, that is due to looks and durability for a device which is basically still my daily driver! So, if a Pyra ever gets an aluminium case, I'll definitely get one whether it's cooler or not!)
 
I think you forget that in our reality you have to compromise for every thing. In this case you have to find a working compromise between heat mitigation and case strength/development time. If you add more openings in the case you add instability and complexity. You also add a lot of time for a redesign of the molds. In addition you add weigth, weigth some of us do not want to lug around. A too big heatsink may also play havok with WiFi and no one in his right mind wants that right? Or you will ahave a lot of people look this way::|

Edit:
For everyone who wants to do heavy stuff(compiling big software, number crunching, ...) please consider proper hardware for this tasks. Use the pyra as a terminal for a server/cloud instance/... which does the cross compile or number crunching much faster than a full powered Pyra can ever achive.
 
On the video, you can see that the "heat cable" is flexible, so no mechanical force from the outside can affect the boards.
 
On the video, you can see that the "heat cable" is flexible, so no mechanical force from the outside can affect the boards.

What application do you have in mind that would be appropriate to running a Pyra at 100% processor usage on both cores for long enough to be 'continuous'?

Since you have a pair of dev board class systems, I'm assuming you have something in mind.

Also, which dev systems do you have again so we don't have to search? What are the cooling solutions on them?
 
You got me - He banned me for saying "what does he have in his pouch?" - an obvious reference to a koala. An offense if I've ever heard it. It sounds even worse HERE, taken out of context - but the language police are always out for blood.

No I "banned" your alternative account "flapjack" (which you just admitted owning as that pouch comment is on that account) for posting significant amounts of drivel in on-topic threads.

You're welcome to stay, just don't be a moron.
 
You just admitted things as well. Thank you.
[doublepost=1512595926,1512595396][/doublepost]
What application do you have in mind that would be appropriate to running a Pyra at 100% processor usage on both cores for long enough to be 'continuous'?

Since you have a pair of dev board class systems, I'm assuming you have something in mind.

Also, which dev systems do you have again so we don't have to search? What are the cooling solutions on them?

One would think that at some point, some people would try to max out the settings on this handheld device. Squeeze some performance out of it where they can. Especially considering that the specs are somewhat lacking now. I'm not buying this device for the sheer processing power however - that's clear.
[doublepost=1512596066][/doublepost]
You're welcome to stay, just don't be a moron.

Well thanks. I'll try not to be. I didn't understand your hair-trigger response to me last time - so maybe you should adhere to your rule(s) as well. Thank you Wally.
 
Yes I admitted to doing my "voluntary" job as a moderator which other members asked me to fulfil as you were spamming the boards up with nonsense with your alt account (which is against the rules but we ran with it for ages as we figured you might have changed). Not much of a hair-trigger as you say.

Happy to share the nonsense with people if anyone wants to see it is not visible to the public.
 
And that cooling works perfectly fine without a fan and has heatsinks small enough to fit into a case (and don't forget that inside a case the hot air naturally can't move away, so put your devboard into a plastic box as well)?

If there was an easy way to handle this, how come even multibillion dollar smartphones company can't handle this?

IMHO maybe 1.5 - 1.7GHz is enough for this iteration/design. Once you get the handhelds out the door, you might find that people are able to add their own cooling solutions. Maybe you and Askarus/staff will find solutions - shunting some of the heat to the SD card slots...anything metal that goes out to the outside of the case? Maybe these options have already been discussed? My apologies if I'm bringing up topics that are old - I haven't been back here in a while (a year).
[doublepost=1512597369,1512597271][/doublepost]
Yes I admitted to doing my "voluntary" job as a moderator which other members asked me to fulfil as you were spamming the boards up with nonsense with your alt account (which is against the rules but we ran with it for ages as we figured you might have changed). Not much of a hair-trigger as you say.

Happy to share the nonsense with people if anyone wants to see it is not visible to the public.

I'll take the high road here - it's nice to be back. Thanks for the warm welcome Wally.
 
You are misinterpreting what I said.
We have the opportunity to dissipate the heat better, we should take it to get the most out of the device.

I'm not sure how much you know about mold production, but just adding holes at the backside for heat dissipation needs another tool (which is a couple thousand EUR) and you'll see a small recession where that tool is applied.
We're not printing 3D cases here. With molds, you have a top and a bottom. Think about how they're made: They're being drilled with CNC machines from the top and bottom.
If you want to add any holes SIDEWAYS in the case, you can't drill that in the mold, it's physically not possible.

Therefore, for every hole, clip, or whatever that can't be milled, you need an inlay too which needs to be produced additionally and put into the mould each time for production.

This is the way the Pandora have been designed : to have the best device.

No, with the Pandora, compromises had to be made as well.
It's mostly built around the OMAP3: It uses features that the OMAP3 supports (SD Card slots, etc.)
Everything else (touchscreen, Wifi, analog sticks) are standard features for a mobile device as well.

We've added a lot more things in the Pyra (mobile internet), even things that the OMAP5 doesn't support out of the box (like three SD Cards and an eMMC).
The Pandora has nothing like that, AFAIK.

The Pandora was amazing, but there were a lot of compromises that had to be made as well.
 
Back
Top