How will the Pandora 2 differentiate itself from the market ?
- Thread starter thatgui
- Start date
WizardStan
Mega GP Mania
- Joined
- May 24, 2008
- Messages
- 16,731
The heat capacity of lithium ion is around 1000 J/kg*K. The battery weighs about 0.25kg, assume an average room temperature of 20C and an ideal resting temperature of 4C, a difference of 16C, and you're looking at an energy cost of about 4000J, or 10KWh per year.
Producing a lithium ion battery involves holding it at about 700C for several hours (plus some additional machinery, but it's mostly negligible against the cooking), maintaining that temperature requires about 175000J or 0.4KWh for an 8 hour cooking period (I think it's actually less than 8 hours but I can't find the exact length of time required, just "several hours"). Let's multiply that by 10 because I'm missing a lot of additional overhead such as mining and shipping, and the cost of making our lithium polymer battery is around 4KWh.
4KWh is less than 10KWh. It costs more energy to keep your battery cool for a year than it does to produce an entirely new battery. If you recycle your lithium batteries as you should, the cost goes down further, and avoids unnecessary waste in the landfill. Your argument about storing a battery vs using it and buying a new one when it wears out for environmental reasons doesn't hold up.
Looking at this...let me get this straight: a Lithium Ion Battery produces heat all the time while it's stored at 4 degrees Celsius? o.Ò
I could understand this calculation, if the fridge would just run because of this and has the exact size of the battery
or if you constantly take the battery in and out of the fridge, but since this would require opening the door all the time...man, that's some weird fantasy there.
BTW: My fridge never runs at 4, it's about 7.
I haven't got much knowledge on this field and am always happy to learn, but some points in this calculation seem very fishy to me.
I wouldn't store the battery of the pandora in the fridge anyway, i just don't want to stress it all the time.
I just do this with my notebook-battery, because i nearly NEVER use it (maybe once a year?) and do not have to put it in the device at all at the rest of the time, because it's powered by cable.
I could understand this calculation, if the fridge would just run because of this and has the exact size of the battery
or if you constantly take the battery in and out of the fridge, but since this would require opening the door all the time...man, that's some weird fantasy there.
BTW: My fridge never runs at 4, it's about 7.
I haven't got much knowledge on this field and am always happy to learn, but some points in this calculation seem very fishy to me.
I wouldn't store the battery of the pandora in the fridge anyway, i just don't want to stress it all the time.
I just do this with my notebook-battery, because i nearly NEVER use it (maybe once a year?) and do not have to put it in the device at all at the rest of the time, because it's powered by cable.
Last edited by a moderator:
WizardStan
Mega GP Mania
- Joined
- May 24, 2008
- Messages
- 16,731
No, the battery wouldn't produce heat, the ambient air is producing the heat; that's why I made special mention of room temperature. It takes energy not only to reduce the temperature but also to keep it there. Make no mistake, I'm not claiming it's a perfect calculation, there's a lot of wiggle room in my numbers and insulation will help a lot, but on a high level it should be enough to demonstrate that actively cooling your battery doesn't necessarily have the direct saving you seem to think it does.
Assuming the charge chip is finally doing what it is supposed to, there should be no problem. It'll stop charging when full and give the battery a chance to relax. There wish trickle charging issues with the original firmware but tweaks have mostly fixed all those problems now, except for some of the more recent CC Pandoras which are just weird to begin with, unfortunately, but that's a different problem.
Back on topic: these are good points. The P1 already hit the bullseye (minus some problems) of two market groups.
+ The group wanting a handheld linux computer
+ The group wanting a premium retro gaming experience
Of course there is significant overlap between the two.
Craig's goal is to address a third market group with both ICP2 and P2
+ The group wanting a premium android gaming experience with hardware controls.
The last group is potentially huge, but the numbers are more uncertain, depending on how good the Android support actually is, how many opt for the ICP2 alternative and how effectively online viral advertising spreads news of the device.
Integrating Android and Linux:
1) Dual Boot: Simplest to implement, high compatibility, fragmented user experience
2) Run Android apps under Linux (android hardware abstraction layer): incomplete Android compatibility?
http://www.linuxfordevices.com/c/a/News/Myriad-Group-Myriad-Alien-Dalvik/
3) Run Linux under Android: difficult to provide direct access to framebuffer / overlays for linux emus?
http://androlinux.com/
http://www.xda-developers.com/android/got-linux-the-linux-on-android-project-can-help/
4) Hypervisor - run Linux + Android concurrently
http://www.openvirtualization.org/
http://www.cnx-software.com/2011/10/30/run-2-os-simultaneously-on-arm-omap4-with-codezero-embedded-hypervisor/
Each approach appears to have advantages/disadvantages. Choosing any of the last three would likely help differentiate the P2 from other (at least current) products.
I think OPT should be seriously considering these options.
Comments?
EDIT: moved to a new thread
Last edited by a moderator:
Grench
Forum Addict!
- Joined
- Oct 3, 2008
- Messages
- 6,629
#4 would be ideal if we could do it right.
http://en.wikipedia....wiki/Hypervisor
Either as a Type 1 or with a stripped down command line only bare bones hypervisor OS.
In theory we're talking about a quad core CPU. With a bit of user due dilligence, neither concurrent running OS should be consuming more than one core's worth of tasks when it isn't in the foreground - in theory since the background OS isn't getting actively used, it should drift into sleep mode.
Being able to simply toggle between Android and Linux - without rebooting, without booting one OS inside the other (even though they would both be inside the hypervisor) - that would be very cool.
With a quad core CPU, enough RAM (4GB?), 'disk', a hypervisor and patience, it should be able to run 4-5 OS's simultaneously - just don't expect to have many applications running concurrently inside of those OSs.
Still - being able to toggle between Android and Linux on the fly without disrupting active processes in the background OS - THAT would be a perfect 'dual OS' solution.
This topic spur has moved to:
Last edited by a moderator:
Thatgui I don't know where the thinness of a device means it's a cellphone concept originates from.
As for the adapter, you presented a special case problem and I presented one of several special case solutions.
As for Craigix's intentions, people (like yourself) are saying he said it would be "this and that" I'm asking for the evidence saying he said "this and that" because all I have him saying is that it will be thinner. Which is an easy thing to do, even with a larger battery.
When I say define the terms, I mean don't say things like "x has a minimal effect on the environment" without backing that up in strict terms or evidence. What one thinks may be true may in fact be just the opposite, or only true in specific circumstances. This is why these terms need to be defined.
Yeah I might be wrong and my dimensional analysis did an exit stage left but the numbers I get after multiple checks are:
Cost at 4000j per day equals 0.40515 Kwh/year and at the low end 0.0406 Kw/year (actually this is wrong because I divided by an additional year). Let's go with the high number as I might have had one too many decimal places involved.
Cost at 175000j per 8 hours works out to 53.229 Kwh/year (checked twice). The order of magnitude to produce a battery is 2 times that of refigerating it. Also a couple of liters of H20 would store MUCH more caloric heat at room temperature than a battery (water is really good at that) so I would expect the cost of refrigerating/freezing water that is at room temp to be much more than that of the battery. We also have the volume of the refrigerator to consider and all the other items already stored there that would contribute to lowering the cost of cooling the battery.
As for the adapter, you presented a special case problem and I presented one of several special case solutions.
As for Craigix's intentions, people (like yourself) are saying he said it would be "this and that" I'm asking for the evidence saying he said "this and that" because all I have him saying is that it will be thinner. Which is an easy thing to do, even with a larger battery.
When I say define the terms, I mean don't say things like "x has a minimal effect on the environment" without backing that up in strict terms or evidence. What one thinks may be true may in fact be just the opposite, or only true in specific circumstances. This is why these terms need to be defined.
Yeah I might be wrong and my dimensional analysis did an exit stage left but the numbers I get after multiple checks are:
Cost at 4000j per day equals 0.40515 Kwh/year and at the low end 0.0406 Kw/year (actually this is wrong because I divided by an additional year). Let's go with the high number as I might have had one too many decimal places involved.
Cost at 175000j per 8 hours works out to 53.229 Kwh/year (checked twice). The order of magnitude to produce a battery is 2 times that of refigerating it. Also a couple of liters of H20 would store MUCH more caloric heat at room temperature than a battery (water is really good at that) so I would expect the cost of refrigerating/freezing water that is at room temp to be much more than that of the battery. We also have the volume of the refrigerator to consider and all the other items already stored there that would contribute to lowering the cost of cooling the battery.
Last edited by a moderator:
thatgui
Advanced Member
- Joined
- Apr 2, 2009
- Messages
- 3,048
Not exactly, it was a comment on an argument you gave about outsourcing user filesystem work to an external device to avoid removing/inserting microSD Cars (for the sake of people having limited fine motor skills). Please see it in that context.
true, but wasn't that the point of it as making a device thinner also makes it slimmer - or not ?
As expressed before I don't see that this will be an easy thing to do, and as no one is arguing against it I see no reason to believe otherwise.
I understand this, but I wanted to to be more term specific
Thatgui: The special case I was referring to was when you said this.
That functionality remains, as would transferring microSD<->USB if a USB port is present on the P2, and allowing for the case that you meant USB flash/jump drive when you said "USB" (as no other device is present). If no standard USB port is present you will need an adapter to exchange microSD<->USB or even SD<->microSD. You could also turn a microSD card into a traditional jump drive (USB flash drive) by inserting it into a converter. Here are a few examples I found. Lot's of images here. Or even a multicard adapter here. Failing that you could use one of these adapters to mate your jump drive to the unit.
If something is thinner I would agree that it could be argued that it is thinner. However sleek and thin aren't the same thing. So when you said this:
As for making the P2 more slim, consider the thickness of the Samsung Galaxy S3. It's 8,33 mm thick (or thin if you prefer). It is able to be so thin because it uses the quad-core 1.4ghz version of the Exynos 4 SoC. For perspective here is an image to show how tiny it is compared to the connectors surrounding it (this board cost $129 btw). This is an even smaller version of the same platform starting at $69 dollars. Here is a spec sheet for the quad-cor Exynos chip that shows 1080p full HD out, 3D graphics etc.
Now the $129 version Odroid-X daughter board, when mounted to the main board is thinner than the size a standard USB slot would take. Take note how much real estate a full SD slot takes.
Specs+ Dimensions
As you can see the limiting factor in regards to thickness are the connectors themselves. The limiting factor towards maximizing board real estate is the width of the things like SD slots that populate the board. I'm not sure how Craigix plans on making the device thinner, but I can definitely see how it is possible, even including full gaming controls, full hd out, 5 inch screen and a larger battery.
To define terms you can explicitly state what you mean or provide a citation. For example one could argue that lithium itself is rather benign, but the manufacture of it requires that copper and aluminum have to be mined to make them (taking a lot of energy and chemicals to mine it effecting the environment). Another might argue that mining lithium has "just a very small fraction of the entire environmental footprint of a battery-driven car." Another might argue that where lithium is mined from are some of the most beautiful and ecologically unstable sources in the world. I could go on but will stop at this last point. Depending on where you are the method of recharging these batteries are the most significant, coal power plants would be bad for example and even worse as the demand for lithium power (cars for instance increases). Sources [1][2].
So is lithium power's effect on the environment a big thing or a small thing? Well. It depends on how you define the terms.
So you are on vacation, with no other devices present but need to transfer files back and forth like so SD<->SD/USB. I assume the question you pose is how to do microSD<->SD/USB, and you feel that another device would be needed to do so. Now since you said no other device is present, I can exclude the case where you are talking about transfering pics between a pandy and a camera, and that the transfer would be either between the cards present on the pandy itself SD<->SD, or in this case microSD<->microSD
(or even microSD<->SD if both slots are present-if not you will have to use an adapter shown
below). So yes it would require you to add that to your peripherals bag, besides your power adapter etc.
below). So yes it would require you to add that to your peripherals bag, besides your power adapter etc.
That functionality remains, as would transferring microSD<->USB if a USB port is present on the P2, and allowing for the case that you meant USB flash/jump drive when you said "USB" (as no other device is present). If no standard USB port is present you will need an adapter to exchange microSD<->USB or even SD<->microSD. You could also turn a microSD card into a traditional jump drive (USB flash drive) by inserting it into a converter. Here are a few examples I found. Lot's of images here. Or even a multicard adapter here. Failing that you could use one of these adapters to mate your jump drive to the unit.
If something is thinner I would agree that it could be argued that it is thinner. However sleek and thin aren't the same thing. So when you said this:
I was looking for you to show when/where this was said. For instance a device can be thick and be sleek at the same time.
As for making the P2 more slim, consider the thickness of the Samsung Galaxy S3. It's 8,33 mm thick (or thin if you prefer). It is able to be so thin because it uses the quad-core 1.4ghz version of the Exynos 4 SoC. For perspective here is an image to show how tiny it is compared to the connectors surrounding it (this board cost $129 btw). This is an even smaller version of the same platform starting at $69 dollars. Here is a spec sheet for the quad-cor Exynos chip that shows 1080p full HD out, 3D graphics etc.
Now the $129 version Odroid-X daughter board, when mounted to the main board is thinner than the size a standard USB slot would take. Take note how much real estate a full SD slot takes.
Specs+ Dimensions
A Korean hardware manufacturer called Hardkernel is launching a high-end alternative. The company’s new ODROID-X board comes with a Samsung Exynos 4 processor, a quad-core CPU clocked at 1.4GHz. The board also has a quad-core Mali 400 GPU, 1GB of RAM, six USB host ports, an ethernet adapter, headphone and microphone jacks, and an SDHC card slot for storage.
With four times as much RAM as the Raspberry Pi and a much more powerful processor, the Hardkernel board seems like a nice option for more computationally-intensive usage scenarios. The system is still highly compact, measuring at about 3.5 x 3.7 inches.
Because the Hardkernel board is based on the Cortex-A9, which uses the ARMv7 architecture, it is capable of running the latest version of Ubuntu in addition to Android. Ubuntu doesn’t support the Raspberry Pi’s ARM11 architecture (the Raspberry Pi foundation recommends using Fedora on its board).
Of course, users can expect to pay more for the higher specs of Hardkernel board. It is available for $129, which is a lot more than the Raspberry Pi’s hobbyist-friendly $35 price tag. It’s still less than the $183 PandaBoard ES, however, which has a dual-core 1GHz ARM CPU and 1GB of RAM.
With four times as much RAM as the Raspberry Pi and a much more powerful processor, the Hardkernel board seems like a nice option for more computationally-intensive usage scenarios. The system is still highly compact, measuring at about 3.5 x 3.7 inches.
Because the Hardkernel board is based on the Cortex-A9, which uses the ARMv7 architecture, it is capable of running the latest version of Ubuntu in addition to Android. Ubuntu doesn’t support the Raspberry Pi’s ARM11 architecture (the Raspberry Pi foundation recommends using Fedora on its board).
Of course, users can expect to pay more for the higher specs of Hardkernel board. It is available for $129, which is a lot more than the Raspberry Pi’s hobbyist-friendly $35 price tag. It’s still less than the $183 PandaBoard ES, however, which has a dual-core 1GHz ARM CPU and 1GB of RAM.
As you can see the limiting factor in regards to thickness are the connectors themselves. The limiting factor towards maximizing board real estate is the width of the things like SD slots that populate the board. I'm not sure how Craigix plans on making the device thinner, but I can definitely see how it is possible, even including full gaming controls, full hd out, 5 inch screen and a larger battery.
To define terms you can explicitly state what you mean or provide a citation. For example one could argue that lithium itself is rather benign, but the manufacture of it requires that copper and aluminum have to be mined to make them (taking a lot of energy and chemicals to mine it effecting the environment). Another might argue that mining lithium has "just a very small fraction of the entire environmental footprint of a battery-driven car." Another might argue that where lithium is mined from are some of the most beautiful and ecologically unstable sources in the world. I could go on but will stop at this last point. Depending on where you are the method of recharging these batteries are the most significant, coal power plants would be bad for example and even worse as the demand for lithium power (cars for instance increases). Sources [1][2].
So is lithium power's effect on the environment a big thing or a small thing? Well. It depends on how you define the terms.
Last edited by a moderator:
thatgui
Advanced Member
- Joined
- Apr 2, 2009
- Messages
- 3,048
I guess we are going round in circles here, It was a reaction to one of your ideas in the spoiler:
"Let's look at microSD functionality/convenience for example. With microUSB hosting you could plug your Pandy into your laptop or what have you and perform file manipulations, copying etc without having to remove an SD card to do so. Say you have a Resident OS that is dedicated to an SD slot, swapping that slot in and out is problematic aand reduces you to only one easily removed SD slot."
Nothing more, and I presented you with an usecase were that fails.
Guess you are right on that, but you do tend to use these words in an interchangable manner too. That does not make my fault right of course, but it gave my the impression that you don't care that much about the differences:
So your point is ? Powerfull SoCs are thin nowadays - no big news. But after that you are getting to the point. So why make the comparison with the S3, to make a general point again ? How useless is that ? The SoC of the current Pandora is not much thicker (and again I remind you that making the SoC changable will add to space that is needed for it), so it boils down to other things, and those things won't shrink much in the next Pandora, even if craig learnt something, if he wants it to be smaller he has to cut down on features.
Leaving in my opinion less and less arguments to why anyone should favor a Pandora over an average smartphone combined with an attachable controller
You presented a specific use case where my spoilered text would fail, is this not true? I then provided a specific solution(s) for your specific case.
Can you provide an example of where I'm using sleek interchangably? For instance the Zipit Z2 is sleek but by no means thin, nor is the Xperia mini pro thin, yet it is sleek. I might have used the term as the community did and then argued against that point. I can't be certain without examples. I admit I may have done so, but need a reminder.
About the S3 cellphone. I think you missed the point where I showed that the same SoC of the S3 is being used in a kind of Raspberry Pi inspired device (I even provided 2 product examples). So the S3 was used to show how thin a device could be compared against thicker devices that use the SAME CHIP. This would be obvious if you clicked the various links I dug up that illustrates that point. In doing so you would have seen you could reduce the thickness of the device without cutting down on functionality. Which has been my consistent argument all along.
Edit:
Here.
Can you provide an example of where I'm using sleek interchangably? For instance the Zipit Z2 is sleek but by no means thin, nor is the Xperia mini pro thin, yet it is sleek. I might have used the term as the community did and then argued against that point. I can't be certain without examples. I admit I may have done so, but need a reminder.
About the S3 cellphone. I think you missed the point where I showed that the same SoC of the S3 is being used in a kind of Raspberry Pi inspired device (I even provided 2 product examples). So the S3 was used to show how thin a device could be compared against thicker devices that use the SAME CHIP. This would be obvious if you clicked the various links I dug up that illustrates that point. In doing so you would have seen you could reduce the thickness of the device without cutting down on functionality. Which has been my consistent argument all along.
Edit:
Here.
Last edited by a moderator:
thatgui
Advanced Member
- Joined
- Apr 2, 2009
- Messages
- 3,048
no it is not true, you gave example, i presented a use case where it fails, you started altering by bringing other factors into play (adapters)
correct, and consistently useless as it leaves out nearly every factors that comes into play when talking about making the Pandora thinner. You don't need to prove over and over again that a device with the same chip can be smaller if nearly everything that really adds to a device volume is left out. As you say the odroids size is largely determined by the size of its connectors so why pull the S3 in if it only has one external connector which is specifically made for such thin devices, while that was no design consideration for the odroid ?
The S3 audio jack for example is smaller than the ones used above. Surely there are lower profile USB ports available by now (there's two examples of smaller full sized components right there). Does the P2 require a power jack? Why not charge over USB? Or microUSB? Can the control pad be made thinner? Have you considered that? The thickness of the S3 includes the screen itself, have you considered that? Did you consider that the odroid-u uses microsd and microhdmi?
There's really quite a few options available.
thatgui
Advanced Member
- Joined
- Apr 2, 2009
- Messages
- 3,048
sigh..
I considered a lot of them but most of your examples make mostly room in horizontal, not in vertical dimensions. At least if craig is persitant in the usb-port it is a given, dpad and Nubs will most likely be the same as in P1,ICP,ICP2 (cost factors). The S3 is no clamshell device so you will need to add a few layers of plastic additonally. Dont forget about the Keyboard, as craig stated somewhere to reduce costs the ICP2 keyboard will most probably find its way into P2, but looking at the pictures, the construction didn't look much thinner to me.
FaeMinx
Rainbow Liberation Instigation
OMG *.O
I'm blind.
Sent from my HTC Sensation Z710e using Tapatalk 2
I'm blind.
Sent from my HTC Sensation Z710e using Tapatalk 2
patrick295767
Member
- Joined
- Jan 10, 2013
- Messages
- 223
Will the Pandora 2 have a better keyboard?
would it be possible to make it typing friendly with one added row to the current Pandora 1?
regards
would it be possible to make it typing friendly with one added row to the current Pandora 1?
regards
It will almost certainly have the same one as the iCP2, which looks to be a lot better, but we can't tell until the iCP2 is released.
