I can almost, with 90% certainty saythat the p2 should use an amd apu!


Last edited by a moderator:
 I wonder what if ed was to come out and support this idea, would the community follow or would they stick to their guns -so to speak.
The only way I see ED supporting a non-ARM P2 is if x86 or amd64 processors become much more efficient (e.g. a modern GPU and four 1.5GHz CPUs running flat out at 7W), in which case, I think the community would support him.
 
 I wonder what if ed was to come out and support this idea, would the community follow or would they stick to their guns -so to speak.
The only way I see ED supporting a non-ARM P2 is if x86 or amd64 processors become much more efficient (e.g. a modern GPU and four 1.5GHz CPUs running flat out at 7W), in which case, I think the community would support him.
1. is if he did

2. read the first post, 8W max cpu or 13W max gpu ~60% cpu, and I am making the case for  the dual core version -side note @1GHz the quad scored 0.83 in cinebench using ~8W for the whole system, where the e2-1800 scores 0.5 @18W and the ivybridge 3217u scoring 1.4 @17W. so the dual core should score just below the 18W e2-1800 while using much less power

so I just found some performance metrics on the a6-1450 apu, it uses similar power to the z-60(~17W MAX system included) but has similar, NO better performance to that of the brazos 2 e2-1800. Though it is out of the scope of p2 the results are impressive and unfortunately at this time, metrics have not been revealed for it dual core variant but from a guesstimate, it will have a super low tdp and power consumption similar to the exynos 5250 while offering a modern, ogl 4.3 capable gpu (that has foss driver in dev. for the GNU free software fans).

[the a6 was using around ~8W @1ghz with 4 cores maxed out on cinebench via perfmon(might not be super accurate) and 13W @1ghz with 4 core at 60% and gpu maxed out]

from those metrics and with the guesstimate that the dual core version will have 2 less cores(obvious), lower clocked gpu, lower clock memory(ddr3L[RS?]) and they are both in a soc so it should be a decent for for an embedded use case!

edit: the 8W/13W numbers are for the entire system, not just the apu...although further more accurate testing could solidify those results.

source http://ultrabooknews...temash-testing/
 
The only way I see ED supporting a non-ARM P2 is if x86 or amd64 processors become much more efficient (e.g. a modern GPU and four 1.5GHz CPUs running flat out at 7W), in which case, I think the community would support him.
A 7W CPU would drain the current battery in about 1.5 hours. You'd need something at most half that (and 1/3 would be ideal) to even begin coming close to the current Pandora's battery life.
If monstercameron can find a decent CPU that eats just 2-4W he might start to sway some minds.
 
The only way I see ED supporting a non-ARM P2 is if x86 or amd64 processors become much more efficient (e.g. a modern GPU and four 1.5GHz CPUs running flat out at 7W), in which case, I think the community would support him.
A 7W CPU would drain the current battery in about 1.5 hours. You'd need something at most half that (and 1/3 would be ideal) to even begin coming close to the current Pandora's battery life.

If monstercameron can find a decent CPU that eats just 2-4W he might start to sway some minds.
whats to stop the p2 from moving to a 20-25Whr battery from its ~14Whr? one?
 
whats to stop the p2 from moving to a 20-25Whr battery from its ~14Whr? one?
The size. It would have to be a battery like in a netbook, but then again you could just buy a netbook an be happy.


That CPU would suck Pandoras battery dry in less than 1.5 hours, and that's just the CPU. Add the screen, Wifi, etc., and you'd have a device with less than an hour battery time - unless you increase the size to that one of a netbook.


An x86 mobile device with a decent battery life simply with the size of a Pandora simply can't be built with current technology.
 
The only way I see ED supporting a non-ARM P2 is if x86 or amd64 processors become much more efficient (e.g. a modern GPU and four 1.5GHz CPUs running flat out at 7W), in which case, I think the community would support him.
A 7W CPU would drain the current battery in about 1.5 hours. You'd need something at most half that (and 1/3 would be ideal) to even begin coming close to the current Pandora's battery life.

If monstercameron can find a decent CPU that eats just 2-4W he might start to sway some minds.
http://boards.openpandora.org/index.php/topic/13006-battery-life-compromise/?p=245805

7W flat out for ARM

8W browsing for x86
 
Last edited by a moderator:
The only way I see ED supporting a non-ARM P2 is if x86 or amd64 processors become much more efficient (e.g. a modern GPU and four 1.5GHz CPUs running flat out at 7W), in which case, I think the community would support him.
A 7W CPU would drain the current battery in about 1.5 hours. You'd need something at most half that (and 1/3 would be ideal) to even begin coming close to the current Pandora's battery life.

If monstercameron can find a decent CPU that eats just 2-4W he might start to sway some minds.
http://boards.openpandora.org/index.php/topic/13006-battery-life-compromise/?p=245805

7W flat out for ARM

8W browsing for x86
not browsing, maxed out in cinebench @1GHz...
 
Last edited by a moderator:
whats to stop the p2 from moving to a 20-25Whr battery from its ~14Whr? one?
The size. It would have to be a battery like in a netbook, but then again you could just buy a netbook an be happy.


That CPU would suck Pandoras battery dry in less than 1.5 hours, and that's just the CPU. Add the screen, Wifi, etc., and you'd have a device with less than an hour battery time - unless you increase the size to that one of a netbook.


An x86 mobile device with a decent battery life simply with the size of a Pandora simply can't be built with current technology.
I dont see why the p2 couldnt have a bigger battery, apple did it when they went from ipad 3 to the 4, besides fujitsu uh900 had an extended 25Whr battery.

dimensions 195.37 x 87.98 x 13.35 mm


weight 253.80g -kinda high

http://www.pocketables.com/forum/showthread.php?t=6740

DSC_0016-2.jpg
 
Last edited by a moderator:
That thing costs £1k and lasts 2hours web-browsing.

For an extra £100 you get a double-sized battery (max. life maybe 6 hours), making the whole thing weigh 225g more than the pandora.
 
Last edited by a moderator:
That thing costs £1k and lasts 2hours web-browsing.

For an extra £150 you get a double-sized battery (max. life maybe 6 hours), making the whole thing weigh 225g more than the pandora.
that battery is being sold on ebay for ~$40 but yes this thing is pricey...and the battery is heavy but thats nothing r&d cant solve or atleast mitigate.
 
There's not much you can do about battery weight without inventing a new type of battery - the uh900 already uses a Li-ion
 
There's not much you can do about battery weight without inventing a new type of battery - the uh900 already uses a Li-ion
I would say that our battery technology is near its peak. It's limited by its chemical energy: there is only so much non violent reactant that could be stored in a little box.

So, I don't see how any majors leaps in battery capacity will happen any time soon. 

Also, even if x86 chips can somehow achieve better power consumption than ARM ones, ED should not take the risk from a business stand point. Let microsoft, sony and those big guys bites the crab first.
 
edit: the 8W/13W numbers are for the entire system, not just the apu...although further more accurate testing could solidify those results.

source http://ultrabooknews...temash-testing/
Doesn't make any sense that the system would use 7-8W while web browsing (that's what he says, "Average power utilisation of the platform if around 7-8W for online usage with 50% screen brightness") and 8W while maxing out four cores on Cinebench. The difference in occasional wifi access while browsing is not enough to make up for what should be a massive difference in CPU utilization. Therefore I call BS on one of these claims. I'd rather wait for a real analysis instead of jumping on the first jumbled and inconsistent vague notes of a now unavailable live review. Of course you're clearly happy to jump on anything.

If it has power consumption similar to Exynos 5250 then it shouldn't have much worse battery life while browsing, using the same screen size and battery capacity as the Series 5 Chromebook. End of story.
 
This thread is still alive?


The Pandora lives and dies by its community. Any decisions made must be made in respect of the community that supports it.


We have software specifically programmed for ARM. You cant just flick a swith and everything comes out peachy. There is already a precedent.


No real member of the community would want to just throw away all the thousands of hours spent on creating content for our Original Pandora.
 
It's kind of like proposing to shove a sports car engine into a hybrid. People here like efficiency, and less efficient is less efficient no matter much you mitigate, simple as that. If newer technology gives us more power in our battery then we should seek to leverage that in reducing the Pandora's size, as she takes a lot of criticism for being big-boned already.

And as traylorpark said, 900 packages is hardly just a "quick recompile". Some stuff like DraStic and our enormously optimized PCSXReARMed you can straight up say goodbye to, although maybe there are replacements, I haven't looked into it. But why would we simply throw that away? So we can run DirectX games natively? Stupid, stupid, stupid. Go and lobby Razer to make this kind of thing. OP fits a whole different set of goals, and I think just about every angle has been thoroughly explained. There should be no more misunderstanding; this is not a Pandora 2 proposal.
 
It's kind of like proposing to shove a sports car engine into a hybrid. People here like efficiency, and less efficient is less efficient no matter much you mitigate, simple as that. If newer technology gives us more power in our battery then we should seek to leverage that in reducing the Pandora's size, as she takes a lot of criticism for being big-boned already.
There are many difficulties for the OPT to move to x86. The risk seems rather high to me. Even though it sounds good, but to make the our project plausible, they should be considered.

Here are my concerns.

1. Cost of components and manufacturing: X86 chips are more expensive, and they require more complex machinaries to populate. OPT is a small team, so they do not have much overhead for risks.

2. Android and Linux for ARM can provide most if not all of the functions X86 has. There are more and more games and programs that are built under linux. Not to mention valve is considering making linux based steamboxes.

3. X86 has just joined the competition, I would not expect them to make things that will totally blow ARM away.

4. There are more chip providers for ARM than X86. So getting good deals and services from them should be easier.

5. Moving to X86 may invoke potential legal issues. If some of us installed pirated games and pirated Windows 8 on P2, who knows what microsoftis going to do. I would not trust their judgement. Microsoft and Apple are running by evil, greedy, short-sighted fiends with red glowing eyes.

6. The OPT is much more famous in the ARM scene than in the x86 consumer base. Moving to X86 may be seen as a betrayal by many. 

7. I would deep fry and eat my own PS3 if any of the X86 chips will be open sourced. (may be not literally, its not likely to happen that's all I'm saying)

8. I think OPT can reuse more parts from pandora to build P2 if they stick with ARM, this can save a bit of money. Buying new components and start all over just to jump the X86 wagon would be less sensible.

9. Some of us may leave if P2 became another windows based UMPC.

10. There is nothing wrong with ARM, so why will OPT  even bother risk everything? This is really the most important point I made. In any industry, business owners will not buy new machines just because the new standards are slightly better. They will wait until the maintenance of the old machine is costing them too much. I don't see how the pandora is crippled by ARM. It's doing great, and ARM is going strong as ever. 

I'm not oppose to people who have different ideas from mine, however, moving to X86 is just not feasible strictly speaking. 

If there is one thing I could wish for the Pandora 2, I want the community to continue grow in strength. I want all the wonderful people who have been working recklessly be known, and be respected. The community will decide what specs and hardwares shall be in the P2.  ED and Craigix have no need to worry about how to appeal to the senseless mass crowd of Apple fanboys and fashion victims. As long as P2 is created by the community, it will earn its place in the world. 

That being said, its up to ED and the rest of OPT to handle where to buy and who to buy the parts from, and who to trust for making P2. They have to manage the scope of the project, so I would not be surprised if some promises will not be met. 
 
Back
Top