Pandora Needs To Go X86 (eventually)

[Raz]

If you just need it for YouTube... There's probably a workaround. The iPhone has a way to do it using JavaScript or a Safari plug-in known as "iMobileCinema".

In other words, Flash is going to become possible, not necessarily from Adobe. Truth is, I'd like flash too and I'm dependent on it.

If you want the Pandora to completely change what has been worked on for years, find another device. It took a while for the Pandora team to make the Pandora. Sure, they're gonna have future releases, but much easier when they already have something to work with.

I use Windows x86 myself and I don't think it's crappy. It's not perfect, but nothing is. I wanted a PS2 and XB0x emu for the panny, not gonna happen for me... So either give up on Pandora x86 and enjoy the Pandora or find another device.

In my opinion, it doesn't matter what the device is running as long as it can do what I want - listen to music, play videos, gam

Sorry for going all over the place, but these people have worked hard, adn well they've done a lot already IMO.

 

[vputz]

Heh.

As a fellow Zaurus C1000 survivor, I feel MD's pain. But the situation has changed a great deal from when OpenZaurus was the only real game in town, software selection was minimal, and nothing worked.

The situation has already improved greatly, with various Linux branches sporting ARM variants (DebianARM, which I used on my NSLU2, was remarkably full-featured). The Beagleboard could be the single most important hobby computing device in years and years, and will only get more significant. And it will get more significant because it is different and offers different capabilities.

If the Pandora was atom-based, I wouldn't be nearly as interested.

 

[mali]

I'll repeat it for co-writing:
Hardware accelerated Flash10 for ARM Cortex exists and will be available to consumers soon.
There is much movement behind the scenes. Be prepared for cool stuff on Pandora

 

[gibberish]

Starcraft, Morrowind, Diablo series, Fallout series, Neverwinter Nights series, Battlefield series... all of that would fit on an 16 gig SD card, and you'd still have more than enough room left over for every single NES, SNES, and Genesis rom.

nope, every snes rom adds up to about 10gb. genesis roms about another 10, nes is close to 1gb. morowind is about 1 gb, nwn is at least 4gb.


without wanting to derail the thread here, i want to correct you on your figures.

no intro snes stands at 2.8GB with 3425 zipped roms
no intro genesis stands at 1.02GB with 1616 zipped roms
no intro nes stands at 261MB with 2544 zipped roms

bearing in mind the roms do not need to be unzipped to be played, then he is right - there is room for all of those systems archives and more.

 

[Bronek]

Ok, so to clarify:

EVENTUALLY Pandora needs to go x86. Eventually means "not right now", because the embedded x86 blitz is just beginning.

I strongly disagree. If you need a netbook, buy a netbook. In diverse environment, each architecture has its merits - including both Intel and ARM. If you have any doubts about ARM's future, check here http://business.timesonline.co.uk/tol/busi...icle5740769.ece .

Can someone close this thread already, nice please?

 

[joa]

Offtopic: But Please if an Admin hasn't already can they send Hando an email to fix the insanely annoying broken forum.

Those added quotations, and broken links.

please!

 

[WizardStan]

Successful troll is successful.
Seriously, he stopped replying 3 pages ago, and everyone is still arguing with him.
MD made poor arguments and prophetic ideals based on outdated information, and when all his points were countered, instead of revising his arguments, he took his ball and went home. And now you're all still standing in the field trying to decide who's turn it is to play third base

 

[Tor]

ARM is a great architecture, objectively speaking. But, from a general-purpose software dev standpoint, is a dead end. [...]

And still ARM CPUs outsell x86 CPUs by a factor of almost three. There doesn't seem to be a slowdown either.

For numbers verification, check out sales numbers from e.g. the arm yearly report.

 

[ldesnogu]

And still ARM CPUs outsell x86 CPUs by a factor of almost three. There doesn't seem to be a slowdown either.

IIRC ARM sold about 3 billion processors last year. Did really Intel sell one billion CPUs?

BTW this doesn't prove ARM is better

 

[Tor]

As for the 'difficult to develop for ARM' argument, here's my take:

I'm also familiar with the Zaurus, which was the first Linux-on-handheld device I owned. Yes, developing for that one isn't that easy. But it's a red herring. For the last two years I've owned a Nokia internet tablet, which is very different from a Zaurus, and this is the type of device the Pandora will be, so I think I have a pretty good idea of how it will be to develop for the Pandora: Developing for the Nokia N8x0 is pretty much the same as developing for a standard Linux laptop or desktop. This is the type of device we're talking about. Not a Zaurus-type of device.

I just keep my Linux application source on the same laptop and build either for x86 or for ARM. There's _no_ difference worth talking about. The differences that exist are only in that the Nokia UI library needs some tweaking in the code that's not needed for the desktop Linux version, and that has nothing to do with the target being an ARM CPU.

Lately I've also started to work with the Pandora SDK. Guess what? It's just like developing and building applications for the desktop. There's _nothing_ that would make a nanogram of "improvement" if the target (Pandora) had happened to be an x86 instead of ARM.

IIRC ARM sold about 3 billion processors last year. Did really Intel sell one billion CPUs?

Apparently they sold a little above 1 billion CPUs, which indeed sounds very high.. it's not as easy to verify that number as the ARM number though, so for all we know the difference may be even higher.

QUOTE

BTW this doesn't prove ARM is better


According to the way some marketing folks I know thinks, it must be..
But of course it doesn't mean that at all, but it does indicate that the market for ARM developers will not diminish any time soon.

 

[cosurgi]

ARM is for the winners! good that it was moved offtopic.

If you want x86 just buy eee or wind, or whatever. Pandora is not good for you, it's that simple.

 

[adamorjames]

Repetitiveness...

EDIT: For the last time Flash is going to be ported to the ARM architecture by Adobe.

 

[monkeyo2]

Screw x86 and Flash, Pandora 2 should be IA-64 with Bonzi Buddy!

 

[Fishbong]

I agree with Material Defender, the Pandora should definitely go x86 as soon as x86 is as mobile as ARM.







:lol:

 

[Svartalf]

Future models of Pandora dont NEED to go x86 (eventually). Pandora 2.0 needs to use whatever makes the most sense.

Heh... Odds on, it'll be an OMAP4 SoC. Same architechture, but much more of it for only a slight loss on the power consumption front. Moreover, some of them offer beefy dual core A9 configs- and that sounds really appealing on a handheld if you're not giving a lot away on power consumption to get it.

If you choose OMAP4, you get to keep all the current apps. Legacy support's good for 1-2 generations- Intel and Microsoft didn't QUITE catch that one. Not to mention, I seriously doubt that anything other than ARM or MIPS is going to fill the role we're going to be looking for. It's certainly not going to be X86. Moreover, all one has to really do to scoop up some semblence of Windows support is dust off Darwine. It's certainly doable.

QUOTE

This will probably not be X86, no matter what you seem to keep telling yourself.



Especially since there's going to be credible answers for most of your X86 needs now and when the next gen shows up- with a Cortex-A9, you can realistically emulate X86 for many situations on the ARM.

 

[Karel Jansens]

Anyway, Im not arguing that the current breed of x86 chips and chipsets are able to take on ARM. I am saying, the battle has just begun and it should be painfully clear who is going to be the winner here.

Who gives a SHIT about winning? Were going to develop our own software to meet our own needs. We get it, the Pandora wasnt perfectly tailored for YOU. Nobody cares. Anyone who actually enjoys playing video games or using a UMPC will tell you they will take longer battery life over the ability to run windows applications for one hour.


The Pandora isnt about winning some imaginary race, it is about meeting the demands of a community that has been drooling over it for years. Maybe it doesnt meet your needs - tough. It meets most of ours.

Although I agree with your basic sentiments, the way you voice them makes it sound as if "Material Defender" shouldn´t be allowed to table his opinions and preferences. If that were the case, there wouldn´t be a community here at all.

I concede that he doesn´t make a very good case -- one that has been put forward before ad nauseam -- and that he represents at best a small minority, but still...

Edit: Sorry. I should have read more of the thread. In my defense, I thought it´d pretty much ended after your reply.

 

[sold]

Dammit, the only reason people keep crapping on about x86 is because of the amount of software that will work out of the box. But don't get me wrong.

IMO x86 based mobile computers may eventually be on par with arm mobile devices . And we may see ARM based desktops with performance levels on par with x86 desktops.

The only thing keeping x86 ahead to date is the amount of software that is available. There will come a time when software developers will be programming for all cpu architectures so they can (within reason) tap into all available markets. Be it desktops, laptops, netbooks, mids, umpc's, gaming consoles, gaming handhelds, mobile phones or freakin' microwaves

Eventually,there may even come a time when devs and publishers decide to release old software for all available platforms commercially, or they may even adopt the open-source mentality for older software that isn't commercially viable.

Competition is keeping things very interesting ATM, but eventually there will be nowhere to go. Hardware manufacturers will hit a wall. And that is when we will see a paradigm shift towards open source software and hardware.

Someone could create a computer capable of being controlled by thoughts and powered by love that could automatically download all the music, games, movies and porno I could think of (which is a lot) while typing up and sending all my emails, writing and printing my letters for me, doing all my bookkeeping while updating my freakin' facebook account AND sucking my balls through my USB 4.1 ballsucker, and it would be completely useless until someone wrote some decent software for it.

 

[Karel Jansens]

I'm also familiar with the Zaurus, which was the first Linux-on-handheld device I owned. Yes, developing for that one isn't that easy. But it's a red herring. For the last two years I've owned a Nokia internet tablet, which is very different from a Zaurus, and this is the type of device the Pandora will be, so I think I have a pretty good idea of how it will be to develop for the Pandora: Developing for the Nokia N8x0 is pretty much the same as developing for a standard Linux laptop or desktop. This is the type of device we're talking about. Not a Zaurus-type of device.

To be fair, Nokia has made a habit of breaking backwards software compatibility on the Internet Tablets several times over and quite a lot of developers have left the platform because of that (just go check the amount of applications that are "stuck" in the 2005, 2006 or 2007 versions).

Nokia treats the tablets as their phones and expects customers to buy a new device, rather than get an OS update for their device (again, go check the number of filed bugs that will be "fixed in Fremantle", meaning the OS for a (hopefully, but not by me) upcoming new tablet.

 

[mali]

Dammit, the only reason people keep crapping on about x86 is because of the amount of software that will work out of the box. But don't get me wrong.

IMO x86 based mobile computers may eventually be on par with arm mobile devices . And we may see ARM based desktops with performance levels on par with x86 desktops.

The only thing keeping x86 ahead to date is the amount of software that is available. There will come a time when software developers will be programming for all cpu architectures so they can (within reason) tap into all available markets. Be it desktops, laptops, netbooks, mids, umpc's, gaming consoles, gaming handhelds, mobile phones or freakin' microwaves

Eventually,there may even come a time when devs and publishers decide to release old software for all available platforms commercially, or they may even adopt the open-source mentality for older software that isn't commercially viable.

That's exactly it
Cortex is the first ARM product to enter x86 territory. The software will follow sales

 

[nicfer]

x86 is the dead end. Soon or not, we'll even see higher end desktops using ARM which will be capable of running Crysis at maximum setting, while x86 being trapped due to the fact that 6.0ghz processors wouldn't make any difference over a 3.2ghz one. Sure, it would be fast as hell but what it could be useful? Scientific use, which I don't think that a normal user will want to run.

One question, is ARM already 64 bits?