Starting an x86 handheld project

[3_bit]

Steam for ARM Linux - wow they should do that.  Steam for Linux X86 works very nicely - and the thought of playing HalfLife in OpenGL on a Pandora 2 is very cool.

Keep dreaming. It's not going to happen unless Valv Open sources the code OR the android gamers demand it. Valv probably thinks Razer has mobile gaming covered with the Edge. Why port HL1 when you could just recode the whole thing and keep the current level and file save system and probably save time?

 

[kuru]

Atom or similar in Pandora form factor? I will throw money at it.

 

[Binky]

I'm thinking that detailed specs may be the wrong place to start...

 

[2bit]

ever considered just making one for yourself ( see if it is pratical ) if it custom order , then if the demand is big enough kickstarter ?

 

[monstercameron]

I nearly missed this thread. the only viable x86 chips would be:

intel atom medfield/clovertrail/clovertrail+ - arm level performance and power consumption

amd gt16r - low cpu power, decent gpu

the amd z-60 - higher power draw, decent cpu, decent gpu

or wait a few months

silvermont/valleyview - opensource gpu driver

amd temash apu - very high gpu performance

for diy maybe modding a netbook or similar is the best route. As for building a handheld, I just want to remind the op of the p1's journey, it takes more than a bunch of guys on a forum to build a shipping product. It would seem, the most likely route to take is to save/borrow/kickstart a bunch of money to build a prototype, then save/borrow/kickstart a bunch of money to promote, develop for and maintain such a product.

maybe you could do a kickstarter campaign about a video of you building the prototype to fund the actual building of the prototype.

 

[elwing]

I wish you good luck. I won't invest money on yet another handeld even if a x86 one would be nice...

 

[Exophase]

In other words, if you need a small form factor motherboard available off the shelf you're out of luck. A real design is going to take some level of money, experience, and a willingness on the part of the SoC manufacturer or a distributor to sell you the part.

 

[monstercameron]

here are some other pico itx mainboards

http://www.axiomtek.com/products/ListProduct.asp?ptype1=202&ptype2=204&ptype3=137

dont think you can just order one though from them directly but maybe from amazon

http://www.amazon.com/Axiomtek-PICO100-G-Series-1-0GHz-Audio/dp/B007P5ZS5U

 

[Exophase]

Even that N2600 board needs a 13.5W supply. While not as bad as the quad Nano that's still pretty serious. Also needs an external boost regulator to power off of a battery..

 

[monstercameron]

maybe you can start by modding an asus eeepc 701 or some other cheap netbook.

netbook-tablet mod http://www.mydellmini.com/forum/dell-mini-9-hardware-upgrades/13639-dell-mini-9-tablet-project-complete-lots-pics.html

netbook gaming http://liliputing.com/2009/11/liligaming-5-recent-netbook-friendly-games.html

netbook dual mod http://www.slashgear.com/harlequin-dual-touchscreen-netbook-mod-is-diy-courier-0965511/

netbook snes mod http://www.forevergeek.com/2009/06/snes_netbook_mod/

netbook phone mod http://liliputing.com/2009/12/forget-tablet-mods-turn-your-netbook-into-a-huge-cellphone.html

replacing the panel might be difficult so maybe portable hdmi or usb monitors could work.

http://www.mimomonitors.com/products/mimo-touch2

http://www.amazon.com/Lilliput-Um-70-Monitor-non-touch-Screen/dp/B000AYHY7I/ref=cm_cr_pr_product_top

http://www.amazon.com/Lilliput--camera-Component-Composite-Included/dp/B0062FA55M/ref=sr_1_13?s=electronics&ie=UTF8&qid=1361515290&sr=1-13&keywords=hdmi+viewfinder

http://www.tvlogicusa.com/product/product.php?model=VFM-056W/WP

 

[Jonttu]

Haha, well thanks for the replies guys. I'm thinking I will have to postpone this project a little. Really I just want to get some input from others on a project like this and as well as see if anyone else is interested in doing this project with me. I really bought the nano board just to test the platform, but at some point I will try to make it portable. 

 

[hmc]

Well, the idea is interesting, but after trying the UMID M1 and then the Viliv N5, I am not too optimistic about your project. For me, the only advantage of an x86 platform of that size is being able to sync Outlook data from work into a "mobile" Outlook. That is whai I miss on the Pandora. But in general, an ARM platform is much more suitable for a small mobile platform.
I am not a hardware expert, but my impression is that with an x86 board you have to care about heat and size of components so much, that other aspects such as ergonomy and overall quality need to get into the background, making the device less usable after all. At least the was IMHO the case with M1 and N5.

However, maybe a private project would be much more successful than those industry projects. Who knows. I recently have seen such a case, where a bunch of "normal" people have developed and built a very, very nice ARM-based Linux gaming console that is even suitable for productivity, and that's better than any industry based device I have seen recently. ;-)

I will be glad to provide tipps and experiences, if you start such a project. But as I said, I am not an expert, so I am not able to assist much.

You may have a look at my homepage. I have some pages about the Viliv N5 and UMID M1 there, that may give some ideas to think about.
http://www.hermocom.com
Either search the site for N5 and M1, or browse the knowledge base's UMPC section.