Starting a Project

[dominic777]

Hey guys,
Seeing that mobile (phone) processors are getting much more powerful, Me and my friends are thinking about starting a kick-starter project.
The concept is we would use a Intel atom z3770d, 64gb micro SD, 6000mah battery, and a smartphone motherboard and ram parts from china to make a small box that can attach to phone, monitors etc... and run windows 7 or 8. (preferably 8).

1. Would the hardware be strong enough to run windows 8?
2. I am confused as to how to make the device connect to a smartphone and use the smartphone as a screen and run windows 8.
3. If this does take off, I want to supply it under 250$. Too low? I am aiming for people who want a extremely portable computer but for cheap.
4. Could we be using a smartphone motherboard to run windows 8, or would we have to design our own?

PS: While I am not new to computers, I am new to mobile technology, although I am not a complete noob... Please be nice!

Thanks in advance!

 

[Someone Somewhere]

A few comments:

SD cards have very few P/E cycles, and are slow. You'll wear it out quite quickly. For mobile solutions, eMMC is the usual option, though PCIe/SATA Express is gaining traction.

You almost certainly can't use a cellphone board. However, an Intel NUC board might work ( http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-D33217CK.html or http://ark.intel.com/products/78953/Intel-NUC-Kit-DN2820FYK ). Haswell versions exist, but are pricier.

Using a phone as a screen is going to be an interesting task. However, via WiFi, you should be able to get VNC/screen sharing going. Just remember that a tiny screen isn't going to work great.

 

[dominic777]

A few comments:

SD cards have very few P/E cycles, and are slow. You'll wear it out quite quickly. For mobile solutions, eMMC is the usual option, though PCIe/SATA Express is gaining traction.

You almost certainly can't use a cellphone board. However, an Intel NUC board might work ( http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-D33217CK.html or http://ark.intel.com/products/78953/Intel-NUC-Kit-DN2820FYK ). Haswell versions exist, but are pricier.

Using a phone as a screen is going to be an interesting task. However, via WiFi, you should be able to get VNC/screen sharing going. Just remember that a tiny screen isn't going to work great.

NUC board seems a bit too big... is it possible that something smaller exists?
I would like to be able to attach it to a back of a phone as a case, or at least small enough to fit in a pocket and hand.
I could even use Tegra processors since they are small if they are powerful enough to run windows 8, which I think they are.

 

[ssddx]

the real question is.....

what could this product possibly offer that you cannot get in a windows 8 tablet already?

essentially that is what you're trying to jury-rig together.

 

[manofchalk]

1. Running it will be fine, though I think Win7/8 wont be viable for other reasons. Storage requirements are one I can think of, a fresh Win7 install is something like 20GB.
2. You will probably need to develop software so that it can accept a video feed through the USB connector and pass user input back to your device. Its not that uncommon a thing already, just typically its done via wireless solutions.
3. No idea, but I think it might be too low.
4. Again, no idea. I imagine the chip and whether its X86 compatible is what matters, but I just don't know.

I think Tegra only works for Android at current, dont think they are x86.

 

[dominic777]

the real question is.....

what could this product possibly offer that you cannot get in a windows 8 tablet already?

essentially that is what you're trying to jury-rig together.

Yes, but I could possibly allow people to have one for a very cheap price, while other tablets go for above 300-400 for new. And it is not limited to only 1 tablet, it could connect to any phone, tablet, monitor you have and run it. And at under 200, and possibly at 100, this will be one of the most affordable computers out there.

 

[ssddx]

what makes it better though?

-if the user already has a smartphone why do they need such a device? sure you can get it to run windows however if windows was such a big deal the user would likely already have a windows smartphone and not need your device. also android devices already have software compatible with things such as msoffice, adobepdf, etc which eliminates the need to run windows on the device.

-for $100-200 the device will be very underpowered so what is the advantage over other devices such as an intel nuc which are already very budget oriented mini computers?

-the device would require you do connect to a phone via wifi or bluetooth (which will have lag) or would requrie somehow working with usb but that route may be plagued with issues. it would also require at minimum hdmi out. it would also be a bit large to strap onto a phone and honestly i dont see what you would gain by doing so. keep in mind that the size of cell phones is quite varied so how will it attach exactly?

-netbooks and desktops are both available in the $200 range already featuring intel atom processors. what could your product offer to compete with these assuming you cant beat them on price.

-keep in mind that since its a startup project your prices will be much much higher and you wont likely have the same sort of deals on components that large companies do. in short your products will need to cost more not less so i dont think they can compete on a price to price basis. your product would need to offer something novel, something new that isnt on the market already.

 

[dominic777]

1. Running it will be fine, though I think Win7/8 wont be viable for other reasons. Storage requirements are one I can think of, a fresh Win7 install is something like 20GB.
2. You will probably need to develop software so that it can accept a video feed through the USB connector and pass user input back to your device. Its not that uncommon a thing already, just typically its done via wireless solutions.
3. No idea, but I think it might be too low.
4. Again, no idea. I imagine the chip and whether its X86 compatible is what matters, but I just don't know.

I think Tegra only works for Android at current, dont think they are x86.

what makes it better though?

-if the user already has a smartphone why do they need such a device? sure you can get it to run windows however if windows was such a big deal the user would likely already have a windows smartphone and not need your device. also android devices already have software compatible with things such as msoffice, adobepdf, etc which eliminates the need to run windows on the device.

-for $100-200 the device will be very underpowered so what is the advantage over other devices such as an intel nuc which are already very budget oriented mini computers?

-the device would require you do connect to a phone via wifi or bluetooth (which will have lag) or would requrie somehow working with usb but that route may be plagued with issues. it would also require at minimum hdmi out. it would also be a bit large to strap onto a phone and honestly i dont see what you would gain by doing so. keep in mind that the size of cell phones is quite varied so how will it attach exactly?

-netbooks and desktops are both available in the $200 range already featuring intel atom processors. what could your product offer to compete with these assuming you cant beat them on price.

-keep in mind that since its a startup project your prices will be much much higher and you wont likely have the same sort of deals on components that large companies do. in short your products will need to cost more not less so i dont think they can compete on a price to price basis. your product would need to offer something novel, something new that isnt on the market already.

Correction: After some research, I believe I can provide it for less than 90$. If you ask me, that is incredibly cheap for a computer. A usb connection to a phone will not have as many problems as you say, In my opinion. Say if a person has an ipad, or a samsung tablet. Unless it is a windows operated tablet, they may wish that they could use a windows tablet. But, since it costs more than 200-300 for it, they may want to get a bit less powerful add-on for only 90$. Also, things like intel nuc is not-so portable, and cannot be connected to tablets, and it cannot fit on your palm. I searched, but a computer this size for this price does not exist yet. And I am targeting cheap cheap prices.
I am thinking about the ODROID-U3, which includes RAM and a decent processor for only 65$. All I need on top of that is a wireless card and eMMC.

 

[ssddx]

you keep mentioning about cheap prices.

what kind of hardware can you possibly include for such a cheap price and how exactly are you going to be making a profit after all the overhead costs of r&d and programming? designing a new piece of hardware isnt cheap.

also did you give thought to the price of the windows license you will need to include with the device?

i just dont see how you plan on making the device both "good" and "cheap" while still actually making any cash on the venture. if big brand names havent done so then how are you planning it?

how is your device better than a $200 netbook with win7 if competing at the same prices (i think your pricing plan will double or triple quite easily with costs involved)

i'm not saying its a completely terrible idea i just A) dont see a mass market need for such a device B) dont see how you can compete with larger companies on a price to price basis without having crud quality and C) manage to make it work across all devices without some rather major hurdles to overcome.

do what you want. just saying that unless you absolutely know what you're getting yourself into you could be in for a rude awakening.

 

[Someone Somewhere]

The issue is that ARM chips don't run Windows. They can run Windows RT, but that's a whole lot more locked down, and IIRC you need microsoft's go-ahead to release a product with it.

Also, there's a substantial lack of bandwidth to move a decent resolution screen across USB2.0. Don't expect FullHD video to look good, even if you get it working.

You'll probably have to write your own protocols for display sharing over USB. And compressing/uncompressing video over USB in a phone, in software, is not going to work well. It's touch and go (and gets quite hot) on a DisplayLink box with custom silicon. A cheap android phone with a dual-core chip isn't going to do much.

The most I think you could reasonably get going is a box that sits in your bag and uses an ad-hoc WiFi connection to a screen-sharing app on a phone/tablet.