Why the next project should be a 8-10" detachable tablet


Ya, I didn’t think about the trackpad. That does make it difficult unless Windows dials in the tablet UI.

For 3rd party keyboards I was simply thinking that Eve could design the tablet in a way that it works with an existing line of keyboards. Or maybe just a common form factor that is used by several companies. That way there would be a readily available option for anyone that needs one.


Not necessarily, Dell did that since they want the display to be directly powered by the eGPU. If you don’t mind having the display wired to the iGPU (no G-Sync), then you could use Nvidia Optimus technology. Surface Book is an example of an Nvidia Optimus detachable.

The problem here is you need to have some sort of mechanism to prevent the dGPU from getting detached while in use. When the dGPU is not in use, you can safely unplug it. The easiest way (but not foolproof) is to put an LED indicator on the gaming accessory to indicate when the GPU is in use.

That being said, I have never seen Optimus being implemented on non-Intel iGPU, so I’m not sure if that’s possible or straightforward as it is on Intel CPU. Technically it should, as Nvidia doesnt care which CPU you buy (e.g no political reason to prevent non-Intel CPU)

Surface Book pulled it off, if we need some inspiration :wink:

Since the MX150 only supports 4 lanes of PCIe (typical GPU is 16 lanes), we only need a fraction of the pins that the Surface Book has.

For power delivery, the power consumption of MX150 is relatively trivial, I’d guess the power bus pins from the PCIe connector itself is sufficient. This point shouldn’t be a concern.


Your right in that it’s possible but the R&D Cost and time goes up and so doest the device cost. Although having said that using a custom pcie based connection will probably be cheaper components wise because you won’t need a tb3 controller or have to pay any royalties if any still apply.

I think for Gen one a Lakeview tablet plus secondary screen is a good starting place. Followed by a gaming controller and some sort of keyboard 6 months down the line . (a keyboard isn’t necessary at launch in my opinion but I guess its down to votes) the secondary screen is pretty easy over usb-c and it will make the device compelling so I definitely think it should be the first accessory


From what you guys are saying, it sounds to me like TB3 is they only solution that really makes sense. The PCIe option sounds like itv would be contrived and not worth the effort. To do it right it seems like this device should only be produced once the proper Intel platform components are available.


If we go with Qualcomm 8cx, eGPU might not be necessary in the first place. (Yes I didnt check the numbers before suggesting the idea, my fault). Plus the idea of TB3 in a $500 10" tablet, when you think about it, is a bit far-fetched.

Anyway, here is the performance results of the MX150, compared to Adreno 540 (Snapdragon 835), compared to Intel HD 615 (Intel Core-Y)

MX150 / Adreno 540 / Intel HD 615

  • 3D Mark Ice Storm: 224588 / 56046.5 / 70743
  • 3D Mark Cloud Gate: 19232.5 / 4259 / 6387
  • 3D Mark Fire Strike: 3511 / 513 / 853
  • 3D Mark11 Performance: 4497.5 / 791 / 1242.5

Now, Qualcomm claims the performance of the Adreno 680 is approximately 3.5x of the Snapdragon 835 (Adreno 540), so I multiplied it to the numbers above to get the estimated performance.

MX150 / Adreno 680 (estimation) / Intel HD 615

  • 3D Mark Ice Storm: 224588 / 196163 / 70743
  • 3D Mark Cloud Gate: 19232.5 / 14906.5 / 6387
  • 3D Mark Fire Strike: 3511 / 1795.5 / 853
  • 3D Mark11 Performance: 4497.5 / 2768.5 / 1242.5

As you can see, the performance figures is closer to the MX150 than HD 615, it could even nearly match the 10W ‘1D12’ variant of the GeForce MX150 (instead of the 25W ‘1D10’ variant) in some scenarios. This is not very surprising, considering Qualcomm generally has the strongest GPU among its ARM competitors.


New Win Tablet/2-1:



Sad to see Chinese tablets going 16:9 again lately. There was a brief period where they all had 3:2/16:10 display


As mentioned in the other thread, it should definitely include wireless charging.


What about removable battery instead? So the device ships with 2 batteries and a battery charging dock. One battery stays in the dock, the other is inside the tablet. When you run out of battery, just swap them out. You go from 0 to 100 in less than 10 seconds.

The advantage:

  • Device is never tethered, this is especially useful for gaming use, as gaming is both incredibly uncomfortable when tethered, and it uses a lot of power
  • Batteries can be smaller --> reducing weight/thickness
  • Battery cycles are split between the two batteries --> reducing battery wear
  • Less heat in general --> reducing battery wear
  • Slow charging --> reducing battery wear
  • Battery is usually the first component to degrade, so removable battery would tremendously help extending the longevity of the device.
  • Gamers/Pros can keep 3-4 batteries for all-day intense use
  • Cost of battery charging dock+battery is about the same, or even less than an extra wireless charger alone, and you get a whole new battery here.
  • You can put those pair (battery charging dock + battery) in places you often go to (e.g work, car), much like an extra charger, so there is always a fully-charged battery waiting.

Ideally it should also have a tiny secondary battery (less than 10 min battery life) on the device so you don’t have to power off the device when swapping the primary battery. Here’s one example:


What would swappable batteries do to the ability to build the tablet with a rating like IP67 or IP68? I would want to use a device like this in the field and I don’t see it being easy to get a waterproof case designed for it. I think it would be just as easy to carry a battery bank as an extra battery. Speaking of field work…LTE?


The Galaxy S5 did, granted its not a tablet, but waterproofing is possible with removable battery. But as far as I know, most mainstream tablets (iPad, Surface) don’t have those waterproof rating anyway, so I doubt its possible/viable here, even with sealed-in battery.

I think it would be just as easy to carry a battery bank as an extra battery.

  • You are still tethered, either with cable, or with bulky case
  • All the battery wear / battery degradation-related points above
  • You still need to remember to charge the battery bank
  • Most battery banks, especially the ones that can charge tablets/laptops are thick
  • Finding a battery bank with USB-PD isnt exactly easy

Speaking of field work…LTE?

This would be easy if we go with Qualcomm platform, as it has LTE (even 5G) built-in. That’s why all Qualcomm Windows tablets have LTE. Not sure about Intel Lakefield.


I agree on all those downsides of battery banks. I just really want waterproofing as well. If a case could at least be designed for the tablet in it’s most basic form then I’d be all about swappable batteries. :grin:

What are your thoughts on housing material? I love metal but it also kills battery life in the cold. Plastic and glass have a real advantage in that department.


Interesting ideas about battery and wireless charging. how about we split the difference between removable battery and battery bank by offering a battery cover that magneticaly
latches to the back. this cover would also support wireless charging and could also support reverse wireless charging so you could charge your phone. a case means no expensive modifications to the tablet an no impact on structural integrity or water ingress. the most significant benefit is no additional cost added to the base device those who want extra battery life or wireless charging can pay a little extra for it but other folks don’t need to pay for features they may never use. also no wires. nothing extra to carry around as it fits neatly to the back of the tablet.

Also bear in mind that almost every accessory (secondary screen, gaming controller & keyboard) will have its own battery that will likely be even larger than one in the main tablet. the reason those three accessories have to have a battery is either because of balance in the case of the keyboard and power requirements in the case of secondary display and gaming controller.

for LTE again we could have have a snap on LTE case, the significant drawback here however is you wouldn’t be able to use two accessories at once so you would not be able to use the gaming controller and LTE at the same time which is kind of silly. so the answer then would be either an LTE sku of the basic tablet or LTE built into every accessory. on balance i think a simple LTE SKU would suffice.


I honestly never heard of battery life issue with metal housing. What is the problem and the cause exactly?

I personally prefer plastic, as it is very flexible, which is the reason why car bumpers are made of plastic. It is also the perfect material for removable back, as you could take advantage of the flexibility property of plastic, instead of requiring a latch in case of metal (e.g Moto RAZR, Samsung Captivate)

But I dont think plastic is going to sell well in the market that obsesses more with the quality feel than actual quality. With that in mind, I choose metal as glass is relatively more fragile.

This is possible yes, but imo it looks like an overcomplicated solution for something that can be solved cheaply using a “technology” that has existed for decades.

Wireless charging in particular imo is the best example of an overcomplicated first-world-solution to a first-world-problem.

Also the battery inside the tablet still needs to be charged, which means the device needs to be tethered regularly still, and some of the battery-wear-related points above would still apply.

for LTE again we could have have a snap on LTE case, the significant drawback here however is you wouldn’t be able to use two accessories at one

We will see which CPU we ended up with. If we went with the Qualcomm platform, we wouldn’t even need to worry about this point at all, as every device will ship with LTE capability built-in (unless disabled by the OEM)


So this is a 8-10 inch gaming tablet? Is that correct? I’m just curious who a tablet this small would be targeted to.
If it’s gaming, what can you do for performance?


use your hat :stuck_out_tongue_winking_eye:


its a multi purpose highly portable tablet. i would say its more multi purpose than any tablet currently on the market. depending on the configuration it can function as the following:

  • A basic 8inch ish windows 10 tablet that is affordable ($400 ish). main use content consumption and light productivity whilst on the go
  • With a controller add on it becomes a portable gaming machine similar to a nintendo switch. main purpose is gaming on the go, performance is dependant on cpu (intel vs ARM) and the potential for an external mx150 GPU in the controller.
  • with a second screen add on it and become a courier like device, we will have wait and see what Microsoft’s software plans are for dual screen windows devices but i suspect they will be all about productivity and multitasking. the argument will probably be similar to using dual desktop monitors as it increases productivity if you can do two things at once. (microsofts centaurus devices is rumoured for fall 2019)
    -with a keyboard addon it becomes a mini laptop with extended battery life.
  • with a battery addon it can become an endurance tablet for travelling on long haul flights etc.

depending on demand and support from Microsoft it also could become:

  • with a speaker dock addon it could become an echo show like device.
  • with a eink cover it could become an ereader.

this device has the potential to be many things to many people.


I wonder if we can have a crowd-developed app (read: open source project) for an Alexa or Google Assistant app. It is possible to access them as a third party, so all you need is basically just a pretty UI and perhaps plugging it to Windows APIs and/or third party (e.g Philips Hue)

(note the app will be optional, so if you are paranoid about privacy, this app will never touch your device unless you specifically install it)


I was hoping for the Cortana home hub but it looks like it may have been canned, hopefully t was delayed as part of core os. We first heard about home hub two years ago now.

An open source initiative sounds interesting, sounds like a lot of work though and you’ll need it to work above the windows lock screen which is probably impossible.


If the tablet supports display-in, we could have Google Assistant embedded to the dock (although it won’t be a “dumb stand” anymore).

When there is no tablet docked, it will operate as a Google Home / Amazon Echo type of device, powered by Google Assistant. When you dock the tablet in, and the tablet detects a display-in, it will automatically prioritize the external display, which then turns the whole package into a Home Hub / Echo Show-type of product. There could be a hardware switch on the dock to toggle the display output, which allows the tablet switch back to Windows.

That being said, its probably much simpler and easier to simply add an LCD panel to the dock, and I just realized this is a stupid idea.