GPD WIN 2


The strong point of the Pyra was never the CPU. Even when development started, there were faster ones available already.

The Pyra is being planned as an upgradable BASE system. The reason for that?
Well, we had to reinvent the wheel. The Pandora was nice, but the whole system has been developed around the OMAP3.
Using a new processor means a full redesign - and you can clearly see with the Pyra what that means if you've got a small development team involved (and not a multi-billion-dollar company like GPD), it takes a LOT of time!

That's why it's important to get the first unit right, so upgrades can be done a lot faster.

The main idea of the Pyra is still to have a good and usable base system, as open as possible (hardware and software-wise). It should be easily moddable and hackable (add a camera, etc.).
The main goal is NOT (and never was) to have to the fastest device available.
The case design files, schematics, etc. will all be opened up.
Want to change the case for some mods? Sure, change the design and print it.

Another thing that's important for me is the removable battery, for example.
AND the quality of the battery.
Mind you, there are Pandora batteries produced 2008 that haven't even lost capacity yet. Most of the ones from 2013 all still work like fresh new ones.

Whereas I already have quite a few customers complaining that the GPD Win which is only a year old have lost a lot of capacity already (or the battery even grew in size).
And GPD hasn't figured out a way yet how to send me replacement batteries for the customers...
If it's similar with the GPD Win 2, then it could very well be you can't really use it for more than 2 years.
We don't know yet, but it could happen.

The batteries for the Pyra have been produced by the same company as the ones for the Pandora.

The GPD Win2 is a powerful pocket computer, yes, and I guess there's no one here who denies it will simply kick Pyras' ass in terms of power.

But when it comes hacker friendlyness, openness, flexibility, etc., the Pyra easily kicks the GPD Win 2 away.
It all depends on what you're interested in and what you're planning to do with it :)

You pretty muched summed up most reason I am still waiting for the Pyra instead of getting anything else. It has been some tempting to switch gears but just like with the Pandora it's worth the wait.
 
I do find it a little sad that threads about new devices tend to turn into a pissing match between the PYRA and them. For a tech enthusiast like me, I just appreciate things for what they are.

The PYRA and WIN2 are different, they do have there points over each other, so just go with what suits you is the obvious advise there.

P.S In regards to GPD, just to clarify, they are not a billion dollar company - not even remotely close to that, they have only just grown to 30 staff last year. They operate out of a crappy office in Schezhen and are basically a design house, outsourcing all manufacturing. I should add that GPD have been testing batteries from a variety of manufacturers to get it right this time for the WIN 2, so hopefully that turns out well. I think the core issue with theb GPD WIN 1, was that the battery was jammed in there
 
I do find it a little sad that threads about new devices tend to turn into a pissing match between the PYRA and them. For a tech enthusiast like me, I just appreciate things for what they are.

Well, I guess that always happens, as there'll always be people who try to compare it :)

The PYRA and WIN2 are different, they do have there points over each other, so just go with what suits you is the obvious advise there.

Yep, same with the GPD Win back then, and nothing has changed since that :)

P.S In regards to GPD, just to clarify, they are not a billion dollar company - not even remotely close to that, they have only just grown to 30 staff last year. They operate out of a crappy office in Schezhen and are basically a design house, outsourcing all manufacturing. I should add that GPD have been testing batteries from a variety of manufacturers to get it right this time for the WIN 2, so hopefully that turns out well. I think the core issue with theb GPD WIN 1, was that the battery was jammed in there

:D Sorry, I meant "million", that was a typo :)
And that's a lot more realistic (as they have a couple of million USD on their hand).
And more than 10 times more we have available for the Pyra :)

Well, for the batteries, it's a bit of hit and miss.
I'm not sure you can really test it upfront. Even most smartphones' batteries don't last longer than two or three years, even if they're made by major companies.

The Pandora battery was definitely a very lucky hit, I have never expected something like that... all I knew is that batteries shouldn't last longer than 2 year, but the Pandoras simply work and work and work.
So it was pretty clear we won't change the company here :D

Glad to hear they're looking into better batteries as well :)
The XD also has some issues, though they don't seem to be as bad as with the Win :)
 
Ive seen user mods where people wanted their GPD Win1 devices to run cooler than the hot temperatures they end up running at. Also with the Battery swelling/failing issues I'd wonder about the quality of other parts.


Anyone other than myself worried they wouldn't get long-term use out of this product because of quality issues of past products?
 
Anyone other than myself worried they wouldn't get long-term use out of this product because of quality issues of past products?
That's the main concern for me with the GPD Win2. I don't want to judge them before they deliver but I got a very bad first impression when trying the GPD XD of a friend. He was still happy with his device but mostly because it was quite cheap. That said, it seems like people are happy with their GPD Pocket so far and praised the build quality. This is why I want to believe the Win2 will be a win on the build quality front as well. We'll see.
 
GSM isn't crap where i've lived in the US (i'm on t-mobile), but i am concerned about the network letting me put a pyra on it. i'll maybe have to go bug people in person, but i'll let the community know.
When I asked in person and have some basic specs to tell them it would work on their network they said I could use it, all I had to do was buy the SIM card. I am not exactly convinced since their job is to sell stuff, and I though there was more that needed to be done when adding things. Since there are a handful of us looking at using T-Mobile maybe we should get together so that they can see it isn't a single individual (maybe they will pay more attention to us) and we can get real answers from someone on the technical side. It would really help if someone had a Pyra in hand (@Grench, would you be willing to help when you get your prototype, even though you will be using a different company?). The state of software at that time might also be a hurdle.
 
If the GPD Win 2 will be about the Size of the Win 1, hear I got a Picture to compare divergence between Pandora and GPD Win (side Note: im currently using Pandora again for every day use)
9eb50d8609d8e7a3029dae123d52e9d6.jpg
e7f3807c8ab8e3ffbd0aba709af3722f.jpg
 
That's the main concern for me with the GPD Win2. I don't want to judge them before they deliver but I got a very bad first impression when trying the GPD XD of a friend. He was still happy with his device but mostly because it was quite cheap. That said, it seems like people are happy with their GPD Pocket so far and praised the build quality. This is why I want to believe the Win2 will be a win on the build quality front as well. We'll see.

They certainly have improved the quality with each release, which is great, so the GPD Win 2 will probably be a lot better than the XD or Win.

I also plan to update the Pyra as good as possible (case changes, etc.), keeping it all compatible to the original one though so that everyone can simply grab a new case or whatever and update it :)
Useful community ideas or hacks welcome, of course. So once released, let's all try to improve it as good as possible :)
 
Unless new cases are more brick-like I would want to stick with the original. Changes in ports might require me to change, though.

I would like to see people that want things like cameras get what they want. Although I know many want x86, I don't think I am in that camp anymore. On the other hand, it would help sales, even if it ships with Linux (it would, right?).

Let's just get the first iteration of the hardware and see how we feel about it. Later we can try to get it to compete with GPD devices if necessary.
 
Although I know many want x86, I don't think I am in that camp anymore. On the other hand, it would help sales, even if it ships with Linux (it would, right?).
The advantages would be Linux closed soure games (Steam) like Rocket League and many indies. And foss video driver. These 2 points are big, I was surprised recently after installing Steam on my older Linux laptop that a fair share of my library (I mostly game on a recent Windows tower PC) was actually available on Linux as well !

Not sure what the drawbacks would be. But it seemed like we couldn't source an Intel CPU anyway. And I think selling an Intel powered Pyra without Windows support would be something most potential buyers don't understand. Then we would get request after request to get Windows working even if the device was not designed with that in mind.
 
Last edited:
I was surprised recently after installing Steam on my older Linux laptop that a fair share of my library (I mostly game on a recent Windows tower PC) was actually available on Linux as well !

You can also install Steam under Wine on Linux and get access to the vast majority of the games that don't show up in the Steam for Linux menu if you need to fill in some of the gaps.
[doublepost=1514916575,1514915782][/doublepost]
It would really help if someone had a Pyra in hand (@Grench, would you be willing to help when you get your prototype, even though you will be using a different company?). The state of software at that time might also be a hurdle.

We'll see how the time and space things work. If it is going to be weeks between prototypes and mass production, it might not be worth prototypes traveling. Space - or at least connecting points in space - could be an issue. Prototypes, people, sim cards & cell towers - they're unlikely to all be in the same convenient location.

We still don't know who the other prototype purchasers are. They have been very stealthy. There may very well be more than one of us in the US. Having a $1700 prototype travel around the country is a bit more onerous than having a case & keymat mule travel about. There is the obvious expense of insurance and process of assurance, but there is also the concept of giving up a new toy before I have a chance to play with it much. We'll categorize this as a, "lets see where we're all at in time & space then."

There are a few things that have to happen before... Prototypes have to be built, tested, shipped, received, etc...
After, I will have to get the prototype running on a carrier. I have an AT&T data only SIM that I plan to use as a starting point. Have to start somewhere. Getting there on this is going to require getting the software side of things working at some basic level as well.

We're still a long ways out from needing to travel a prototype to test on additional carriers.
 
If the GPD Win 2 will be about the Size of the Win 1, hear I got a Picture to compare divergence between Pandora and GPD Win (side Note: im currently using Pandora again for every day use)
As I understand it, the Win2 will be slightly larger than the Win1 due largely to the larger screen. I think it also has larger keys, because there appears to be space there to fit an extra layer of keys above the keyboard of the Win1 so that that right hand bar could be moved there if it were all designed in, and yet on the Win2 the game controls are much more cramped against the hinge than they are on the Win1.

But thanks for comparing the win1 against the Pandora. They're both previous generations of what most of us are talking about, but it's a worthwhile starting point.
 
You can also install Steam under Wine on Linux and get access to the vast majority of the games that don't show up in the Steam for Linux menu if you need to fill in some of the gaps
Aye, you can, but how well they work is a different thing, and as always, ymmv
 
The Win 2 seems like a really great device and, if I was only a gamer, I wouldn't bother with the Pyra anymore. But since I am looking for a mini pc with Linux, it is more important to me to have great input methods (I already trust the Pyra keyboard a lot more than the Win 2 keyboard) and long battery life. That said, I hope that there are really no compromises on desktop usage with the OMAP5 (including stuff like 720p60 video playback for streams within the browser).

I will definitely get a Pyra and probably also a Win 2 for steam gaming. I just hope that we will receive the Pyra soon. I am a second batch preorderer (number 6xx) and I hope that it doesn't take another 6 months after the first batch to receive the units. Anyway, I am very excited for 2018 in terms of hardware.
 
I think it also has larger keys, because there appears to be space there to fit an extra layer of keys above the keyboard of the Win1 so that that right hand bar could be moved there if it were all designed in, and yet on the Win2 the game controls are much more cramped against the hinge than they are on the Win1.
For the Win1, Pandora and Pyra, the physical keyboard design, key count and placement were pre-ordained and the mapping argued/decided later.

I suspect that the Win2 keyboard benefits from a mapping first design decision. A distinct surface area was likely dictated for the keyboard contacts, but not physical keys. The key count, layout and positioning could then be decided by mapping and scaling. The result would then be fed back into board design for the contact pad placements. Since there was no keycap surround (non-island keyboard), plastics could progress even while physical key placement within the surface area was in flux.

I think it frustrated a lot of people in the Pyra keyboard threads that the count, size and physical locations of the keys were hard-fast declared before any discussions of logical layouts began.
[doublepost=1514927483,1514927162][/doublepost]
I am looking for a mini pc with Linux
And that is the two main bits that turns me off the Win2. Opaque answers about Linux support and the "Windows" in the product name. I dread purchasing anything with i* or Win* in the product name. Both feel deliberately deceptive - like marketing through theft of branding.
 
Opaque answers about Linux support and the "Windows" in the product name. I dread purchasing anything with i* or Win* in the product name. Both feel deliberately deceptive - like marketing through theft of branding.

Then you'll be certainly looking forward to GPD's next offering, the iWin Pi 3.
 
https://www.theregister.co.uk/2018/01/02/intel_cpu_design_flaw/

While not GPD's fault there's a bug in the proverbial ointment that just showed up.

tl;dr:
Basically security features associated with the security of kernel memory that should be handled on die will need to be handled in software for both Linux and Windows to protect against an exploit associated with kernel memory access in speculative execution. This is expected to result in a rather significant performance penalty of around 15-30% associated with the memory paging and software mode penalties once the relevant security patches are pushed in the next month.

AMD indicates they're uneffected by this exploit. In turn Intel's going to need to delay Cannonlake for a good bit if they want to fix this issue prior to bringing Cannonlake to market, as this issue has to do with parameters associated with how the CPU operates.
 
Last edited:
Ouch. I really hope Intel fixes that in future CPUs or else we could end up having the fix running even when not needed (as AMD user for example) with people claiming "it would be too much work to maintain both variants".
 
Back
Top