GPD Win (x86 Computer / Palmtop)


It wouldn't surprise me in the slightest if they took their initial design to fabrication right away, then when promoting it try and give the impression that it was interactive to gain good relations.

They listen on major requests, we did ask for x86 and clamshell with a keyboard, and they are delivering on that. But they were too quick to push their concept to production it seems and ignored the feedback from their concept.
 
They are making it physically bigger, slimmer, adding a keyboard, adding ports, different pcb, bigger screen, etc.
The same things as the pyra did. Things that effect the case mold.

We weren't talking about the molds. We were talking about the DESIGN. The CAD files.
And changing the size of a CAD file when you already have a similar design can be done very quickly.

We couldn't do that. The original case design files were made by DaveC, he owns them, and the format he uses is a completely different than the company in Greece.
Which means: They couldn't just use the existing design files and tweak them, they had to completely redo them.

Maybe it is better understandable for some here when talking about software:

Let's say there are two similar games by two companies and they both want to release an updated version with some enhancements (replace some graphics, add small new features, etc.).
The current version of thoses games represents the Pandora (or the GPD XD).

The new version of the games with some small modifications is the Pyra (or the GPD Win).

Now let's assume the sourcecode for the game for the Pandora is not accessible.
To release an updated version, the full game needs to be recreated first. Completely. From scratch. If it's a complex game (and the design IS complex), this will take years.

The other company (GPD XD) has the full sourcecode available and can fully reuse that, adding some new features. This can be done very quickly. So their new game (the GPD Win) can be created within a few days / weeks.

Note: I'm ONLY talking about the case design, not the PCB, not the molds. Just the case design.

If you compare the GPD Win case and the GPD XD cases, they are very similar. Some sizes have changed, some holes have moved, there's now a huge cutout for the keyboard.
But those are all simple changes that can be easily be done within a few hours, once everything has been decided.

The only more complex thing here is the keyboard. And the keyboard apparently is not finished (it's just a block with some dividers, you can't press any key, it's just a dummy).

If you know how CAD works, what they are showing here is nothing special nor was that lightning fast, it was just a normal turnaround time.
 
@ED

I have to agree with SNESFAN. The WIN looks nothing like the XD. Being bigger is just the tip of the iceberg in terms of differences. The case is a completely different design.
In fact I have never heard anyone on Dingoonity say it was similar, rather thoughts have been put out there on what other device they may have got that design from or whether they came up with it totally themselves.

For example I have heard people say it looks similar to a chinese electronic dictionary.
 
Last edited:
I've taken the GPD XD apart, and when I compare the GPD XD case to the posted GPD Win case, they look very similar.
What is there that you think is a lot different?

The only change to the backside is a bit different roundness, size and angles.
That's two or three clicks in a CAD program.
The shoulder buttons are a bit different, the hinge is different, the holes in the backside are different... that's about 1 - 2 days work-

That's the backside.

The keyboard side has the exact same DPad and nub cutouts, they only added a recession here.
Then there's a huge cutout for the keyboard, but that can be done within a few minutes.

Well, the display are hasn't changed much as well, except for the hinge.

Don't forget:
Changing sizes, roundness, angles, etc. is just a few clicks in CAD.

And to begin with, the GPD XD has an incredibly simple case. The backside is so simple that even I could design it within 2 days from scratch (and I only know a bit of CAD, only enough to make simple designs).
Don't get me wrong, it's not bad having something simple. In fact, it's great, it saves costs, you can't do much wrong and it can be done fast.

I don't want to make the case design bad, I just want to point out that it wasn't ultra fast, it was done in a normal speed for CAD, for such a simple design.
 
We couldn't do that. The original case design files were made by DaveC, he owns them, and the format he uses is a completely different than the company in Greece.
Which means: They couldn't just use the existing design files and tweak them, they had to completely redo them.
And if started from scratch, why no complete redesign of the Pyra? It just looks to Pandora-ish to me with all the "brickness" and so on. Even without changing the overall size, there could have been some more "streamlined" alterations here and there imho. Especialy now that the Pyra case seems to turn out thicker than needed (battery/underside). All the possibilities, now it's to late of course but still makes me sad somehow.
The GPD WIN may be worse at the end compared to the Pyra but at least it does not scare people away with it's design. ;) To sell devices to a wider audience outside of the hardcore fandom, you need at least a minimum of "I WANT THAT" factor floating into the design. ^^" If done right, you could even sell the biggest bullshit, just with good looking design at the outside, ask Apple. xD
(Pyra is no bullshit of course, don't get me wrong, it just deserves an appearance that shows what a great device it is.)
 
So everything is completely different except for the dpad/nub cutouts - therefor it's much the same. Got it.

Sorry , but IMO your drawing an extraordinary long bow ED. I understand your point that it might be quick for a professional to design a case from scratch , therefor it wouldn't take long, and also that its quicker to work off something existing rather than nothing.
But to say the cases are very similar is a stretch. If it takes one - two days for a proffessional to design a simplish case as you said, then to say it would only require a professional 2 days to mod the XD case to a WIN case is not saying anything. I have seen the internal aspects of the WIN case, it is very different to the XD, (and that goes for the PCB shape also - different point , but just to note it). The bottom half also has a lot more going on than just rounding as you can see in these pics (https://c1.iggcdn.com/indiegogo-med...it,w_620/v1456715118/t99mydlwpwgj6jo5krvl.jpg https://c1.iggcdn.com/indiegogo-med...it,w_620/v1456715160/cpyxoxt7k3fv4qbwe97t.jpg https://c1.iggcdn.com/indiegogo-med...it,w_620/v1456307838/zabesujsqbnawxcgqlwb.jpg) . You will also notice the angled lip on the face which means they dont need the bumper they had on the XD.

As I said, the rumor is that they got the completely different design from a chinese electronic dictionary , probably the Besta.
 
Last edited:
It seems a little like they'd done the design, started prototyping, then had a meeting about how to sell it. New ideas tabled were: crowdfunding, and community feedback to the design. So Peter, Paul; you work on those. Jim, you say mass production is eight weeks from now? Nice.
 
@SNESFAN and @vcoleiro1 , take Ed´s word for that, it is as he described.
Redesigning the pyra case based on pandora, even with one at your hand or an incompatible file, is a huge time consuming thing compared to have the proper files of ´similar looking´ devices.
Design is more then the ´looks´.
 
I'm not doubting it was a huge undertaking by ED and his outsourced company to take an existing device recreate it in CAD it probably was very time consuming to recreate it millimeter by millimeter for a new application. I work with architects, I hear their complaints when upgrading or changing software.

But once the pandora was recreated, how much additional time was invested to make the changes to create the pyra off of that? That wasn't a question I was previously asking, but I'm genuinely curious now.

My complaint is only the belittling of the amount of effort on GPD's end, that's all. Suggesting all the changes from the XD (if it was indeed the base) could be done in minutes or hours seems slightly discrediting and at the same time suggesting the time and effort into making the pyra modifications (not recreating the pandora in CAD) was this huge endeavor when the changes appear to be similar scales.
 
I didn't think anyone was belittling anything. ED was explaining, as he said several times, that the case was not made "fast" but in exactly the expected amount of time based on what they had and were doing. In other words, if you have something to start with it will take less time than starting from scratch. In case anyone can't follow that, go back and look at where this first came up.
[doublepost=1458707173,1458706895][/doublepost]Actually, here, I will do it for you.
That's incredibly fast. Suspiciously fast but I don't know why.
It's not that incredibly fast.

They're working on the design since November, it's VERY similar to the existing GPD XD, the pieces are very simple and the only thing that's really new (the keyboard) is not designed yet, that's just a block that shows where the keys are, it cannot be pressed.
I'm not even sure they're plastic injection yet, from the pictures it looks like a rapid prototype.

So yeah, those can be made from an existing design within a few days.
I'm also sure it's an 3D print and based on a altered existing design. But the Pyra case is also based on the Pandora design and the greek company needed many months to alter everything into the current Pyra case, making the actual molds not even counting.
No, it isn't. The Pyra case was a complete recreation, we didn't have any of the original design files for that.
It's been designed from scratch, based on Pandoras design.

GPD HAS the original GPD XD design files. Changing measurements or moving some things around (like the cutout for the DPad, the nubs, etc.) can be done within minutes if you have the proper design files.

My wife does CAD, so I know how that works.
You've got some reference points and the CAD software changes everything else itself. You want to make it 2cm wider? Just change the parameter and the CAD software will adopt everything else that relies on that.

So they didn't really have that much work to create the design.
In other words, don't judge their timeframe by how long it took to do the same for the Pyra, as they had something to work with, even though they did make changes.

Edit: I screwed that up. I will fix the quoting errors above, so give me a few minutes.
Edit, the second: I think I got it right now.
 
Last edited:
ok, maybe I am looking at it weirdly then. Me and you read the same posts. I pulled from them the impression that ED is suggesting that they took the XD design, point and clicked a couple times, and an afternoon later you got a GPD Win, few days later you can have a printed case to demo.

If ED is saying it is quite the complex thing to make those changes, then I apologize.
 
You need to look at what he was responding to, as well, not just what he said. Ultimately, what he was saying was that the process for making the Win case, if it is based on preexisting designs from something like the XD, takes less time than creating the Pyra case based on the Pandora case but starting with nothing. Yes, he was saying it was simpler. This is not belittling, it was simply to say that you can't compare the two and expect them to take the same amount of time. He didn't even mention all the little tweaks to the Pyra that required redesign, printing, testing, etc., such as the logo, the shoulder buttons, rounding those razor sharp edges that have caused pain and injury, yet my soft sensitive hands somehow continue to be exposed to and come out unscathed. (That last one is what I am most interested in since it didn't bother me on the Pandora, so it should be great now, or at least I hope it is for those that had a problem.)

If you feel he was attacking GPD, continue to assume he was.

Another thing that would speed things along, even if working from scratch on the Win case was mentioned earlier; they are in close proximity to the case designer/manufacturer. It might be that they have the designer in-house, and that they make regular visits to the people that will actually make the cases and understand the limitations. That close relationship between those making the guts and the case, plus dedicated personnel working solely on their case design (if they have this) should cut the time it takes. Actually, people working 100% of the time on the Win means the whole thing should happen a lot quicker.
 
ok, maybe I am looking at it weirdly then. Me and you read the same posts. I pulled from them the impression that ED is suggesting that they took the XD design, point and clicked a couple times, and an afternoon later you got a GPD Win, few days later you can have a printed case to demo.

If ED is saying it is quite the complex thing to make those changes, then I apologize.
While I wouldn't have said it that hard, it mostly is, once you decided what design changes you want to have.

As mentioned: Both the GPD XD and the GPD Win (and all tablets out there) have incredibly simple designs. Which is not meant to discredit anyone. It's great, it makes things cheaper to produce, easier to assemble, easier to design.
Why do you think chinese companies can spit out thousands of different tablets for 60 EUR end customer price? Because there's not much in there.

It's also the reason most phones and devices don't have replaceable batteries anymore: to save costs.

If you ever did CAD, you know how quickly something like that can be done:
The base layer is a rectangle. It gets some walls around it. The edges can be rounded (that's a normal tool in CAD), and you can cut out holes within seconds as well, or add pegs on any surface.

In CAD, everything you do is based on something. You can define the height or thickness of the pegs to all be the same. Once you change the parameters, it automatically happens to all of them. And you can move them around as you please.
You can even define locations, i.e. that one peg is always 1cm away from the edge, or even 1/3 of the total length of one edge (so if you change the length of an edge, the peg moves accordingly)

It's all non-destructive. If properly done, sizes, etc. can all be changed at ease.
So once designed, changing the backside from being a 10x15cm brick with hard edges to a 12x16cm form with rounded edges really is just changing three parameters: the length of the edges and the diameter of the roundness tool.
And those two look totally different, but the base is the same.
If you don't believe me, I can make a video for you showing that.

So, try to think in CAD way and then compare the XD with the Win again:
There's not that much that has changed, even if it looks totally different.
Which is great, and what I'm aiming for with the Pyra case as well: To be able to use it for any future enhancements without recreating everything (same with the PCB and the CPU upgradibility).

Doing it like that is not discrediting, it's how things should work for a company: Use your assets to make future projects faster, cheaper and better.

What takes longer than the design itself is the decision and thinking about what you want to change and how to solve possible issues. The design itself doesn't take long.

Now, as the question arose why we reused Pandoras design when we were recreating it instead of giving it a more modern look and what's so complex about it:

What's pretty simple for us is the LCD part. That's simple and it also was done very quickly. The battery lid and stylus are two parts that could also be created within a day.

The complex part is the battery compartment and the part with the keyboard in.
Those two are a lot more complex than any modern devices cases out there.

Why do they need to be so complex?
Well the battery compartment is needed to have a replaceable battery.
Instead of just having the battery glued into the case (like the GPD, the GCW, the Caanoo, most smartphones, etc.), there needs to be a compartment with aligned contacts.
So the backside can't be a simple rectangle with walls anymore, you need to add the compartment, the battery lid, etc.

So a replaceable battery makes things more complex (and adds thickness) which is why it's rare these days.

Then the keyboard: On the GPD, it's a simple cutout where the keyboard will be put in.
I assume it's one of those keyboards like you had in older smartphones (N900, e.g.), where the keyboard is one block you put in.

We could've done that as well, but didn't for two reasons:
1. I like Pandoras keyboard (with a keymat) better than those mobile keyboards (and according to others I asked, it's the same for them as well)
2. I have zero experience with those keyboards. Our money is very limited, and you've all seen with the Pandora what unexpected issues can occur when you do something new. So I needed to work with something we already had experience with, to avoid possible pitholes.

Another thing that added a lot of complexity actually is the size:
We have a lot of features on the PCB (4G, two SD card slots, replaceable CPU board, multiple SD Card slots, etc.). The easy easy way would have been to increase the width of the device, as that adds more space for all these parts.
But I didn't want to do that, as it would've made it hard pressing the middle keys of the keyboard (that's still my main concern with the GPD Win).
Thickness had to be added due to the battery size, but I set a limit for the width that had to be kept.

That meant: Nikolaus and the case designers had to work together, move the parts around so that they all fit in the narrow space and do not cause any issues (we don't want breaking cases).
That was pretty complex, but it's what I wanted to achieve.

GPD probably took the more simple approach:
The put everything they need on the PCB and then defined the case size.
Which is okay as well, but the middle keys of the keyboard could cause a problem here. We'll see, but that's my concern.

So, the Pyra is thick because of the replaceable battery, the replaceable CPU Board and the type of keyboard we use.
And it's complex, because I didn't want to increase the width so that all controls and keys are still easily accessible.

And it's similar to the Pandora because I can't afford to do any experiments right now, so better keep things as similar as possible to avoid issues and mistakes but improve it as good as possible.

And finally, I actually like Pandoras design very much.
I don't know why everything these days has to look so rounded, etc.
I still think for something like the Pyra or Pandora, a case like we have it looks and suits best :)
 
I agree about the Pyra's case design - I like it. I also liked the Pandora's design too. I never had a problem with apparent sharp corners on the pandora- playing for a long time though led to the hands getting sore due to the placement of the shoulder buttons and how one held the device.
 
So, try to think in CAD way and then compare the XD with the Win again:
There's not that much that has changed, even if it looks totally different.
So it's basicly quite the opposite of the Pandora<->Pyra design, they look similar from the outside but are completely different at the inside. :D

Then the keyboard: On the GPD, it's a simple cutout where the keyboard will be put in.
I assume it's one of those keyboards like you had in older smartphones (N900, e.g.), where the keyboard is one block you put in.

We could've done that as well, but didn't for two reasons:
1. I like Pandoras keyboard (with a keymat) better than those mobile keyboards (and according to others I asked, it's the same for them as well)
2. I have zero experience with those keyboards. Our money is very limited, and you've all seen with the Pandora what unexpected issues can occur when you do something new. So I needed to work with something we already had experience with, to avoid possible pitholes.
What could have gone wrong with that? ;) (*joke* - I know EVERYTHING can go wrong.) Well, besides the missing keyboard light, I see some advantages of such an dense keyboard - you save alot of costs for mold making because it's basicly only one hole for the entire Keyboard. maybe an option for future devices, even if the experience is missing, I see some potential there.
 
Well, besides the missing keyboard light, I see some advantages of such an dense keyboard - you save alot of costs for mold making because it's basicly only one hole for the entire Keyboard. maybe an option for future devices, even if the experience is missing, I see some potential there.

Keyboard light would not be a problem, that's not any different from Pyras keyboard here (my Droid 4 also has backlight). You create a transparent mat and print with black and transparent white onto it.
(However, there's no space to put some plastic on top, so it rubs off after a few years - which actually happens on my Droid 4).

But where do you get such a keyboard from? How do you create it?
The keys need to be pushable but the plastic needs to be strong enough so that they don't break and fall out.

I've got no experience with that, and there are no finished keyboards we can simply buy, so this IS dangerous.
 
Not to be misunderstood, I have nothing against the Pyra/Pandora keyboard, it's simple and it works which is important. AND you can replace it without any problems ( separated ABXY buttons would have been an additional plus). This GPD keyboard seems to be somehow mounted/soldered as a single piece onto the PCB because it must be held in place there without a keymatt and without separate holes for each key. So while the mold of the case may be simpler and cheaper there, the actual keyboard is not. Potential is there but it would need some custom solution, I also doubt you can just buy existing keyboards of that kind and fit it onto your handheld. Would be interesting if possible, especialy with an modular (removable) "non-rubberdome-keyboard".
 
Back
Top