What Flaws Do You See In The Pandora?


My biggest complaint isn't with the unit itself, but with the fact that there still isn't an easy-to-install devkit that includes everything a newbie needs to make a game on a desktop yet*.



*I have seen some install scripts passed about that are targeted at specific distros, and of course, we can compile on the unit itself, but I think we'd see an uptick in development if there were something like Dev-C++ or a preconfigured version of Eclipse that had arm-linux-gcc and friends, along with a visual .pxml creator/editor/validator.
 
el_pango said:
My biggest complaint isn't with the unit itself, but with the fact that there still isn't an easy-to-install devkit that includes everything a newbie needs to make a game on a desktop yet*.
* Because you didn't look hard enough ;)
Use the script linked in the Pandora's chapter of this.
And thanks PokeParadox for that ;)
 
Last edited by a moderator:
There are a few flaws in the Pandora that keep it from being precisely the last handheld electronic device a person would ever need for the foreseeable future. These flaws are entirely forgivable considering how few devices are providing an open handheld gaming/generic computing experience with a physical interface comparable to the Pandora, but when it is considered how likely another device from another company will probably come along in the near future to fill the Pandora's small but not adequately served niche and how mobile technology has advanced in the past few years to smaller, cheaper, and more efficient hardware, it becomes increasingly important that the few flaws in the Pandora become addressed as soon as possible (preferably by the first revision of the Pandora, AKA Pandora 2) so that OpenPandora Ltd. won't be pushed aside by another, more aggressive company with more funds and that people will still have plenty of reasons to buy a future Pandora revision.​
I will first address the flaws in order of my preferences. Then I will summarize the flaws in order of a proposed logical business priority.​
The first flaw I would like to address is the button layout. Currently, as it is, the button layout of the Pandora is simple and adequate for most games in general, but that does not mean it couldn't or shouldn't be improved. I have used and seen quite a few different controllers over the years, and I will show precisely which traits are best from a large pool of gamepads, which traits are to be avoided, and hopefully go into enough detail for you to imagine what a revised button layout may look like. The first aspect of the button layout that needs to be addressed is the layout of the face buttons. Yes, few retro gaming enthusiasts (or even gamers in general) are willing to criticize the classic diamond layout, but I believe this is simply because of its widespread use on so many controllers and the feelings of nostalgia retro gamers have for the Super Nintendo. I'm not putting down the Super Nintendo. I love its controller too:
800px-SNES-Controller.jpg
After I used a gamepad with six face buttons, I found that the diamond layout is actually lacking. Sega was the first that I recall using a six button layout:
800px-Sega_Saturn_Controller_-_Type_2.png
I recently experienced a modern gamepad using this layout that I particularly favor due to the top row being the same size as the bottom:
8828-Ryu.gif
This layout gives the most extra advantage to fighting games specifically (Street Fighter needs this bad.), but it's not unusual for games to need a couple more buttons especially if you only have two shoulder buttons. Your objection to this layout may be either that you don't play many fighting games or that most of your games require few buttons. I assure you that even if you play good old NES games, the other four buttons don't get in the way. The next aspect of the button layout I'd also like to address is the lack of analog nub buttons. This gamepad is the first one I remember with buttons in the analog sticks:
800px-Playstation_image3.jpg
I feel this is much less important than having six face buttons, but it's a very nice feature to have for games that can make use of buttons in analog sticks. FPS games tend to use them a lot for sniping, and some games use them for changing the camera and/or crouching a character. It's simply a feature that I've come to expect from modern gamepads. I'll understand if designing a new analog nub capable of this would be a pain in the ass for OpenPandora Ltd. to actually implement, but as long as I'm making a post about the few flaws in the Pandora, I'll talk about everything I can think of. The last aspect of the button layout I'd like to talk about (and probably the least important) is the number of shoulder buttons. It's not just modern gamepads that tend to make use of four shoulder buttons but a surprising amount of modern games have as well, so it's nice to have them available. Obviously we can't have this:
618px-PS1All3ControllersTopButtons.jpg
Or this:
xbox-360-wireless-controller-20050513052443718_640w.jpg
Both of those layouts with four buttons would ruin the portability of the Pandora. I want my Pandora to fit in my pocket, so here's what I'm suggesting:
ccfeature_002.jpg
(I don't know what that thing in the middle is by the way.) I'm not sure how it would feel to reach and press those two circular buttons on the shoulders in an actual game, because I don't have a Wii game that really plays with them yet. However, this allows the Pandora to have two extra shoulder buttons without needing more space. It's worth mentioning, but the importance of implementing the two extra shoulder buttons if the Pandora already had six face buttons is questionable. (I'd still prefer to have them.) I know what some of you may be thinking. All three of these suggestions in all amount to six more buttons, but I assure you all that I'm not advocating the same mistake as this gamepad:
top-10-tuesday-worst-game-controllers-20060221054050260.jpg
It is not unusual for someone to quickly judge that gamepad's problem as only having too many buttons. It does have too many buttons (fifteen in all minus start/select), but most of the problem is that twelve of them are located in an inconvenient and indistinct/illogical positions. Pressing the three at the top is simple, but trying to find the right button for a particular action on the bottom twelve is hard. The bottom twelve are easily confused with each other due to the lack of distinct and convenient positioning. When we see the Xbox 360 controller,
800px-360_controller.svg.png
it is obvious that at least ten buttons (I'm not including start/select/home.) are easily manageable and not causing the gamepad to be overly complex. Together, all three of my suggestions only result in two more buttons than the Xbox 360 controller, and I hope to have explained in enough detail before now to show why 12 buttons is not too much. Before I finish talking about the button layout, I must caution against one single thing:
800px-Nintendo-DS-Lite-w-stylus.png
Don't see what I'm pointing out? It's the four face buttons. I couldn't find a picture with an adult hand on it, so you'll have to trust me if you've never used a Nintendo DS Lite. Those four buttons are tiny and too close together. This is an example of a safe and conservative button layout done terribly wrong except for children, and even children like big buttons sometimes. Please, OpenPandora Ltd., don't copy the DS Lite button size for Pandora 2. That Street Fighter IV gamepad up there has excellent button size. It's my favorite gamepad.​
The second thing I'd like to address wasn't really a flaw not too long ago, but exciting new screen technologies are coming out that will soon make the Pandora's screen obsolete. These new screen technologies are going to allow readability in sunlight and lead to tremendous battery life. I will only show one example of these new screen technologies as I don't know which one is most favorable: http://www.pixelqi.com/ I firmly believe that the next revision of the Pandora needs to take advantage of some new screen technology to be relevant when it is eventually released. It won't be long before nearly every mobile device utilizes something like Pixel Qi. Also, if anyone wants to weigh the pros and cons of various touch screen interfaces that's fine, but the Nintendo DS showed me that resistive touch screen technology is very adequate.​
The third and last flaw that I would like to address is the lack of mobile telephony, mobile internet, and a navigation system, such as (and probably preferably) GPS. This is probably the most important flaw to address for the next Pandora to be relevant at all. I have no idea why Apple is marketing their touchscreen only interface iProducts as gaming devices when they could slap a PSP Go style slider on the back of one to hide/show the gaming controls and compete seriously with Nintendo's ridiculous two screen gaming weirdness (Seriously, Nintendo, stop it. I don't like it.). The point is that at any moment some intelligent smartphone company could come along at any second and release a device with gaming controls. This entire time the Pandora has been in development must have been some fluke because of all the time that has passed without any smartphone company waking from its slumber and making the first 'Gameboy-Phone' for insane profit. Before you raise your hand, the N-Gage was garbage that no serious gamer would want to touch. It doesn't count, because it doesn't have real gaming controls anyways. Look at it:
Nokia_N-Gage.png
Do you want to play it? No, you don't. I assure you that it's some fluke that Sony hasn't given us the PSP phone yet (but I think I heard they're working on it anwyays). How long till Apple steps up the the plate and bats a homerun with a real gaming device? They already attracted all those developers and profit with a touchscreen interface. The last thing that needs to be said is that I don't want to carry both a Pandora and a cellphone. It's only making less and less sense as time marches on to not include all these extra features in a single device due to the miniaturization of technology.​
Now that I've mentioned all the flaws I will summarize them in order of a proposed logical business priority from most important to least important.​
  1. The lack of mobile telephony, mobile internet, and a navigation system (GPS).
  2. The screen.
  3. The button layout.
    • The amount of face buttons.
    • The lack of analog nub buttons.
    • The amount of shoulder buttons.
Before I end this post, I would like to take the time to say that the clock speed of the processor and the amount of RAM in the Pandora are entirely sufficient and not flawed. A Pandora revision would of course have a greater processor due to how cheap and fast mobile technology (especially ARM technology) is getting, so it will be upgraded by default if Pandora 2 comes about and is not worth stressing. Lots of RAM is hardly necessary for gaming purposes, but we will most likely get it in a Pandora 2 anyways. The Xbox 360 only has 512 MB. The problem is that we don't have enough people developing for the current Pandora to make the most efficient use of the hardware available. We could have more battery life and faster emulators with the right code.​
 
Brutus said:
The lack of mobile telephony, mobile internet, and a navigation system (GPS).
I would not have bought it if it were a phone. It's a major additional cost for something I wouldn't use. We're not talking a couple of extra bucks for an unused feature, it's an expensive undertaking to make a phone which complies with all the requirements, and the parts aren't cheap either. $50 for a feature I don't want is a lot of money, especially when it comes with additional draw backs. I'm not alone in this. This is the only thing I will adamantly argue against because this has come up many many times before with the argument "it would increase the number of users" but the reality is that, while it would increase the number from one niche, it would decrease the existing niche. Whether the increase balances out the decrease enough to make the extra difficulty in building a phone worth it is not something I think OPT was (with good reason) willing to accept.

Brutus said:
The amount of face buttons.
4 is plenty enough for me. If I need more, the keyboard is right below it, or the EXT port has a bunch of GPIO lines you can connect additional buttons to (if Craig or ED ever offer to sell us some of those EXT connectors. C'mon guys!) but I can see why someone might want more. I disagree, and if it had had six I would have said the flaw was that it had too many buttons, but you can't please everybody :p

Brutus said:
The lack of analog nub buttons.
It theoretically has analog nub buttons. At least, the hardware should be capable of it. The software which controls the nubs (which is just a binary blob, so we cannot modify it) is the limiting factor here. So the flaw should be that they went to a company that didn't agree to give up the software source for their nubs, not the nubs themselves.

Brutus said:
The amount of shoulder buttons.
There are pads on the board for a second set of shoulder buttons. The reason they aren't connected is because no one could agree on exactly how they should be placed. I think they made the correct decision here: they couldn't please everybody so they gave them the tools they needed to please themselves however they wanted.
If you want to add a second set of shoulder buttons, it isn't extremely hard: a button, a soldering iron, and a drill, and you can place it wherever there's room. Not to be done by an amateur, but it doesn't require a master soldering expert either.
 
Last edited by a moderator:
Brutus said:
(I don't know what that thing in the middle is by the way.)
It moves two clip-type-things that are set into the back. I can only assume that it was meant to hold on to possible expansions, if anyone wanted to produce them. It serves no actual purpose in practice.

and compete seriously with Nintendo's ridiculous two screen gaming weirdness (Seriously, Nintendo, stop it. I don't like it.).
To my knowledge, it's only the most successful console of all time. :p Here's to much more.

EDIT;

WizardStan said:
I would not have bought it if it were a phone. It's a major additional cost for something I wouldn't use. We're not talking a couple of extra bucks for an unused feature, it's an expensive undertaking to make a phone which complies with all the requirements, and the parts aren't cheap either. $50 for a feature I don't want is a lot of money, especially when it comes with additional draw backs. I'm not alone in this. This is the only thing I will adamantly argue against because this has come up many many times before with the argument "it would increase the number of users" but the reality is that, while it would increase the number from one niche, it would decrease the existing niche. Whether the increase balances out the decrease enough to make the extra difficulty in building a phone worth it is not something I think OPT was (with good reason) willing to accept.
Just wanted to add that Stan definitely isn't alone in this. In fact, if my memory serves me correctly, I gather that telephony features were actually voted against by most people when the machine was being planned.
 
Last edited by a moderator:
WizardStan said:
Brutus said:
The lack of analog nub buttons.
It theoretically has analog nub buttons. At least, the hardware should be capable of it. The software which controls the nubs (which is just a binary blob, so we cannot modify it) is the limiting factor here. So the flaw should be that they went to a company that didn't agree to give up the software source for their nubs, not the nubs themselves.
?!

What I recall was that MWeston noticed that the nub behaved differently depending on whether it was being touched or not. If not being touched by a thumb, it would give no output, but if being touched without being moved, it would output a stream of zeros. His idea was that you could detect that stream of zeros and perhaps use it to emulate a nub button, but it's certainly not an actual nub button.

Or maybe you're talking about something else I haven't heard of.

PS: I agree, phone would be trouble, but 3G modem would be wonderful, especially since that could be used for VoIP by people who really want to converge their devices.
 
Last edited by a moderator:
Tempel said:
Or maybe you're talking about something else I haven't heard of.
Most joysticks have two variable resistors: an up/down, and a left/right. If the resistance is at 0, it is all the way up or left; if it's at 255, it is all the way right or down; if it is at 128, it is in the centre.
These nubs have 4 resistors: up, down, left, and right can all be increased or decreased independently on the hardware. The software in the chip that comes with it takes the value from the up, and the value from the down, does some math, and comes up with a Y value. Same with the X value. It stands to reason that you could push down on the nub (since it's rubber) increasing all four resistances simultaneously. The software could, in theory, be designed to recognize when all four resistances are increasing and register that as a "button". Of course, it can't, because we don't have any way of reprogramming the nubs, but it could have.
 
Last edited by a moderator:
I'm in the same mindset as the others regarding the SIM/Phone inclusion. While a 3G/4G modem would indeed be nice, I wouldn't want to deal with the extra cost or whatever bulk it might come with, and I certainly don't want it to function as a full-fledged phone. An option would be nice, like a USB dongle or something, but I wouldn't want it built in. For one, as much as possible I like to keep my phone free of most duties other than phone and internet, because I never know when I'll need it to have a decent charge. Stuff like web browsing on 3G, playing music and games, and chat all take a lot more power than just phone or checking email, and I'll be offloading music, games, chat, and other duties to my Pandora. I can live with a Pandora with a dead battery. It's difficult to go without a phone.
 
Back
Top