It used to be that you needed a well-equipped expensive new beige-box PC if you wanted to play Doom at all. Now, you can do so in a form factor with a footprint smaller than a credit card, as demonstrated by this nifty little build from Adafruit.
The build relies on the Retro-Go firmware for ESP32 devices, which can emulate a range of machines, from the Nintendo NES and Game Boy to the NEC PC Engine, Atari Lynx, and, yes, Doom itself. It can even run Doom mods, via the WAD architecture used by the game.
It was a simple matter of porting Retro-Go to run on the tiny QT Py ESP32 Pico board, and everything fell into place. With six tactile buttons, it’s capable of not just running Doom, but running it at full playable speeds including that classic soundtrack. The 1.3″ 240×240 screen looks surprisingly crisp and does a great job of displaying the game while keeping everything readable.
It’s one of the smaller Doom-capable portables we’ve seen; we reckon you could stuff this in the change pocket in your jeans if you tried hard enough. We’ll never quite get over seeing the world’s most loved FPS running on commercial kitchen hardware, though. Video after the break.
19 thoughts on “Pocket-Sized Doom Is Actually Playable”
I looked on Adafruit and I see the QT Py, but not the interface board. While I would enjoy to toy with one for the sheer geekyness of it, I would actually live to try to hack this to build interfaces for small pieces of equipment. Does anyone know where the interface board (with the screen and buttons) is available, or was it custom?
As of 3 days ago (1/23) they are still testing the prototype design.
It looked like the interface board (with the TFT) has been used for several projects. If it is not ready, OK. I had not gotten that from the presentations and online info. In the meantime I will think about possible fun uses ;-)
“Beige box PC”?? You’ve not built a PC for the past 20 years…..have you.
That’s the type of PC that Doom ran on when it came out, about 25 years ago …actually.
They’re talking about back when doom came out. You haven’t done much thinking in the last 20 years….have you?
She mentions in the video that it’s a prototype. It’s custom.
sorry. I thought she was talking about the ESP32-pico board, not the interface. Anyway, I’ll look into it again in a couple of weeks. Thank you.
If you can’t wait for adafruit to release the board, ODROID-GO also runs the firmware.
Are these even available anymore?
Don’t quite understand why everyone is always using 4 separate tact switches instead of a 4-way tact joystick.
4x SMD buttons is a far cheaper than a joystick. You can get buttons for around 1 cent each in a qty100 order, and the cheapest joystick I’m seeing comes out to more than 20 cents. Button footprints are also fairly universal, whereas most joysticks I’m seeing will require a specific footprint, so you could easily be forced into design revision between runs.
I do agree with the costs that you are mentioning. But these kinds of devices are not high-volume devices. Nobody is going to sell tens of thousands of these. Maybe a thousand? Probably less. So saving 19 cents on the BOM can only be seen as a theoretical excercise.
The law of the greater numbers can also be applied in reverse. If they sell 1000 units, the extra cost for manufacturing is only 190 dollars. It’s not like it’s going to save or cost them hundreds of thousands of dollars on a manufacturing run.
But a joystick would make the thing *actually playable*. Which is a huge added value for a buyer… And I’m sure that if a buyer would have to choose between a device with 4 tact switches and one with a joystick, they would be prepared to pay 50 cents more for the one with the joystick.
The seller would still sell as much as he would have before, but maybe even more because of the increased playability and therefore increased usefulness of the device. And so the seller would actually make 30 cents per unit more profit, or 300 dollars, if we talk about 1000 devices.
If I were in the market for something like this, I would choose the device with joystick hands-down, and would gladly pay 50 cents more. I probably would even buy it if it were a dollar more expensive.
And those 300 dollars more profit would easily pay for the at most one hour of work to change the footprint and the PCB layout.
You haven’t convinced me yet. ;)
and if I were a designer, at this point I would look at the board and see if there was some way I could tweak the board design and allow you to choose either 4 tact switches OR a joy stick, but I’m just that way ;-)
What’s the reliability of a 4-way tact switch? The 4 separate tact switches closely emulates what the big boys (Nintendo, Sony, Microsoft) do with their consoles.
If I were going to put a d-pad on top of either, I’d prefer 4 switches because that joystick is going to feel like hot garbage.
“The 4 separate tact switches closely emulates what the big boys (Nintendo, Sony, Microsoft) do with their consoles.”
No they don’t. Because with the big boys the button is part of a whole. Beneath, on the PCB, there might be tact-switches. But on top of the tact-switches is a +. And around the + is a case.
With the big boys, the joystick is a composite of PCB, tact-switches, D-pad cross, and front- and backcase.
Ever noticed that a D-pad generally either has a frame around the buttons with a small knob in the middle, or a hole in the PCB where a post on the backcase will protrude. And that the D-pad cross is actually resting and pivoting on the small knob or post? It’s crucial to the proper working of the D-pad.
The big boys have a whole system, not just 4 tact-switches in a + format. And the whole system is actually more expensive than a tact-joystick.
I’m still not convinced. ;)
There’s another question that formed in my head. Is such a small display really so much cheaper than a display with same resolution, but four times the size? I’m quite sure that bigger displays are manufactured in much larger quantities, e.g. for the automotive industry, and might therefore even be cheaper than those small displays.
Isn’t the display size really chosen because the designers went for a cute-factor, and not for cost-reduction? But does ‘cute-factor’ really sell so much better than ‘usefulness’ (e.g. being able to read the game’s menu text without having to search for a magnifying glass)?
In this case, I’m not sure of the answer. I’m sure Adafruit knows much better what makes their customers buy than I do.
But I can’t read sh*t on that display. :D
I do not know what motivated their choice, but I have seen these small screens in places where they wanted some feedback but in a small footprint (the new Unifi security and networking stuff is that way) — they basically have a 1U footprint, and need to stuff everything onto the edge.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)