Greetings,
TL;DR: I will be representing TASVideos at the
Awesome Games Done Quick 2014 charity marathon for the
Prevent Cancer Foundation on January 5th starting at 8:25 PM EST and I am requesting donations from you (yes, you, reader :) toward a $10,000 USD goal during the TAS block where I will be showing TAS's on consoles. Donating during the block will enter you in a raffle to win various prizes, including the first production run bot from true in the form of a ROBBerry Pi!
In August I started discussions in IRC about the idea of TASVideos supporting the AGDQ 2014 charity marathon and I'm very happy to see the level of support from across the TASVideos and SDA communities. Before I go into details, I should explain what AGDQ is and my motivations for doing all of this work. The website
SpeedDemosArchive.com is the largest community of players completing games as fast as possible using human skill alone. Several years ago they started a yearly charity marathon now known as
Awesome Games Done Quick and the marathons have driven more than $1 million in donations to various charities, primarily to the
Prevent Cancer Foundation. As cancer claimed the lives of three of my grandparents when I was young I am motivated to support SDA's charity efforts any way I can. As part of this, I've had good discussions with adelikat about having TASVideos actively support the AGDQ 2014 marathon through things like a website banner on the front page and representing the TASVideos community during the marathon itself.
I
pitched the idea of a TAS block in the AGDQ 2014 Games List thread and mikwuyma (who is generally in charge of game selection and the marathon itself) responded on the
next page by initially allocating a 30 minute block (which he
later bumped to a full 45 minutes) during primetime on the opening day of the marathon on 2013-01-05 at 8:25 PM EST.
Update: Many people contributed to this project over the course of the last several months, including Ilari, zid', micro500, DarkKobold, adelikat, and true. Here is what we've accomplished, with the original post at the bottom to provide historical context for the discussions in the rest of this thread.
I attached a
Raspberry Pi computer and the
board that true built to a
ROB and
dubbed it ROBBerry Pi!
ROBBerry Pi can play back NES and SNES games and will be used to play back
Gradius with adelikat commentating,
SMW total control thanks to Masterjun, and a bid war between
SMB3 and something as-of-yet undecided.
On the N64 front, micro500 has his
existing bot and has also started work on a new revision. We'll be playing back
SM64 0-star.
My wife and I will be at AGDQ 2014 for the entire marathon. She's made a few raffle prizes and I will be
volunteering at various stations.
What I need now is for people to spread the word. The TAS Block will run on 2014-01-05 from 20:25 - 21:10 EST (8:25 PM - 9:10 PM EST). My personal goal is to raise $10,000 USD in those 45 minutes. It's an optimistic goal, and it can't happen without the help of the TASVideos community. Donating will enter you in a raffle to win various prizes, including the first production run bot from true.
The second half of the original post is below for historical context
We even have some
good ideas for
donation incentives!
So far, we have the following suggestions on things we can show in a TAS oriented block:
- NES bot running Gradius, potentially with commentary from adelikat if he can attend Update: He'll be there!
- N64 bot running Super Mario 64 0-star
- SNES bot showing Super Mario World, possibly with total control based on work Masterjun is doing Update: There are exciting developments on this front!
For my part, I have experience and access to a hardware lab and am committed to doing the following:
- Post regular updates of my progress (in an entertaining and educational manner)
- Build an NES bot (using either the Arduino hardware or my own attempts using the ENIO board)
- Build an N64 bot (either the existing design by micro500 or my own design)
- Build an SNES bot (based on the design feos came up with)
- Attempt to build all three on common hardware (as a TASBot) using a Raspberry Pi and engage the Pi community in the project
- Attend AGDQ 2014 early around January 4th, 2014 to help them with setup
- Read donations during the marathon
- Set up hardware and give a demonstration of all of the bots that work during a TAS block
Here are the parts I still need to pull this off:
A copy of Gradius that will sync with adelikat's run Update: micro500 lent me his cartridge - Thanks!
A copy of Super Mario 64 (an SDA member named Corrodias on #SDA offered to donate an N64 console for this project)
A copy of Super Mario World (I own one, but it's not with the rest of my SNES games and I have been unable to find it)
Four sacrificial SNES controllers (I only need the cords so broken controllers are fine) Update: I bought SMW, SM64, and two controllers from a local enthusiast and I bought a third controller using Bitcoins from SDA member AniMeowzerz
- A Raspberry Pi to backfill the one I plan on using to create a new TASBot platform (still needed)
Potentially a GERTBoard for the Raspberry pi to protect the Pi's sensitive 3.3v GPIO pins from voltage spikes No longer needed
- Wiring harnesses for each console, possibly with console-specific circuitry as needed
An Arduino (as a "backup strat" :) Borrowed from a co-worker with much thanks
I'm seeking donations from the community to help obtain these items as some of them might be difficult to find.
In particular, getting the correct copy of Gradius could be a challenge, and not many people actively sell broken SNES controllers (I'm good here now). I'm planning on financing the plane trip itself from California to
Washington, DC and covering room costs for a week at the Crowne Plaza as well as covering the costs of all of the materials needed but at the same time I would not turn down donations.
Update: A
n anonymous Bitcoin donor (namely,
Sonny from Butterfly Labs) covered all of the existing travel and materials costs! The only remaining costs I have are food and transportation while we're there and a few smaller materials (lots of yarn for raffle prizes, primarily :). If you're inclined to donate but do not have any of the needed materials I can receive Bitcoin donations at the address 1CknXDyb5jTG1Nw1hs437r6fnz7npBSEfA and can publicly thank donors during the stream if you make your identity known.
Thank you in advance for the support,
A.C.
******
The breadboard came up fine, even with lots of probes attached.
We started with just the clock connected.
A few things surprised me - for one thing, the scope showed that the Game Genie ran the CPU at around 112 kHz, but Dr. Mario ran at 43 (I think) and Mario ran at 78 as tested with a single controller connected. The scope had a USB port and a print ability, so I saved off a few screenshots. Here's the state while running the Game Genie.
This shows the clock in yellow, the data in purple or blue, and the latch in green (although apparently we'd bumped that when we saved the images from the unit itself). In the above image, the latch only hit once, but when I plugged in a second controller while playing SMB3, the clock went to 16.1 kHz and the latch triggered twice in a row.
While the scope was attached I was able to get everything working with the dip switches so flipping an appropriate switch caused the correct button press to be sent to the NES. I even managed to stomp the first Goomba in SMB3 using just the dip switches.
Now that I'm completely confident that I have reimplemented an NES controller on a breadboard, my next step is to tear this down and further investigate WiringPi and get cracking on getting level shifting going for the CPU and latch lines into the Raspberry Pi. I need to get the +5 volts headed from the NES to the Pi down to a safe level of between 2.5 and 3.3 volts as well as boost the 3.3 volts the Pi will send down the serial data out line up to 5 volts. I have limited parts on hand but I'm going to try to get something working with what I already have and then plot my next parts order.
I am still actively looking at the option of making a common connector from the Pi to a common plug type and converting each console's controller type to act as the wiring harness and console-specific circuitry holder, but the one problem with this plan is this may complicate working with multiple controllers. The other option would be to create a single PCB that could support NES, SNES, N64, and perhaps Sega, but that's a bit ambitious so I'll worry about those decisions later.
Thanks for the feedback and keep it coming!
A.C.
******
That's a solid-core door sitting on two filing cabinets for a desk with the Twilight Zone pinball machine to the right. Note the breadboard with a previous project's wiring still in place. It took me a while to clear off all of the jumper wires but I was able to get them sorted by length.
I decided to sacrifice an NES MAX controller for this project (I have two normal rectangular NES controllers and that particular MAX controller always seemed difficult to use for me). I desoldered the cord and removed the existing 4021 CMOS shift register IC from the board with the assistance of my helping hands gripper.
An NES controller uses the shift register to collect the 8 buttons that can be pressed (U, D, L, R, Select, Start, B, and A) and feed them to the NES one bit at a time through a single serial data out pin. Since my goal is to emulate what an NES controller can do I decided to start with that chip on the breadboard. I initially relied on the wiring diagram from micro500's original instructables post. I started by getting the wiring harness pinned down and the 4021 placed with a few wires run.
In this state I wasn't supplying the column marked A with any voltage so the dangerous-looking yellow jumper on the right was innocuous. It should also be noted that in this state it didn't do anything useful and I wanted to fix that so I took a couple of days to do some research. I was attempting to short one of the 8 parallel in pins of the 4021 to ground in the hopes that it would trigger a button press but nothing was happening and I wanted to fully understand why and what I should be doing instead. With the aid of an army of people in #tasvideos giving me guidance (including micro500 himself, DarkKobold, true, zid', Ilari, and several others I'm probably forgetting) I got a better grip on how to work with CMOS integrated chips. The consensus at the time was that the voltage needed to be pulled in the correct direction rather than just relying on shorting to ground. During this string of educational conversations I came across this helpful post about connecting an NES to a Raspberry Pi with links to a project from Seb who put together a very helpful wiring harness design (albeit with the wire colors in German). After many hours of trial and error testing over the course of a couple of days the breadboard looked significantly more populated.
I happened to have an 8 position dip switch and placed it next to the 4021. I also joined A and B on the left with A and B on the right, respectively, which can be seen in this image of the full breadboard. By connecting all 8 parallel pins to 5 volts using 10k resistors I was able to pull those pins high unless the switch was flipped to allow current to flow to ground, thus pulling the respective pin low. I did a bit of testing and this somewhat works, but at this point there is no timing, so what tends to happen is all 8 buttons are held down when the console is turned on or no buttons are held down until the 8th switch (4021 pin 1, parallel in 8) is flipped. This tells me that I now have at least a somewhat working circuit (the behavior of the 8th parallel in pin puzzles me, but I suspect it is due to the lack of timing synchronization).
Now that I have the basics down my next step involves a whole lot of research to determine the best way to connect a Raspberry Pi to this interface, keeping in mind that the NES is the most simple device of the lot and the design must eventually support 4 SNES controllers at maximum data rate. I'll post as I make progress, although it could be a while before I have anything else fancy to show. Thanks again for everyone who assisted me and here's to further progress over the coming days and weeks,
A.C.
******
ROBBerry Pi can play back NES and SNES games and will be used to play back Gradius with adelikat commentating, SMW total control thanks to Masterjun, and a bid war between SMB3 and something as-of-yet undecided.
On the N64 front, micro500 has his existing bot and has also started work on a new revision. We'll be playing back SM64 0-star.
My wife and I will be at AGDQ 2014 for the entire marathon. She's made a few raffle prizes and I will be volunteering at various stations.
What I need now is for people to spread the word. The TAS Block will run on 2014-01-05 from 20:25 - 21:10 EST (8:25 PM - 9:10 PM EST). My personal goal is to raise $10,000 USD in those 45 minutes. It's an optimistic goal, and it can't happen without the help of the TASVideos community. Donating will enter you in a raffle to win various prizes, including the first production run bot from true.
The second half of the original post is below for historical context
