Hardware anomalies with a Controller switch

Discuss technical or other issues relating to programming the Nintendo Entertainment System, Famicom, or compatible systems.

Moderator: Moderators

Post Reply
User avatar
mitch3a
Posts: 50
Joined: Mon Jan 23, 2017 8:08 pm
Location: Boston, MA
Contact:

Hardware anomalies with a Controller switch

Post by mitch3a » Wed Feb 20, 2019 7:28 am

We noticed something REALLY weird in a game yesterday and I think it's related to hardware issue. If you're not familiar with Metroid, it's an open world platformer where you collect different powerups as you go along. There are some well established screen scrolling glitches (called door wraps) that are used to skip certain areas (I don't think these have anything to do with what happens here bc we've collectively done them probably millions of times at this point).

In this clip, the player starts from the beginning and has no powerups. In fact, all the way til 5:49 he still has no powerups... but then he comes out of the wall and all of a sudden does have them. He has basically all the power ups and no missiles according to the password that pops up after he dies: https://www.twitch.tv/videos/383635109

His full description of what happened is:
I started with a SNES controller plugged into SNES/NES adapter by http://www.raphnet-tech.com into player 1 and a normal NES into player 2
then after dying to kraid, I switched the SNES to P2 and the NES to P1
and that was it
no cheats, passwords, nothing
just my Metroid cart in my frontloader NES
power off, power on, go
I've also attached the password that popped up after he died. While the video clearly shows he has at least 3 missiles (didn't slow it down yet to verify the actual number), based on the password generator (and some light testing for missile drops), the actual missile capacity appears to be 0. Password generator: https://www.truepeacein.space/

I ASSUME this is a hardware issue. I'd also guess the controller switch has something to do with it (cos this is the only time we've ever seen anything like this), but I can't even theorize how. Anyone have thoughts about what might be at play here? If we can't come up with a decent hardware cause, then i'll look really deep at if this is a software issue... but again, I'd be REALLY surprised if this was the case cos this wrap is at the start of the run and done by a decent amount of people (~20) and probably each have thousands of attempts at it... and this was the only time we saw anything like it.
Attachments
pwssword.png

User avatar
dougeff
Posts: 2710
Joined: Fri May 08, 2015 7:17 pm
Location: DIGDUG
Contact:

Re: Hardware anomalies with a Controller switch

Post by dougeff » Wed Feb 20, 2019 8:31 am

I'm going to speculate that changing the controller bumped the NES just enough to misalign the cart and the CPU read from the wrong address and executed false code for a frame, or so, which somehow overwrote the RAM addresses which caused the glitch.
nesdoug.com -- blog/tutorial on programming for the NES

User avatar
darryl.revok
Posts: 520
Joined: Sat Jul 25, 2015 1:22 pm

Re: Hardware anomalies with a Controller switch

Post by darryl.revok » Wed Feb 20, 2019 9:16 am

dougeff wrote:I'm going to speculate that changing the controller bumped the NES just enough to misalign the cart and the CPU read from the wrong address and executed false code for a frame, or so, which somehow overwrote the RAM addresses which caused the glitch.
I would have thought the same, if I hadn't screwed up my Zelda game by plugging in an NES Advantage, and then finding out later that this is a software glitch and it wasn't just a jostle.

I wish I had more information on the topic though. I can't find an explanation of what happens.

If the bug was introduced while porting Zelda from FDS then they'd both have that in common.

User avatar
mitch3a
Posts: 50
Joined: Mon Jan 23, 2017 8:08 pm
Location: Boston, MA
Contact:

Re: Hardware anomalies with a Controller switch

Post by mitch3a » Wed Feb 20, 2019 9:37 am

Do you have more info on this FDS port issue? I'm aligned with doug that it's PROBABLY something like a wrong pin read, but I'm also guessing it's not a realistically recreateable root cause so it'd be tough to prove. So any other leads that I could rule out would be great.

User avatar
darryl.revok
Posts: 520
Joined: Sat Jul 25, 2015 1:22 pm

Re: Hardware anomalies with a Controller switch

Post by darryl.revok » Wed Feb 20, 2019 9:52 am

It's not the FDS port, it's the controller port. I was just noting that both games were ported from FDS, which may or may not be significant.

This page mentions the Zelda glitch, but no information as to why it happens: http://www.glitchgamer.com/the-legend-of-zelda-nes

tepples
Posts: 22017
Joined: Sun Sep 19, 2004 11:12 pm
Location: NE Indiana, USA (NTSC)
Contact:

Re: Hardware anomalies with a Controller switch

Post by tepples » Wed Feb 20, 2019 10:09 am

Many games burp when there's a voltage dip, and a controller insertion often causes a voltage dip. That's why when I've made tests for specialty controllers, I've had to design them so as to plan for a Reset press after hot-swapping from a standard controller (to operate the PowerPak menu) to the specialty controller.

Post Reply