Super Nintendo Hardware Is Running Faster As It Ages (404media.co) 41
An anonymous reader quotes a report from 404 Media: Something very strange is happening inside Super Nintendo (SNES) consoles as they age: a component you've probably never heard of is running ever so slightly faster as we get further and further away from the time the consoles first hit the market in the early '90s. The discovery started a mild panic in the speedrunning community in late February since one theoretical consequence of a faster-running console is that it could impact how fast games are running and therefore how long they take to complete. This could potentially wreak havoc on decades of speedrunning leaderboards and make tracking the fastest times in the speedrunning scene much more difficult, but that outcome now seems very unlikely. However, the obscure discovery does highlight the fact that old consoles' performance is not frozen at the time of their release date, and that they are made of sensitive components that can age and degrade, or even 'upgrade', over time. The idea that SNESs are running faster in a way that could impact speedrunning started with a Bluesky post from Alan Cecil, known online as dwangoAC and the administrator of TASBot (short for tool-assisted speedrun robot), a robot that's programmed to play games faster and better than a human ever could.
[...] So what's going on here? The SNES has an audio processing unit (APU) called the SPC700, a coprocessor made by Sony for Nintendo. Documentation given to game developers at the time the SNES was released says that the SPC700 should have a digital signal processing (DSP) rate of 32,000hz, which is set by a ceramic resonator that runs 24.576Mhz on that coprocessor. We're getting pretty technical here as you can see, but basically the composition of this ceramic component and how it resonates when connected to an electronic circuit generates the frequency for the audio processing unit, or how much data it processes in a second. It's well documented that these types of ceramic resonators are sensitive and can run at higher frequencies when subject to heat and other external conditions. For example, the chart [here], taken from an application manual for Murata ceramic resonators, shows changes in the resonators' oscillation under different physical conditions.
As Cecil told me, as early as 2007 people making SNES emulators noticed that, despite documentation by Nintendo that the SPC700 should run at 32,000Hz, some SNESs ran faster. Emulators generally now emulate at the slightly higher frequency of 32,040Hz in order to emulate games more faithfully. Digging through forum posts in the SNES homebrew and emulation communities, Cecil started to put a pattern together: the SPC700 ran faster whenever it was measured further away from the SNES's release. Data Cecil collected since his Bluesky post, which now includes more than 140 responses, also shows that the SPC700 is running faster. There is still a lot of variation, in theory depending on how much an SNES was used, but overall the trend is clear: SNESs are running faster as they age, and the fastest SPC700 ran at 32,182Hz. More research shared by another user in the TASBot Discord has even more detailed technical analysis which appears to support those findings. "We don't yet know how much of an impact it will have on a long speedrun," Cecil told 404 Media. "We only know it has at least some impact on how quickly data can be transferred between the CPU and the APU."
Cecil said minor differences in SNES hardware may not affect human speedrunners but could impact TASBot's frame-precise runs, where inputs need to be precise down to the frame, or "deterministic."
[...] So what's going on here? The SNES has an audio processing unit (APU) called the SPC700, a coprocessor made by Sony for Nintendo. Documentation given to game developers at the time the SNES was released says that the SPC700 should have a digital signal processing (DSP) rate of 32,000hz, which is set by a ceramic resonator that runs 24.576Mhz on that coprocessor. We're getting pretty technical here as you can see, but basically the composition of this ceramic component and how it resonates when connected to an electronic circuit generates the frequency for the audio processing unit, or how much data it processes in a second. It's well documented that these types of ceramic resonators are sensitive and can run at higher frequencies when subject to heat and other external conditions. For example, the chart [here], taken from an application manual for Murata ceramic resonators, shows changes in the resonators' oscillation under different physical conditions.
As Cecil told me, as early as 2007 people making SNES emulators noticed that, despite documentation by Nintendo that the SPC700 should run at 32,000Hz, some SNESs ran faster. Emulators generally now emulate at the slightly higher frequency of 32,040Hz in order to emulate games more faithfully. Digging through forum posts in the SNES homebrew and emulation communities, Cecil started to put a pattern together: the SPC700 ran faster whenever it was measured further away from the SNES's release. Data Cecil collected since his Bluesky post, which now includes more than 140 responses, also shows that the SPC700 is running faster. There is still a lot of variation, in theory depending on how much an SNES was used, but overall the trend is clear: SNESs are running faster as they age, and the fastest SPC700 ran at 32,182Hz. More research shared by another user in the TASBot Discord has even more detailed technical analysis which appears to support those findings. "We don't yet know how much of an impact it will have on a long speedrun," Cecil told 404 Media. "We only know it has at least some impact on how quickly data can be transferred between the CPU and the APU."
Cecil said minor differences in SNES hardware may not affect human speedrunners but could impact TASBot's frame-precise runs, where inputs need to be precise down to the frame, or "deterministic."
Underlying issue is technically interesting (Score:2)
But seriously, do more than a handful of people care about video game speed running records?
Re: (Score:2)
Right! If new gamers can beat old records because the hardware is faster, good for them!
Re: (Score:1)
Re: (Score:2)
Sorry, I forgot to ask for pronouns.
Re: (Score:1)
Re:Underlying issue is technically interesting (Score:4, Insightful)
Actually, would they beat or lag old records?
If the human takes time to T to complete a task, based on see-respond neurological reflexes, and in that time T, the timing clock counts a few extra beats, then the the system says the player took longer - player lags old records.
On the other hand, if the player keeps pace with data presentation rates and can complete a task in N clock cycles regardless the precise clock rate, and if there is a system clock separate from the audio clock, then system clock sees an N-cycle task completed in slightly less real time than it used to take - players beats old records.
The biggest variance they found was 182 Hz above nominal 32kHz, about 0.5 percent. When you look at how close some speed races are, like Olympic swim and track, or drag racing, measured to 2 or 3 decimal places of a second, 1/2 percent is significant.
I would be interested in knowing how those resonators change that they are speeding up. A "material" Slashdot mystery.
Re: (Score:2)
All excellent points. It's not clear whether the faster clocks would allow great players to beat records, or fail to do as well. My guess is that it depends on the precise nature of the game, and that in some games, players could do better, while in others, they would do worse.
Re: (Score:2)
I don't think it will make any difference to players. The timing difference is too small to affect muscle memory, and for record times they count the number if frames captured these days. Nobody on record pace is pointing a camera at a TV screen anymore.
Re: (Score:3)
Some capacitor in some timing circuit is dying, as likely as anything else.
Or it is the new post-truth world finally settling in. SNES slow? Fake news!
Re: (Score:1)
Re: (Score:2)
SNES was only 16-bit console using 2 clocks (Score:2)
In what Wikipedia calls the fourth generation, Nintendo was indeed the only company using two clock sources for digital logic in one base console.
Both the NES and TurboGrafx-16 have sound on the CPU die, with everything clocked by the same 21.48 MHz crystal. The Genesis derives its sound clocks by dividing the 53.69 MHz crystal by 15 for the Z80 CPU and the VDP's DCSG channels (square waves and noise) and by 7 for the CPU and YM2612 FM chip. The NeoGeo AES divides a 24.17 MHz crystal by 2, 4, and 8 to form
Re: Underlying issue is technically interesting (Score:1)
Awesome games done quick (Score:2)
Re: (Score:2)
mmm... yes!
Right now the GamesDoneQuick twitch channel is casting a "just chatting" rerun and there are 1500 people watching that rerun live.
The VODs for special events can make about a million views.
On the YT side, the WR that is two month old for Super Mario Bros sits at 330K views. The former WR sits at 930K view. One of the influential former WR for Kosmic sits at 6.3M views.
If you look at speedrun history videos, they do well too. For instance, a history of tetris speedrun from this year did 5M views.
A
Maybe it's not your niche (Score:2)
People keep clicking on the YT videos. And people keep making them. So clearly someone follows this stuff, even if it isn't that interesting to a majority of people.
For the record, I think the article is interesting. But I don't really watch speedrun videos much. I prefer the videos that deep dive on how code of these old games works that enables some of the often used tricks in a speed run or in the frame timing itself. But the audience that finds 6502 machine code interesting is far more niche than the on
Re: (Score:2)
Speedrunning used to be an amusing "wow can they do that?" niche hobby that was worth a look once every blue moon. Now it's a community full of people who take themselves far too seriously.
There must be a reliable max (Score:2)
If it stretches too far out of spec, it will likely start glitching.
Re: There must be a reliable max (Score:3)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
There is no upper limit. Whoever plays an old SNES is being accelerated and Special relativity confirms that time is different for their reference frame. The real question is, can you play it fast enough to go back in time to the year 1991?
Just replace it (Score:4, Interesting)
The resonator is not actually on the chip itself, but is a separate component on the board. Same with the crystal oscillator. You can replace it, much like a recap, to restore and preserve a retro game system.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
A lot of them have modified systems, e.g. HDMI mods for the SNES. A lot of them play on emulators too, with keyboard being a particular favourite controller for many.
For competitive play it all has to be recorded directly from the console video output, or from the emulator. Timing is established by counting video frames after the fact - the on-screen timers shown during runs are just so that the runner can tell if they are on record pace or not. It's called "re-timing" if you want to google it.
This is the audio processor, not the CPU (Score:1)
It affects music tempo (Score:4, Interesting)
In most Super NES sound drivers, the sound CPU's clock also affects the tempo of music. This could cause, for example, a jingle played after the player completes a course to finish 1 frame early. If the game is set up to wait for a jingle to finish before advancing to the next scene, this could make the next scene start a frame early.
well you are not allowed to use it to get high sco (Score:2)
well you are not allowed to use it to get high scores any more!
Won't affect speedrunnning (Score:4, Insightful)
The PPU (which draws the frame) is synced to the CPU clock on the NES. The APU also. The CPU clock is what determines the speed the game runs at (and fun fact there's multiple variants of the NES on the market and if you play a PAL game on an NTSC NES you'll get a game that runs 17% off - the speed running community already knows that and bans the practice.
What this error here will affect is the speed at which audio samples are played - incorrect pitch, and potential artefacts if the audio sample ends before expected.
Re: (Score:2)
I don't know it won't affect speedrunning. Speed running by humans, you are probably right. But TASes might be impacted.
If a game is written in a way where the CPU waits for an audio effect to be completed, then a variation in the clock speed of the sounds processor could shift the end of the sound to a previous or later frame.
Re: (Score:2)
PAL game on an NTSC
Amiga owners are very familiar with this phenomenon.
Re: (Score:2)
The PAL/NTSC thing is interesting. Most PAL ports of games just run 17.5% slower than the NTSC version, they aren't adjusted to be the same speed. So there is nothing to gain from running them on an NTSC machine, but arguably there is an advantage to playing them 17.5% slower on a PAL system.
For some games it has been decided that PAL and NTSC players will share a leaderboard, with times adjusted so that they are directly comparable. For other games they are separate.
ANALOG/ORGANIC COMPONENTS DECAY. No way to check. (Score:1)
Someimes even your kid notices the washing-machine makes less noice if moved half a floor tile thisa-way.
Your kid is not a genius. Laundry will not be affected. Kids shouldn't move appliances, but
DAMN THIS IS INTERESTING.
- TO discuss a ceramic resonance device whose natural degradation leads to CONSEQUENTIAL impact on eventual use (gameplay) without destroying it: awesome. Similar decay in other components (i.e. caps) leads to a different solution.
- TO suggest that "the Speedrunning community" do {the se
Well, what d you expect? (Score:2)
Even crystals age, but ceramic resonators are the ElCheapo resonator variant you use when you do not need precision and want to save a few cents. They are often as much as 0.5% off new (more precise ones exist). A cheap crystal is 0.01% off (100ppm) when new and more precise ones exist.