Oklahoma, USAThere are fundamental differences between PAL and NTSC, as you specifically mentioned in your post. I think the reason they're combined is that, while it's technically possible to do a faster run on PAL (The WR TAS uses PAL) - it's humanly impossible.
Practically speaking, PAL is a slower way to play the game, for humans. I can't remember if "full FPG" on PAL requires humanly impossible inputs (like several alternating left/right inputs on successive frames) or if it requires L+R (which is illegal). Someone who knows more about that than I do can clarify which it is. Either way, it's not possible currently. And if you DON'T use the full fpg (which is not possible anyway) - you'll be several seconds slower than a comparable NTSC run.
In regard to the original question about Switch. I think people get confused on the "it runs slower" "conversion" parts. If you use a stopwatch, and start the stopwatch at the "400" (beginning) and stop it at the axe disappearing (end), that's your time. The reason conversion has to be used is primarily having to do with mods timing the runs. The AXE PATTERNS used to find the exact frame a run is finished for runs under 5:02 or so, are tied to the framerate.
If you were to use the same inputs on a Switch and a original NES console on each frame, you'd end up with the exact same bowser axe pattern at the end (because you did the exact same things in each version). But the Switch would have finished a half second slower than the NES. There are formulas to correct for this used in timing.
Of course, that half second doesn't really matter if you're talking about a 5:23 run, since (1) they're only timed to the second at that point, not millisecond, and (2) it's such a small amount of time relative to the possible increases in PB. When you get down to the 5:0X range, and especially sub-5 range, that half second becomes increasingly important - which is why no "fast" runner would run on Switch, since it's silly to give up that half second - but it really doesn't make much of a difference if you're learning the game... except out of principle, or if you want to eventually go for a sub-5, and would like to learn on the system you'll end up on.
IF full fpg were to become viable for humans on PAL, I'm sure it would spark a discussion on separation of the boards, but for now, it's a moot point. PAL is a worse (slower) way to play the game... so why not include it in the leaderboard? It's literally a detriment.
I'd wait to hear from a mod directly... But... I don't know why it wouldn't be legal. Provided you can't do anything a controller can't - like L+R
Although I don't know how it would be possible to BEAT your pb with a less optimal controller. And I'm not sure that will fix the problem of enjoyment.
P. S. I'd suggest warpless. It's so much more fun than any% to me... And I see you haven't done it yet. And making a few mistakes doesn't RIP your run. It's a different animal. And you'll have several "free" pbs as you start off from 40 minutes down into the 20s
To piggyback on what @Skeeter said, but with more definitive language:
This leaderboard operates on a fairly strict whitelist of emulators and consoles - all shown in the rules:
NES Categories
Original NES, Famicom or FDS
Original NES or Famicom with Everdrive or Powerpak
Wii, Wii U, Switch and 3DS Virtual Console, GameBoy Advance
FCEUX, Nestopia, Bizhawk, OpenEMU
Frontends used in conjunction with these emulators that can alter the performance (ex. RetroArch) are not permitted
NES Classic Edition
Analogue NT Mini
To answer your question directly, you can not use the emulator you referenced, since it is not on the approved list - if you were to submit a run using it, it would be rejected. All of the approved emulators are accurate and easy to download.
As a person that used to verify runs and has seen the problems of "Beat 1-1 as Fire Mario" first hand, I can guarantee this will never be added to the boards here.
However, it absolutely can exist as a separate document, as roopert said previously on this thread
If any of you actually want to do this, make a Google Doc - make it editable by who you trust and consider "mods" of your project, and read-only to everyone else. Make a link to it, and let people submit runs to you. I can tell you, though, that this has been attempted in the past - and been abandoned, because of lack of interest / workload / difficulty of timing.
Another thing that's problematic with IL runs that many people probably don't think about, is that some people record at 60fps, some at 30fps, some at 24fps, some at 50fps, and some at weird rates. Unless you're recording and running at 60fps, you probably won't even see the correct start and finish frame on the video - in a category that requires precision to the frame.
I'm not a mod
But a mod hasn't responded to this yet, and unless a mod says what I'm saying isn't correct - I believe the following to be true...
I would not think that would be allowable. The ability to fake a run is way too high.
Also to quote directly from the rules "Using savestates to the title screen is prohibited. You must use the reset function on your emulator or console and the RESET MUST BE SHOWN IN THE VIDEO"
You can, of course, ALSO upload a video playback of your run this way if you like your run playback on 60fps for some reason - but make sure to record the entire run also.
I would fully expect a run consisting solely of a playback video to be rejected, and I personally believe it ought to be rejected.
Record your screen. If you can't record your screen... do like several runners do and have a physical camera recording the screen.
And then sure - IN ADDITION TO THAT - you can make a playback video using input files to put on your personal YouTube page, or in the description along with the legal video for verification.
There's a technique called "sniping" that if you don't use you'll die NEARLY every time, unless you do a point-blank shot.
You need to hold JUMP, then press FIRE, while CONTINUING to hold jump, then release jump. This will fire a fireball halfway through your jump, allowing the ball to lead your trajectory. Otherwise the ball will be ABOVE your trajectory and you'll die to whatever is in front of you.
It's an odd feeling to get used to. WHILE HOLDING JUMP, you need to fire, then AFTER having fired, release jump - so you'll continue moving upward after having fired.
The more natural feeling action is to release jump at the same time as the fireball, but that will end up having the ball too high in front of mario.
Mods never take the claimed time as the actual time. Every single run gets retimed whether it's a 4:57 or a 6:20 - a time off by 0.17 on start or axe is definitely within the range of normal for error.
Just do your best to put your estimate of what you think the run is (as you have done) and the mods will take care of the rest. No worries :)
Just from what I can see from this post, without giving you a list... there are numerous red flags which would point to your run being faked in some manner.
A sub-5 minute run is a major accomplishment, which typically requires at least months of daily grinding.
Just know the mods will view this run with a high level of scrutiny.
Additionally, to copy/paste straight from the Category rules:
"Times below 5:00.00 are treated with suspicion. It is highly recommended that you stream all of your attempts and provide split files to have your run accepted."
I don't mean this to come off like a challenge... but I don't even know how you could cheat using this video setup. CRT monitors (obviously) build pixel by pixel line by line, so splicing would be easily detectable, as well as hand positioning. I can't think of a more cheat-proof way of filming tbh. Unless there's something I'm missing. Seems very safe - assuming you're also picking up sound, which I assume you are.
But I'm not a mod.
It does emulate games 100% accurately. At least for the whitelisted emulators.
There are several pros and cons for emulator.
The main benefit of emulator is the convenience factor - you don't have to buy a NES, and a CRT, and the game cartridge, and the cables to put it to your computer, you just download FCEUX or Nestopia. Easy
The main problem with emulator is input lag - there is existent input lag while using an emulator that is not present for NES + CRT - which makes the tricks a little harder. However, your brain tends to adjust to the lag, and your muscle memory will adapt to the slight lag in time even for hard tricks like flagpole glitch.
The main benefits of console are the reduction of input lag and --- the purity of the console in its original form (if that is important to you).
The problems are that arguably using a controller can be more difficult because of the placement of the D-pad and buttons (or easier, depending on who you talk to). Also it's more work to setup, and more expensive to acquire.
Still - there are many unbelievably good runners that are almost exclusive to the emulator. @Niftski, @Lul_ecks_dee, and @Tecate to name a few. All rediculously talented.
You actually don't need splits at all for verification.
If you have a suspicious looking run for any number of reasons, a full splits file can help provide evidence in your behalf, but it is not required.
I'd save the old splits files regardless since they hardly take up any hard drive space, but certainly you can start fresh on a new file.
You need a combination of
gvusb2 - https://www.amazon.com/DATA-connection-video-capture-GV-USB2/dp/B00428BF1Y
and
You'll split the audio (red/white) and video (yellow) signal from your NES to go to both the TV and your computer via the GV-USB2 (or equivalent device)
You CAN show the display on your computer only... but it will have horrendous unplayable lag. The video going to your computer will be for the purposes of showing on OBS only.
Neither are accepted as far as I know.
There is a very specific whitelist of accepted emulators and hardware.
(Below copied from rules)
NES Categories
Original NES, Famicom or FDS
Original NES or Famicom with Everdrive or Powerpak
Wii, Wii U, Switch and 3DS Virtual Console, GameBoy Advance
FCEUX, Nestopia, Bizhawk, OpenEMU
Frontends used in conjunction with these emulators that can alter the performance (ex. RetroArch) are not permitted
NES Classic Edition
Analogue NT Mini
That isn't to say that those two are (or are not) accurate... but there's a lot of work that goes into adding emulators or hardware as "acceptable" - and there have even been emulators which weren't accepted, then were, then weren't (Mesen)
In a game where individual frames make a huge difference, the level of accuracy is extremely important - if an emulator emulates everything completely accurately EXCEPT lag, that's a huge problem. If an emulator is extremely accurate, but has a feature that could be exploited (Mesen), that's a problem.
That being said - I think AVS at least (since I have experience with it) is probably pretty accurate for a 60fps system (like NES classic), but I don't know that for sure.
(1) This is a comment for https://www.speedrun.com/smbce (2) They're not accepting new categories (3) 1-1 as Fire Mario exists (4) Warpless effectively already IS beat the game (without warps) as fire mario (5) You can run any% as fire mario, and a lot of people do just that in the slower time runs on the boards
Curved screen with the big bulky back - yeah.
(Although CRT can have flat screen also)
It generates an image in a different way than the modern LEDs - has a response time of pretty near zero... as opposed to LED's which have a slight lag in the creation of the screen from an input.
Slow Mo Guys have a really great YouTube video on the differences filmed in extreme slow motion between CRT and LED... and even features SMB1 on an NES
^ animated GIF of CRT generating an image. LED generates FRAME by FRAME, CRT generates pixel by pixel, from left to right
It's also why you can't play duck hunt (or anything with the light gun) on a LED tv... and you can on a CRT
What @LeonD011 said
You can find original NES consoles on Facebook Marketplace, Craigslist, Retro Game Stores (in the middle of the US, there's a store "Vintage Stock", but others in other areas of the US)
Good quality NES console with controller and Mario/Duck Hunt combo should be around $100 give or take iirc.
You'll also have to procure a CRT tv - otherwise you'll have a LOT of input lag (if you plug the original NES into a modern LED tv)
That being said - unless you're wanting to buy a console out of principle, or plan on running into the sub-5 range... Wii VC or NES Classic may be an easier and more straightforward option.
Both of the above DO run at 60fps instead of a smidge under (original framerate), which will lose you a half second over a 5 minute run. However, if you're wanting to run it and aren't planning on REALLY pushing the speed to the limit - a half second doesn't really matter much. Milliseconds aren't even included until 5:02.x anyway.
The FREE option is to run it on emulator... and the framerate will be the correct framerate. The downside is that there's existant input lag, and it's not "original" if that kind of thing bothers you - but there are plenty of runners that run emulator exlusively, and are EXTREMELY good.
At the same time, there's just something about having an ORIGINAL nes - rather than emulator or a re-creation. Also - the original will have virtually zero input lag, whereas most re-creations have more for a variety of reasons.
Tbh I miss seeing these rediculous runs as a mod. Thanks for letting me have a glimpse back into the world of asshats turning in fake runs. Not sure if you actually thought this tas would be accepted or if you're doing it for attention, but either way - it's a tas and cheating.
