This topic is for the purpose of discussing #2753: MUGG's NES Super Mario Bros. "PAL" in 04:56.48

Could someone encode this run please. I do not have a working smb file to play this with.

I am conflicted. I will not vote. EDIT: I'm voting meh because I think it's suboptimal. I don't know whetherr this should replace ntsc. Also, I feel it's somewhat less entertaining than the ntsc version.

Encoded: http://www.mediafire.com/?pv8g2m2ip57u6uj

Eh, I was entertained. Yes vote.

In before happylee counters with a 12 frame improvement. EDIT: Flag glitch in 8-3 was cool.

I think I would like this more if you used a PAL-specific glitch, but as it is the published run is a bit more entertaining.

So, wait. To be clear, there are no PAL-specific glitches in the content of this run? If there aren't, this isn't sufficiently different to be accepted amongst the glut of currently published SMB material.

I thought getting into the floor was the PAL-specific glitch...?

He definitely uses a PAL specific glitch. Well, more precisely gets a PAL specific opportunity to use a known glitch on every flagpole. The fastest way to deal with a flag pole is to glitch mario to fall of it and run against it, bypassing the walking to the castle. In the U version this is only possible with the help of an enemy (such as a bullet bill, or shell). In the E version, they sped up marios speed per frame to compensate for the less frames per second (a common thing to do for PAL games). With that extra speed, he is able to perform this glitch on any flagpole without help from enemies. So that's the situation here. Is using E rom allows for this glitch and thus being overall faster despite being "slower" (less frames per second). So the debate is should this obsolete the U version, or be published along side, or be rejected. Please discuss.

It is a 85/100 second improvement over the publication. It's a yes vote from me.

After thinking about this for some time, I am quite certain that this run should not obsolete the U version. They are different games, or, more reasonably, this is a kludgy hack of the U version to make it work on different technology than it was developed for. I can go make another hack right now that triples mario's running speed, and a TAS of that hack might be really cool, since now you can do crazier jumps, but it would not obsolete the existing movie. As for what do with this submission, it would be a little sad to see it die. It is, in an annoying technical sense, the very fastest you can beat Super Mario Bros., where that name refers to some cartridge with that title published by Nintendo played on some hardware sold by Nintendo with some officially compatible television. I kind of like Mugg's suggestion in the submission comment that it be linked to from the U version page, if it is made very clear why it is faster. The main concern, I think, should be having new users find the best and most authentic run available first and most easily, and that is the U version.

Good points, but why can't I say the same about official English translations of Japanese games?

Honestly, in my opinion, you can. I'm pretty much for using the closest to the original game as possible, and for a lot of games that's going to be the J rom. Sadly, not enough of us speak Japanese, so in that particular case I guess I don't see a way around the sacrifice. In many cases, I would be happier with english subtitles/explanation file alongside a TAS of the J rom, but I don't realistically expect that to be a popular opinion. I guess the real difference is that while both choices (English vs. Japanese, NTSC vs. PAL) have the same theoretical answer ("use the one that came first"), in one case there is an issue that would massively undercut the appeal of the site/runs (the language barrier), and in the other case there simply isn't - and when a PAL run happens to have crazy different game breakage in it, I'm all for a separate category, as the rules already indicate. We sadly have to make sacrifices in the first case, but there's no reason to do so in the second as well.

They are different games, or, more reasonably, this is a kludgy hack of the U version to make it work on different technology than it was developed for.

I agree with this. This version of the game came out nearly 2 years after the Japanese original release and was modified to literally behave LIKE the ntsc version (hence how a PAL run can be beating an ntsc run). The run is faster because: 1. Mario accelerates quicker, 2. the "instant time starting-flagpole" glitch can be used everywhere. I don't think this run should be published, but I would like to see a link to it on the current ntsc publication. And not the other way around. In a sense, SMB was remade specifically to work for Europe in order to behave as close as possible to the ntsc version. It's quite literally a modified game. They didn't really come out at the same exact time, but years apart. Also , it's too late now I guess... but I'd change my "meh" to "no"

I like both runs. However, I like the (U) run more. First of all, to agree with what Personman's first paragraph said, it feels like the (U) version was the main version and (E) was a suboptimal derivative of it. The clearest indication of this is when 1-2 and 4-2 start, the music is finished long before Mario enters the pipe. The (E) version, for reasons like those, feels "unnaturally fast". Secondly, once you can do the "faster flagpole glitch" (the one that doesn't make Mario walk to the castle) at every opportunity, it cheapens the effect of the glitch. Consider, for example, if the warpless run were done in PAL. A run aiming for fastest time in PAL can probably perform this faster glitch at every level without doing anything too special. In contrast, the current NTSC warpless run needs to take the time to do setup with shells and bullet bills (or even with a cheep cheep in the minus-world run) to make this glitch work. These are also the most interesting and surprising parts of that run. If the PAL run doesn't require the player to be clever to get the best time at the flagpole, then an optimal run in PAL will not take time to do impressive stunts. This, I believe, makes a strong enough argument for why (E) should not obsolete (U). However, when it comes to publishing this alongside (U) or rejecting, I don't think the argument is as clearcut. I lean slightly against separate publication, because I don't feel like the (E) ROM's benefits make for a better-looking run (see my first paragraph). However, I am definitely in favor of linking to this from the (U) run, since players familiar with playing PAL versions of games should especially find this run interesting. All in all, then, I guess my vote is "No". However, I'm quite impressed with this run nonetheless.

Yeah, and even the "one that came first" idea is open to debate. Maybe the one that came first is the closest to the developer's original vision and the one everyone's most familiar with. But maybe later releases fixed things to be closer to the original vision than was possible before, and maybe those fixes made the game more fun. The NTSC->PAL thing really isn't about anyone's vision, though; as I understand it, it's about a computer clock speed relying on the AC frequency arbitrarily chosen by certain countries, so changes made for that reason should be disregarded if possible.

Since this is a PAL run, it shouldn't obsolete the NTSC run. It would have to be a separate publication if it has been accepted.

I agree. I don't think it should obsolete the NTSC run because it should be a different category.

I'm for this run being published but I'm inclined to put it in a different category

Apart from PAL making it easier to abuse one particular glitch and shaving a fraction of a second off from NTSC, what does this run demonstrate over and above the rest of the SMB content on the site? In my opinion, nothing. This does not showcase sufficiently groundbreaking PAL-specific glitches (such as with Blaster Master) to warrant publication as a new branch. It must either obsolete another run (which I can't necessarily see being warranted) or be rejected altogether. I am voting No.

This is an odd little situation here. The speed is a tad bit faster, BGM is faster while the SFX is slower. All because the game isn't exactly the same game, just shoe-horned in to making it work with a 50 mHz system rather than a 60 mHz system. While I was entertained because I got to see that flagpole glitch work more efficiently, I don't know if this is worthy of its own category or obseleting. Personally, I don't think it should obselete because it is like putting a normal kid against a kid who's been fed more sugar before a race, and seeing who comes out on top. It doesn't match up properly. Maybe if this run was a bit more optimized and everything squeezed out of it that we can (I do expect HappyLee to come in with that if this gets published, otherwise I do expect one on the way, he likes doing that) then I could confidently vote yes. Instead, I'm just going to have to vote Meh, not because it wasn't a good run, but because I can see a slightly better run coming in the future, unless this manages to disappear. Sorry I couldn't exactly support this (plus other, more active posters said it better before me...)

After reading the opinions presented so far, I decided to vote no. This run is faster, but it gains its speed by undermining the game's physics that made previous runs so enjoyable. HappyLee is rolling over in his grave at the thought of no more crazy koopa setups in SMB. However, this definitely does deserve to be linked from the published movie, along with an explanation of how and why it's faster. Ultimately, though, I just know that they must be pronouncing "World" and "Time" in pretentious British accents, and I can't stand for that. I want my runs in American English only please.

It must either obsolete another run (which I can't necessarily see being warranted) or be rejected altogether.

Let's be careful with our false dichotomies, good sir. Another valid option (among many) is for this excellent run to be published in addition to the NTSC version. How phenomenally interesting it is that the conversion to PAL resulted in easier performance of that flagpole glitch! I would be greatly saddened if this run were suppressed (as it seems so many people unfortunately and irrationally desire.) Information wants to be free, but it requires free minds to reside in. Close your eyes. Breathe deeply. Free your mind. Visualize people from all the world's nations, holding hands and freely sharing runs of the great ROMs they grew up with and grew to love. PAL and NTSC, living together in harmony ... we've done it!

Another valid option (among many) is for this excellent run to be published in addition to the NTSC version.

I explained earlier why this is not a valid option:

This does not showcase sufficiently groundbreaking PAL-specific glitches (such as with Blaster Master) to warrant publication as a new branch.

To expand upon this:

A run for a proposed new branch for a game should offer compelling differences relative to previously published runs of that game.