This topic is for the purpose of discussing #4159: McBobX's NES Mega Man "low glitch" in 20:35.03

I'm wasn't not surprised by rerecords

I'm more concerned about the fact that this run actually does use glitches: pause tricks to prevent damage knockback and to reset boss invulnerability periods. Perhaps the branch name should be changed to something that describes the run properly, like "no DelayStageClear, no zipping". I'm working on an encode.

I'm more concerned about the fact that this run actually does use glitches: pause tricks to prevent damage knockback and to reset boss invulnerability periods. Perhaps the branch name should be changed to something that describes the run properly, like "no DelayStageClear, no zipping". I'm working on an encode.

Thank you for this review. About the encode,I have done it in HD but my connection is very bad so I can't upload it (if want to upload a video,I go to cybercafe to do so)

About the encode,I have done it in HD but my connection is very bad so I can't upload it (if want to upload a video,I go to cybercafe to do so)

I understand. I have a stable connection here, but a very slow upload speed, so I usually go to school when I need to upload YouTube videos. Regardless, I will be able to upload an HD encode, whenever it finishes (currently at 4% and slowly rising).

EDIT: I see the name has been changed to "no zipping". What about DelayStageClear?

I presume that was an oversight by the author. I'll fix it for him.

I changed it to "low glitch" which is the convention for movies that avoid game breaking glitches while still doing minor time saving ones. Imo though, this is glitchless.

I'm more concerned about the fact that this run actually does use glitches: pause tricks to prevent damage knockback and to reset boss invulnerability periods.

The first one you mention is a pretty clear cut case of a "trick", not a glitch. The 2nd I would argue as a trick too, I don't think it is unintended behavior that select doesn't freeze timers, and the trick simply takes advantage of that observation. But I would be willing to concede to a minority opinion on that one, but the first is clearly not a glitch! Also, I very much agree with the glitch choices in the goals of this movie. And glad to see this TAS finally made.

What's the difference between a trick and a glitch? Those two things seem to fit our site's definition of a glitch. Also, the encode is complete and has been added to the submission text.

On what basis is zipping a "glitch"? It's a feature deliberately coded into the game, so it's not abusing an unintended bug. (Sure, it wasn't intended to be used to zoom through walls and levels, but it's still something that was deliberately coded into the game, rather than being a programming error.)

I remember this run [111] NES Mega Man by Bisqwit in 18:21.03 Can anybody compare the parts of no zipping run? Famtasia is very hard to use.

The run reminds me of that one "buster only, no glitches" run we had previously and it does have to face similar issues. Right off, this run is heaps better than the other one because it is actually allowed to use subweapons, allowing for a smoother run. I actually did enjoy seeing some of the Magnet beam shortcuts in the run. The problem with a category like this to define where exactly to draw the "low glitch" line. The pause glitch clearly is allowed in this run (with useful application), as is the block duplicating glitch with Gutsman's weapon. (At least, I think that's a glitch.) They are definitely not as severe glitches as zipping, and everybody will have used the Pause glitch at least once. The question is whether they add enough entertainment to the run to be considered acceptable for a "low glitch" category. The run looked decently optimized for the most part, except that a few of the jumps didn't look as tight as they could have been. Avoiding enemy drops is key here. Also, in the Fake Mega Man Battle in Wily 2, it looks like the boss should have died in the center of the screen so the screen would clear faster. I can't tell for sure, however. Overall, I found this run entertaining enough to vote yes. I'm betting it could be optimized a bit more, but with new category submissions like this, it's always about whether the category is acceptable or not. With that said, I'm always in favor of seeing new category choices, which is why I'm happy for this run.

You used the pause glitch to beat the game which violates the runs objectives so no vote.

The problem with a category like this to define where exactly to draw the "low glitch" line.

Yes. The line is necessarily quite arbitrary because it depends completely on what's considered an "acceptable" glitch to use in this category and what isn't. And who gets to decide what is and isn't acceptable? Perhaps the runner himself? Although that in itself isn't really the core problem. We could just outright accept the runner's decision and publish it as "low-glitch". The real problem is what happens if someone makes a new submission that uses a different set of glitches (and avoids the rest)? Should it be accepted? Should it obsolete this one? Who makes this decision? We see here that the runner making the decision of what's an acceptable glitch and what isn't causes a problem. Another runner may have a different opinion, and who's to say which one is right or wrong?

Watched it, enjoyed it, think it should be published to Moons. Goal choice will be necessarily a little woolly in a submission like this, but it plays through the game in the spirit of not breaking it, and does so in a way that is not noticeably suboptimal. About potential obsoletion, you treat it as follows: say somebody includes some new trick in a future movie that McBobX considered a glitch, and that glitch saves ten seconds. Then the new movie obsoletes this one if and only if it is more than ten seconds faster than this movie.

I liked it, not as much as I thought I would though. I forgot that Cutman had a level. The level of optimization was not quite as high as the existing run. Not enough to not be published imo, but noticeable nonetheless. Examples: Clear Points, Fireman boss fight end could be closer to the middle, maybe a platform that could be jumped over the top of the screen in fireman, maybe better dropping-off-magnet-beam in iceman, maybe killing Wily 2 boss closer to middle (not sure if explosion leaving the screen matters in MM1), maybe better boss re-fights in Wily 4 getting back to the exit hatch, maybe falling in cutman's level on the right despite spikes and using the beam to go left instead of going left on a couple screens above. I'm okay with pausing for damage invincibility skip. Some boss battles would be really long and repetitive without it (Wily 1 definitely, Wily 4 maybe).

I think it would have been better without the block duplication and pause while hit tricks. Then it would also fit the "glitchless" definition of real-time speedrunners and I can't think of anything besides these that is a glitch or borderline-glitch. The submitted run seems unoptimized, but the route is interesting.

This movie is conceptually something between our very first 2 Rockman publication. It uses some tricks [5] NES Mega Man by Morimoto in 21:52.10 didn't know about, but even in 5M there were some screen wraps this movie avoids. Then, [111] NES Mega Man by Bisqwit in 18:21.03 uses the same pause glitch as here, but it already does plenty screenwraps and zips. Optimization level was not compared, but in fact - is it anyhow visible? I fail to notice how this run is superior to any of the first 2 publications. Primarily, why would we want a run that avoids some time-saving strats? Because it looks more like the old runs we had here, but it applies the current optimization level to them, and tricks that were not known back then. And the result does look different than the old runs. This applies to pacifist runs very well, because the difference is obvious, and the new category requires interesting route changes that are noticeable in real time. Hence, we do like such a movie and it is accepted to Moons. The same sometimes happens to low glitch runs. Latest runs are so glitchy they skip significant amount of gameplay, so people are interested in how would TASing the whole gameplay look like. [862] Genesis Pulseman "no motion glitch" by IdeaMagnate in 33:57.15 [1937] Genesis Sonic the Hedgehog "no zips" by Aglar in 17:36.58 So what is so modern in this submission that was not seen in the old runs? Optimization level alone? It's very hard to tell by an untrained eye, or rather, without comparing every scene frame by frame. Then what is so old-school-ish in this run that isn't seen in modern ones? No screen-wraps/zips/memory corruption? But pause glitches are still here. They still don't let us see how would a modern TASer beat this game, within intended route. I'd say, if one really has something to offer about this game, no glitches should be used, to actually differ from all movies we had. BTW, it's not even about what is possible in real time, because many glitches avoided here are possible. But what can be argued is, what glitch to allow? I'd say, all pause abuses must be avoided too. They were likely not presented as a feature, even though it was programmed this way. And we've already seen them since the second publication of this game. They add nothing new. But again, it all would work only if a runner has something to offer. Which must be possible to tell in real time. In real time I only see that this is probably faster than Morimoto's run. But I'm not sure it deserves a category like this. So my vote is not just No. I vote against this glitch choice. This run seems to be liked, but did anyone ask himself a question: What does this run show that was unseen? This way we could even unobsolete some of the old runs, [111] NES Mega Man by Bisqwit in 18:21.03 is still faster than this one.

I'd say, if one really has something to offer about this game, no glitches should be used, to actually differ from all movies we had.

What about a Magnet Beam-less run (which has been suggested by Nach a few times)? It skips out most major zipping, which means most parts are already much different from what we've had now, it may allow some completely new routes/tricks in some stages, and is also interesting in that it's a concept impossible to perform in real time (as DelayStageClear is required to pass the parts that require Magnet Beam). It also doesn't run into any issues of forgoing arbitrary tricks/glitches or other arbitrary restrictions. (In fact, you could even quantify it as low%, a completely objective category.) The only big problem with this category is that it requires DelayStageClear and other really big glitches, which makes it hard if not infeasible to TAS for anyone not named FinalFighter.

Oh come on, a "low glitch" category now? That's about as arbitrary as it gets. Why not go for a "40% glitch" run instead?

On watching the run fully, it really doesn't give off a "avoids glitches" feeling with the extensive abuse of pause to cancel hit animation (which I consider a glitch) and pause glitch on mercy invincibility (which is certainly a glitch) and some other glitches. It's even more jarring with fights like Guts Man, where the trick isn't even necessary for an optimal run (unless it happens to save lag, not sure) yet abused extensively. The end result looks like it just really arbitrarily skips using some glitches while still heavily abusing other glitches. It's just very arbitrary, and I don't think it's publishable this way. No vote.

After watching the whole run, I agree that this run subjectively chooses some glitches while forgoing others, which makes it too arbitrary a category to be published. No vote.

I changed it to "low glitch" which is the convention for movies that avoid game breaking glitches while still doing minor time saving ones.

"Low glitch" is way too arbitrary a category though. That's why all published runs that avoid a certain glitch (like the ones feos linked to) have the glitches they avoid clearly spelled out in the branch name.

No matter how much whinging about "low glitch" and "look at what glitches you actually used" and all that, it is undeniable that this run provides an interesting and watchable alternative to the sense-destroying runs this game has had in its last few TAS iterations. Frankly I think it should be published and "low glitch" is a fine classifier, because to a viewer it's readily understandable and has high descriptive value. Further specification can be done in the submission text, i.e. a guide to which glitches have been used and which have been deliberately avoided, but as I recall entertainment value and precise execution are the most valued qualities here as stated--let those be the guidelines for further submissions that contest this one as well; any yelling about glitch selectivity is missing the point. So yeah. Voting yes.

Oh come on, a "low glitch" category now? That's about as arbitrary as it gets. Why not go for a "40% glitch" run instead?

I agree; we shouldn't use that label, and instead label it "no foo glitch, no bar glitch" or something.

The real problem is what happens if someone makes a new submission that uses a different set of glitches (and avoids the rest)? Should it be accepted? Should it obsolete this one? Who makes this decision?

I suppose the burden is on the person submitting the new movie to show that it's entertaining enough (and different enough from the existing movies) to be accepted for Moon Tier. Then the judges can decide whether it should obsolete an existing movie or start a new branch. That said, this particular movie strikes me as rather arbitrary, so no vote from me.

Regarding the movie's entertainment, I vote "meh". Now I know there are runs that don't zip all over the place yet still be enjoyable to watch, but with how the current run zipping through a whole lot (while still going through all the stages unlike the first megaman), it was kind of difficult for me to be in awe.

Echoing a few others in this thread, I don't really have a problem with the consistency of the goals. The only glitch that I have some issue with is the reuse of Guts blocks in the Elecman and Wily 3 fights. The two big optimization issues that stood out for me were: * Getting stage bonuses above 5000 points * The Wily 3 fight is also noticeably suboptimal more so than others, I find. That being said, I like the way this run looks. I would want to see a slightly more polished version later, but this is good enough for now, and I'd say it can belong in Moons. By publishing this, it may also encourage other people to improve it more than if it is rejected.