I'm not sure it's actually "resetting" the game and I think RoboSparkle should be able to do that. It's pretty much the equivalent of intentionally dying in order to reload a certain part of the game or respawn somewhere else. Correct me if I'm wrong but I think they do this in some of the Zelda runs.
SDA Forum
umad
The thing about MM is that literally no MM player does single segment anymore. All of them do SS w/ resets because they don't give a shit about SDA. (I'm one of like the only 5 zelda runners who still cares about SDA lol). I feel like allowing SS w/ resets for MM would gain SDA more respect among the zelda community and possibly get more runs submitted. But in the end it's Flip's decision.
In Portal you can leave the elevator in chamber 14 earlier if you save+load right after the level started. That saves about 1 second compared to a normal IL run. Will this now be allowed for IL runs? And will the built-in timer count or will it be timed manually? And would that count as a separate category?
1-Up!
zastbat, resets save 5-10 minutes versus how long of a run without resets?
djcj, no for portal ILs
Indextic, what UA said
Miles, I see where you're coming from, but it's unfair to all SS runs on the site if we suddenly allow SS w/ resets to compete directly. The alternative (hosting both) is simply redundant. A run with resets that is marginally faster (<5% faster than a run without resets) doesn't contribute enough to the overall value of the site to be worth hosting in my opinion. Maybe that will change down the road, who knows.
djcj, no for portal ILs
Indextic, what UA said
Miles, I see where you're coming from, but it's unfair to all SS runs on the site if we suddenly allow SS w/ resets to compete directly. The alternative (hosting both) is simply redundant. A run with resets that is marginally faster (<5% faster than a run without resets) doesn't contribute enough to the overall value of the site to be worth hosting in my opinion. Maybe that will change down the road, who knows.
I don't buy that "it's unfair" argument. Should we have not allowed OoB or "mysterious warping" unless they saved >5%, too? And if the only reason a run is better is because of resetting, isn't that what verification is for?
Right now this just strikes me as "Get the fastest run you can, using everything at your disposal! Except for this. Which isn't against the rules, we're just not letting you use it anyway." I'm completely expecting this to go the way of OoB eventually, though it'd be nice to see that happen sooner rather than later.
Right now this just strikes me as "Get the fastest run you can, using everything at your disposal! Except for this. Which isn't against the rules, we're just not letting you use it anyway." I'm completely expecting this to go the way of OoB eventually, though it'd be nice to see that happen sooner rather than later.
Is PJ
The main reason I think this thread should exist is because some games really don't gain time because of a trick that is only possible by resetting/saving & reloading. I feel like if there was a blanket ruling that resets are allowed in SS runs, we'd get a slew of runs that reset just to get another try at a specific trick or to manipulate luck. I agree that there shouldn't really be a "minimum" amount of time that resetting needs to save, but I do think it should be approved first instead of allowing it for every game.
Just my $0.02.
Just my $0.02.
Gets the cake.
As someone who's completely out of the loop here, I don't see a problem with resetting to retry a trick or manipulate luck. There are several games where the savings of such an action are beneficial on the magnitude of minutes, rather than seconds. That said, if the run looks ugly, slow, etc. because of the resets, then that is an issue that could be addressed in verification.
Take Wind Waker for example. A Wind Waker run can be VERY good, even with resets, just due to the fact that a reset would be 20 seconds of lost time out of the 5.5 hour run time. But, unless you want to start over every time you don't get a proper warp, or every time you miss getting storage on the first try, resets are almost expected.
Take Wind Waker for example. A Wind Waker run can be VERY good, even with resets, just due to the fact that a reset would be 20 seconds of lost time out of the 5.5 hour run time. But, unless you want to start over every time you don't get a proper warp, or every time you miss getting storage on the first try, resets are almost expected.
Is PJ
Yea Wind Waker is the one example I thought of for retrying tricks, actually. I guess the main thing I'm thinking of would be games like 7th Saga (sorry, not my fault that is the first thing that comes to mind) where you can save time by resetting every time you get boned by even minor bad luck. The end result would be a run that is actually faster than a run that accounts for this luck and doesn't reset, but looks much sloppier.
Edit: What Aftermath said.
Edit: What Aftermath said.
1-Up!
Quote from Miles:
I don't buy that "it's unfair" argument. Should we have not allowed OoB or "mysterious warping" unless they saved >5%, too? And if the only reason a run is better is because of resetting, isn't that what verification is for?
Biggest difference in this case would be the number of runs that would be affected. I'm not about to go and count, but I'd say that many more runs on the site could be obsoleted with a simple savewarp (given a comprable level of run quality) than by OoB or mysterious warping. On one hand I see the argument that SDA is about producing the fastest runs but on the other I think we have to do something to respect all of the current and former members who worked hard to submit runs that are/were optimized under the current rules.
Re: Verification - Ideally, but what does it take for a SS w/ Resets to obsolete SS? When should they both stand? These are things we need guidelines for. It's a lot of grey-area to cover.
Quote from Miles:
Right now this just strikes me as "Get the fastest run you can, using everything at your disposal! Except for this. Which isn't against the rules, we're just not letting you use it anyway." I'm completely expecting this to go the way of OoB eventually, though it'd be nice to see that happen sooner rather than later.
5% was an arbitrary mark that I created. I won't claim that it's perfect but it seems to work in the way we intended. Maybe it's not the best way. I'm always open to suggestions but "this is stupid" or "we shouldn't do it this way" doesn't help me much. If anybody has suggestions feel free to discuss.
Gets the cake.
Quote from Flip:
5% was an arbitrary mark that I created. I won't claim that it's perfect but it seems to work in the way we intended. Maybe it's not the best way. I'm always open to suggestions but "this is stupid" or "we shouldn't do it this way" doesn't help me much. If anybody has suggestions feel free to discuss.
I would like to note that, now that I'm aware of the %, then the 1s savings in portal adv ch 14 is more than 5% of that IL run (full chamber being like 15-ish seconds long). Unless IL runs don't count for this category at all, in which case I'm sorry for second guessing you. Just wanted to drop some expertise in >_>
Edit history:
kirbymastah: 2012-07-19 09:16:49 pm
kirbymastah: 2012-07-19 09:16:43 pm
kirbymastah: 2012-07-19 09:16:21 pm
kirbymastah: 2012-07-19 09:15:59 pm
kirbymastah: 2012-07-19 09:14:22 pm
kirbymastah: 2012-07-19 09:11:59 pm
kirbymastah: 2012-07-19 09:11:42 pm
kirbymastah: 2012-07-19 09:10:14 pm
<(^_^)>
If I may suggest something, we have the data for how much time savewarping and such saves for many games, so... for the most part, if a submitted run with resets doesn't obsolete the current SDA run by more than that amount of time, it should at least not replace it.
So if I'm playing wind waker, and the current record is 7 hours, and savewarping saves 30 minutes (I'm pulling random numbers out of nowhere here), and I get a time of 6 hours, 45 minutes, then it should most definitely not replace the current record (put it into a separate category? that's debatable). But if I get a time of less than 6 hours and 30 minutes, then IMO it should replace the current record, since not only does it use a "better route," but also performance overall was improved.
But yeah, overall I personally still see resetting as simply a form of better re-routing, so if it's overall performance is better, then there shouldn't be an excuse to just outright replace the old run. But again, I can see why you'd want to do different categories, to discourage people from submitting runs with worse quality that are faster just from better routes, but IMO that should be left to verification. Basically,
So if I'm playing wind waker, and the current record is 7 hours, and savewarping saves 30 minutes (I'm pulling random numbers out of nowhere here), and I get a time of 6 hours, 45 minutes, then it should most definitely not replace the current record (put it into a separate category? that's debatable). But if I get a time of less than 6 hours and 30 minutes, then IMO it should replace the current record, since not only does it use a "better route," but also performance overall was improved.
But yeah, overall I personally still see resetting as simply a form of better re-routing, so if it's overall performance is better, then there shouldn't be an excuse to just outright replace the old run. But again, I can see why you'd want to do different categories, to discourage people from submitting runs with worse quality that are faster just from better routes, but IMO that should be left to verification. Basically,
Quote from Miles:
And if the only reason a run is better is because of resetting, isn't that what verification is for?
If you're playing Wind Waker, and the no-resets record is 7 hours, and savewarping saves 30 minutes, and you get 6:45 with resets, then I agree that that run shouldn't replace the no-resets record. In fact, I think that run shouldn't go on the site at all because that would mean you made at least 15 minutes of mistakes.
I can only see this as unfair to previous runners if
1. Someone does a run that is only faster than the current run due to advantages from resetting,
2. The verifiers actually accept the run, and
3. It isn't a separate category.
Honestly, I really don't think the first two is going to be a common issue, and somewhat relies on the verifiers sucking. In the rare case that a run is too difficult to determine, either it can be a separate category, which shouldn't make much of a difference in the long run since this would occur with very few runs; or judge runs with resets when there's a current run more harshly (you have the final say in verifications, right?). However, right now you're being unfair to current runners, by arbitrarily deciding whether or not they can make their runs as fast as possible.
I think the opposite situation will be a lot more common: A run saves a lot of time over the previous outside of the resets (like, let's say resets save 30 seconds and the run is 5 minutes faster). At the moment that still wouldn't be allowed, when I can't see how that would be unfair to the previous runner at all. Edit: Also, it currently stops games that don't have a run up on the site in the first place.
1. Someone does a run that is only faster than the current run due to advantages from resetting,
2. The verifiers actually accept the run, and
3. It isn't a separate category.
Honestly, I really don't think the first two is going to be a common issue, and somewhat relies on the verifiers sucking. In the rare case that a run is too difficult to determine, either it can be a separate category, which shouldn't make much of a difference in the long run since this would occur with very few runs; or judge runs with resets when there's a current run more harshly (you have the final say in verifications, right?). However, right now you're being unfair to current runners, by arbitrarily deciding whether or not they can make their runs as fast as possible.
I think the opposite situation will be a lot more common: A run saves a lot of time over the previous outside of the resets (like, let's say resets save 30 seconds and the run is 5 minutes faster). At the moment that still wouldn't be allowed, when I can't see how that would be unfair to the previous runner at all. Edit: Also, it currently stops games that don't have a run up on the site in the first place.
1-Up!
Quote from Paraxade:
If you're playing Wind Waker, and the no-resets record is 7 hours, and savewarping saves 30 minutes, and you get 6:45 with resets, then I agree that that run shouldn't replace the no-resets record. In fact, I think that run shouldn't go on the site at all because that would mean you made at least 15 minutes of mistakes.
This exactly.
I'll comment on the rest later.
Lord Of The Beards
Quote from Flip:
Quote from Paraxade:
If you're playing Wind Waker, and the no-resets record is 7 hours, and savewarping saves 30 minutes, and you get 6:45 with resets, then I agree that that run shouldn't replace the no-resets record. In fact, I think that run shouldn't go on the site at all because that would mean you made at least 15 minutes of mistakes.
This exactly.
I'll comment on the rest later.
When you have the chance i would like to know if Soul Reaver qualifies for SS with resets so that i can find out if I am going to Segment my runs or not.
Sorry if i sound like a dick i couldn't think of a better way to word what i wanted to say.
Claimh Happy
For what little it's worth, I would just like to second Miles' opinion that this is essentially OoB all over again. Eventually SSw/Resets will be the norm, and it will be up to the verifiers to ensure that when a run with resets is put up against one without them, the new run must be a legitimate improvement in order to be accepted. I would very much like this to happen sooner rather than later.
That said, I don't think this topic is useless at all. The runs approved in this topic will be the ones in which resets constitute an alternate category. Even after resets are generally accepted this topic will be rather important.
On a side note, I disagree that a menu reload that clearly does not do anything with another file (unless created during the run) should count as a reset unless you actually have to reset to reach the menu. If it's used to cover up a mistake, the verifiers can deal with it. I don't see why any other use would be controversial.
That said, I don't think this topic is useless at all. The runs approved in this topic will be the ones in which resets constitute an alternate category. Even after resets are generally accepted this topic will be rather important.
On a side note, I disagree that a menu reload that clearly does not do anything with another file (unless created during the run) should count as a reset unless you actually have to reset to reach the menu. If it's used to cover up a mistake, the verifiers can deal with it. I don't see why any other use would be controversial.