Addition of "one segment per map/level category" [locked]

2009-07-12 05:08:12 am

Quote from Spider-Waffle:

Regardless, if more categories has either no relationship or a direct relationship on the quality of runs what is the argument against more categories?

It means viewers have to choose between several runs without knowing which is the best.

Quote:

Can you point out any example where more categories lead to the best run for that game being worse?

That's a pretty unfair challenge. How could you ever directly show cause and effect here?

Quote:

I think the type of runner that makes a lesser run on another category instead of beating an existing run on a more prominent category most likely didn't think they could beat the existing run anyway because they lacked some combination of skill, discipline, patience, time. So they likely wouldn't have ever contributed a run at all if the different category never existed.

And that would be better than them submitting a run inferior to one that already exists, in my opinion.

Quote:

The one segment per level category would likely if anything condense the speedrunners efforts to that very category. For, example, a game which there would be one run for, it would most likely be on the one segment per level category. Nearly all save anywhere games, especially newer ones, runners aren't even willing or capable of making a SS run anyway. Think about it, how many save anywhere games have SS runs, I actually would really like to know which ones do? So the runners would basically be deciding between one segment level or segment anywhere; in the case of FPS, RTS, and puzzle (maybe other genres) save anywhere games, it seems to me that one segment per level is preferred, you would see a lot more runs for it if it existed.

I entirely agree with this. Being able to segment anywhere can be kind of lame because almost by definition you'll never come across any really difficult challenges because you can just segment them away (unless you choose to have arbitrarily long segments and bring those challenges on yourself, which you ultimately get no reward for doing - the run just ends up looking worse anyway.)

To VorpalEdge and ShadowWraith - are you mad? What could possibly be gained by SpiderWaffle beating those runs by using mass segmentation? It would be unfair and the end result would be much less impressive than what is already on SDA.

[quote="Spider-Waffle"]Regardless, if more categories has either no relationship or a direct relationship on the quality of runs what is the argument against more categories?[/quote]

It means viewers have to choose between several runs without knowing which is the best.

[quote]Can you point out any example where more categories lead to the best run for that game being worse?[/quote]

That's a pretty unfair challenge. How could you ever directly show cause and effect here?

[quote]I think the type of runner that makes a lesser run on another category instead of beating an existing run on a more prominent category most likely didn't think they could beat the existing run anyway because they lacked some combination of skill, discipline, patience, time.  So they likely wouldn't have ever contributed a run at all if the different category never existed.[/quote]

And that would be better than them submitting a run inferior to one that already exists, in my opinion.

[quote]The one segment per level category would likely if anything condense the speedrunners efforts to that very category.  For, example, a game which there would be one run for, it would most likely be on the one segment per level category.  Nearly all save anywhere games, especially newer ones, runners aren't even willing or capable of making a SS run anyway.  Think about it, how many save anywhere games have SS runs, I actually would really like to know which ones do?  So the runners would basically be deciding between one segment level or segment anywhere; in the case of FPS, RTS, and puzzle (maybe other genres) save anywhere games, it seems to me that one segment per level is preferred, you would see a lot more runs for it if it existed.[/quote]

I entirely agree with this. Being able to segment anywhere can be kind of lame because almost by definition you'll never come across any really difficult challenges because you can just segment them away (unless you choose to have arbitrarily long segments and bring those challenges on yourself, which you ultimately get no reward for doing - the run just ends up looking worse anyway.)

To VorpalEdge and ShadowWraith - are you mad? What could possibly be gained by SpiderWaffle beating those runs by using mass segmentation? It would be unfair and the end result would be much less impressive than what is already on SDA.

ridd3r.

2009-07-12 06:03:46 am

we have lift off

Quote:

To VorpalEdge and ShadowWraith - are you mad? What could possibly be gained by SpiderWaffle beating those runs by using mass segmentation? It would be unfair and the end result would be much less impressive than what is already on SDA.

I think they just know he won't do it because the whole thing which has come out of this discussion is that runners know the score with segmentation. Therefore I thought most people understood that apart from maybe a slight change in wording in the rules nothing needs to be done as it is a theoretical problem which has never actually arisen.

dex

2009-07-12 07:10:11 am

Invisible avatar

Quote from Spider-Waffle:

The one segment per level category would likely if anything condense the speedrunners efforts to that very category. For, example, a game which there would be one run for, it would most likely be on the one segment per level category. Nearly all save anywhere games, especially newer ones, runners aren't even willing or capable of making a SS run anyway. Think about it, how many save anywhere games have SS runs, I actually would really like to know which ones do? So the runners would basically be deciding between one segment level or segment anywhere; in the case of FPS, RTS, and puzzle (maybe other genres) save anywhere games, it seems to me that one segment per level is preferred, you would see a lot more runs for it if it existed.

AvP2: Primal Hunt, Baldur's Gate: TotSC, Doom II (two runs), Morrowind, Fable (I think it had save anywhere), Fallout 1&2, Half-Life, Max Payne 1&2, Portal, Quake (4 categories), Rise of Nations had saving IIRC, Serious Sam: TFE, Silent Hill 2 (I think there was save anywhere on the PC, though I'm not sure since I always used the regular saving system, could someone confirm? The SS is on PC), Uplink. I might make an SS of Deus Ex, and my super-duper-secret project is actually aimed at obsoleting one segmented run with a single segment (though whether I'll manage this is another matter ;)). That doesn't even include games where save points are very easy to use, but the runners choose not to; or games where the PC version has save anywhere but there's no PC run so I omitted them (MGS springs to mind, it takes 1 second to save there, and you can save on most screens).

The runners currently can decide between a single-segment, or between segmented in a scheme of their choosing. With your category in, they would be able to decide between single-segment, and a segmented in a scheme of their choosing. Except that adding another category would further dilute the 'sea of categories', probably making the level-by-level segmented run the odd one out nobody would watch. If the runners want more challenge, they can use less segments. They can limit themselves. In practice, nobody submits a new run that is just an improvement over another due to huge segmentation, without using a bunch of different tactics. And if they would, they would almost assuredly be ridiculed for it - the SDA community can appreciate a lack of over-segmentation, and can give credit for it, whether during verification or while watching the run.

I also agree with Cabbage, if the runner can't match up an older run, it's better that they don't submit anything instead of submitting an inferior one in a new category. We don't need to give lazy people more categories to make mediocre runs on. One thing I don't agree is that there's no reward for doing longer segments than what would be the best for the runner's sanity - there's the sense of accomplishment "ha! I didn't need to segment here, take that, game!". And ultimately, SDA is all about having fun and getting the sense of accomplishment :P.

And 'blackmail' (if what you said can even be called that) is a pathetic tactic to use in an argument. Not to mention, you wouldn't be pointing a problem, you'd actually be *creating* one. So much for the "for good of SDA" angle...

[quote="Spider-Waffle"]The one segment per level category would likely if anything condense the speedrunners efforts to that very category.  For, example, a game which there would be one run for, it would most likely be on the one segment per level category.  Nearly all save anywhere games, especially newer ones, runners aren't even willing or capable of making a SS run anyway.  Think about it, how many save anywhere games have SS runs, I actually would really like to know which ones do?  So the runners would basically be deciding between one segment level or segment anywhere; in the case of FPS, RTS, and puzzle (maybe other genres) save anywhere games, it seems to me that one segment per level is preferred, you would see a lot more runs for it if it existed.[/quote]
AvP2: Primal Hunt, Baldur's Gate: TotSC, Doom II (two runs), Morrowind, Fable (I think it had save anywhere), Fallout 1&2, Half-Life, Max Payne 1&2, Portal, Quake (4 categories), Rise of Nations had saving IIRC, Serious Sam: TFE, Silent Hill 2 (I think there was save anywhere on the PC, though I'm not sure since I always used the regular saving system, could someone confirm? The SS is on PC), Uplink. I might make an SS of Deus Ex, and my super-duper-secret project is actually aimed at obsoleting one segmented run with a single segment (though whether I'll manage this is another matter ;)). That doesn't even include games where save points are very easy to use, but the runners choose not to; or games where the PC version has save anywhere but there's no PC run so I omitted them (MGS springs to mind, it takes 1 second to save there, and you can save on most screens).

The runners currently can decide between a single-segment, or between segmented in a scheme of their choosing. With your category in, they would be able to decide between single-segment, and a segmented in a scheme of their choosing. Except that adding another category would further dilute the 'sea of categories', probably making the level-by-level segmented run the odd one out nobody would watch. If the runners want more challenge, they can use less segments. They can limit themselves. In practice, nobody submits a new run that is just an improvement over another due to huge segmentation, without using a bunch of different tactics. And if they would, they would almost assuredly be ridiculed for it - the SDA community can appreciate a lack of over-segmentation, and can give credit for it, whether during verification or while watching the run.

I also agree with Cabbage, if the runner can't match up an older run, it's better that they don't submit anything instead of submitting an inferior one in a new category. We don't need to give lazy people more categories to make mediocre runs on. One thing I don't agree is that there's no reward for doing longer segments than what would be the best for the runner's sanity - there's the sense of accomplishment "ha! I didn't need to segment here, take that, game!". And ultimately, SDA is all about having fun and getting the sense of accomplishment :P.

And 'blackmail' (if what you said can even be called that) is a pathetic tactic to use in an argument. Not to mention, you wouldn't be pointing a problem, you'd actually be *creating* one. So much for the "for good of SDA" angle...

arkarian

2009-07-12 11:51:02 am

gamelogs.org

Quote from ExplodingCabbage:

What could possibly be gained by SpiderWaffle beating those runs by using mass segmentation? It would be unfair and the end result would be much less impressive than what is already on SDA.

i'd just like to point out that this wouldn't be unfair in the least. anyone is welcome to do it, not just spider-waffle.

that said, this new category idea seems pretty excessive to me. if a runner wants to do one segment per level, no one's stopping them. if someone beats them by using more segments, clearly one segment per level in that game isn't optimal. if you want to see unsegmented levels that badly, just do an il run.

ShadowWraith

2009-07-12 12:17:51 pm

.

Quote from ExplodingCabbage:

To VorpalEdge and ShadowWraith - are you mad? What could possibly be gained by SpiderWaffle beating those runs by using mass segmentation? It would be unfair and the end result would be much less impressive than what is already on SDA.

SDA gains new runs, Spider-Waffle's supposed 'issue' becomes a reality, SDA either develops rules to either counter or accomodate them, or does nothing as it has been doing thus far. I have no problems with this.

ExplodingCabbage

Edit history:

ExplodingCabbage: 2009-07-12 01:11:08 pm

2009-07-12 12:56:13 pm

Quote from Arkarian:

that said, this new category idea seems pretty excessive to me. if a runner wants to do one segment per level, no one's stopping them. if someone beats them by using more segments, clearly one segment per level in that game isn't optimal. if you want to see unsegmented levels that badly, just do an il run.

Argh. Moronic argument. Go read the thread before you post again, these points were rebuked before they were even made.

1) Doing 1 segment per level is almost never optimal. In fact I challenge you to find me an example of game where sticking to 1 segment per level is optimal, I doubt you can find a single one. (I assume by 'optimal' you mean taking into account the save penalty, or else talking about the optimal number of segments doesn't even mean anything.) Runners choose to do 1-segment-per-level runs knowing that it's sub-optimal because they don't like mass segmenting, and that decision deserves to be respected by other runners (and in my view officially recognised) even if there is no different category. Your dismissive attitude of "Oh, it wasn't optimal so you obviously made a mistake by doing one segment per level and deserve to be obsoleted" is EXACTLY the reason people like SpiderWaffle and me have an issue here. Having played Unreal, I know I could obsolete the Unreal run in less than 48 hours with unlimited segmenting, and yet it is a brilliant, incredibly skilled run. The Warcraft III run is one of the best I've ever seen, and I don't have a fraction of the skill Stupid does, but with segmenting and a few weeks, I could beat that too. Yet the runs I would make would be hugely inferior to theirs if I did.

2) IL runs aren't possible for all games, and those for which it is possible wouldn't have this category. Again, the fact that you can't just do an IL run is why this issue has been raised in the first place.

I have no problem at all with informed disagreement based on a few sound points, heck, I disagree with SpiderWaffle myself on this, but people not bothering to read the thread and churning out illogical arguments (both in this and in the old rules thread) that totally miss the point pisses me off.

[quote="Arkarian"]that said, this new category idea seems pretty excessive to me. if a runner wants to do one segment per level, no one's stopping them. if someone beats them by using more segments, clearly one segment per level in that game isn't optimal. if you want to see unsegmented levels that badly, just do an il run.[/quote]

[i]Argh[/i]. Moronic argument. Go read the thread before you post again, these points were rebuked before they were even made.

1) Doing 1 segment per level is almost never optimal. In fact I challenge you to find me an example of game where sticking to 1 segment per level is optimal, I doubt you can find a single one. (I assume by 'optimal' you mean taking into account the save penalty, or else talking about the optimal number of segments doesn't even [i]mean[/i] anything.) Runners choose to do 1-segment-per-level runs knowing that it's sub-optimal because they don't like mass segmenting, and that decision deserves to be respected by other runners (and in my view officially recognised) even if there is no different category. Your dismissive attitude of "Oh, it wasn't optimal so you obviously made a mistake by doing one segment per level and deserve to be obsoleted" is EXACTLY the reason people like SpiderWaffle and me have an issue here. Having played Unreal, I know I could obsolete the Unreal run in less than 48 hours with unlimited segmenting, and yet it is a brilliant, incredibly skilled run. The Warcraft III run is one of the best I've ever seen, and I don't have a fraction of the skill Stupid does, but with segmenting and a few weeks, I could beat that too. Yet the runs I would make would be hugely inferior to theirs if I did.

2) IL runs aren't possible for all games, and those for which it is possible wouldn't have this category. Again, the fact that you [i]can't[/i] just do an IL run is [i]why[/i] this issue has been raised in the first place.

I have no problem at all with informed disagreement based on a few sound points, heck, I disagree with SpiderWaffle myself on this, but people not bothering to read the thread and churning out illogical arguments (both in this and in the old rules thread) that totally miss the point pisses me off.

Scepheo

2009-07-12 03:20:36 pm

I've been convinced that this category shouldn't exist either, but that's more due to unclarity than being an excessive category. Honestly, I think the best way to solve the issue (and my preferred way) would be adapt the rules a little bit, so more game can have IL tables. Obviously, RPG's can't have them, but FPS's cán. I mean, ammo carrying over can be an issue, but just let the runner play trough the game up to the level he wants to do. What he has there, he can use.

arkarian

2009-07-12 03:43:55 pm

gamelogs.org

just seems silly to me. there are few games that can be organized into levels that don't already allow il runs, and making up more categories to appease a few people playing a few games doesn't make sense.

cabbage: not sure what you're talking about, really. single-segment runs are one thing, but trying to categorize segmentation even further seems counterproductive. segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks. what's being proposed in this thread is basically saying "oh you're allowed to segment a little bit to save some time, but it's just not fair if you save too much time like those people who use many segments."

so the end result is a category with run times between those of ss and normal segmented runs. doesn't sound very interesting to watch. frankly, i'd rather get to see all the cool speed tricks (segmented run) rather than sacrificing some of them for no real reason. if i want to watch pure skill and luck, i'll watch a ss run.

just seems silly to me. there are few games that can be organized into levels that don't already allow il runs, and making up more categories to appease a few people playing a few games doesn't make sense.

cabbage: not sure what you're talking about, really. single-segment runs are one thing, but trying to categorize segmentation even further seems counterproductive. segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks. what's being proposed in this thread is basically saying "oh you're allowed to segment [i]a little bit[/i] to save [i]some[/i] time, but it's just not fair if you save [i]too[/i] much time like those people who use many segments."

so the end result is a category with run times between those of ss and normal segmented runs. doesn't sound very interesting to watch. frankly, i'd rather get to see [i]all[/i] the cool speed tricks (segmented run) rather than sacrificing some of them for no real reason. if i want to watch pure skill and luck, i'll watch a ss run.

Scepheo

2009-07-12 04:35:15 pm

Quote from Arkarian:

cabbage: not sure what you're talking about, really. single-segment runs are one thing, but trying to categorize segmentation even further seems counterproductive. segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks. what's being proposed in this thread is basically saying "oh you're allowed to segment a little bit to save some time, but it's just not fair if you save too much time like those people who use many segments."

Said kinda dumb; it's for people who can't go the stretch of doing a SS run, but who do want to display their skills.

ridd3r.

2009-07-12 04:43:27 pm

we have lift off

Quote from Scepheo:

Quote from Arkarian:

cabbage: not sure what you're talking about, really. single-segment runs are one thing, but trying to categorize segmentation even further seems counterproductive. segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks. what's being proposed in this thread is basically saying "oh you're allowed to segment a little bit to save some time, but it's just not fair if you save too much time like those people who use many segments."

Said kinda dumb; it's for people who can't go the stretch of doing a SS run, but who do want to display their skills.

Sounds a bit like tough luck to me, straight from SDA rules is the following: "A segmented run implies a higher level of risk-taking and a lower tolerance for mistakes. Use as many segments as is optimal to achieve the fastest final time. We will not be more impressed if you use a small number of segments. In particular, do not feel like you must use roughly the same number of segments as a run you are attempting to obsolete. "

If someone is amazing at a game and wants to display their skills then they should do a single-segment run. Segmented runs are all about speed as stated above, speed over skills as opposed to speed created by skills (a single-segment run). The problem of obsoleting a segmented run by using a ridiculous amount of segments has not arisen and the save penalty should at least help to take care of that. There are plenty of runs which have IL tables anyway so this problem of 'unfair' obsoletion that some people seem to be worried about, doesn't look like its going to happen anytime soon, if at all.

[quote="Scepheo"]
[quote="Arkarian"]
cabbage: not sure what you're talking about, really. single-segment runs are one thing, but trying to categorize segmentation even further seems counterproductive. segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks. what's being proposed in this thread is basically saying "oh you're allowed to segment [i]a little bit[/i] to save [i]some[/i] time, but it's just not fair if you save [i]too[/i] much time like those people who use many segments."
[/quote]
Said kinda dumb; it's for people who can't go the stretch of doing a SS run, but who [i]do[/i] want to display their skills.
[/quote]

Sounds a bit like tough luck to me, straight from SDA rules is the following: "A segmented run implies a higher level of risk-taking and a lower tolerance for mistakes. Use as many segments as is optimal to achieve the fastest final time. We will not be more impressed if you use a small number of segments. In particular, do not feel like you must use roughly the same number of segments as a run you are attempting to obsolete. "

If someone is amazing at a game and wants to display their skills then they should do a single-segment run. Segmented runs are all about speed as stated above, speed over skills as opposed to speed created by skills (a single-segment run). The problem of obsoleting a segmented run by using a ridiculous amount of segments has not arisen and the save penalty should at least help to take care of that. There are plenty of runs which have IL tables anyway so this problem of 'unfair' obsoletion that some people seem to be worried about, doesn't look like its going to happen anytime soon, if at all.

ExplodingCabbage

Edit history:

ExplodingCabbage: 2009-07-12 06:58:43 pm

2009-07-12 06:49:22 pm

Quote from Arkarian:

there are few games that can be organized into levels that don't already allow il runs

I would've thought the games for which one-segment-per-level is possible outnumber the games for which IL is possible. At any rate the numbers are comparable, they're certainly not a small minority.

Quote:

cabbage: not sure what you're talking about, really.

I don't know what you could fail to understand. I had two issues, and laid them out, I thought, perfectly clearly. But since you've decided to not respond to either of them and to reply to me as though I'm talking crap, I'll lay them out again:

1) You said that if someone beats a one-segment-per-level run with a mass segmented run, one segment per level is 'clearly not optimal', as an argument in favour of having mass segmented runs obsolete OSPL runs. However, firstly the fact that a OSPL run would be less optimal than a mass segmented run is obvious, and secondly it is irrelevant, since this entire argument is about whether to continue with the rule requiring runners to "use as many segments as is optimal to achieve the fastest final time", or to recognise runners' decisions to do OSPL runs and not treat that decision as a mistake. Saying 'we should accept runs that use more segments to optimise better because they use more segments to optimise better' is circular reasoning. Do you not agree? If not, why not? Don't simply say you don't know what I'm talking about this time.

2) You said that this category was unnecessary because runners could just do an IL run. However, this category is specifically for games without an IL category, so what you said simply isn't true. Again, do you accept this?

Quote:

segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks.

That is one reason for it.

Quote:

what's being proposed in this thread is basically saying "oh you're allowed to segment a little bit to save some time, but it's just not fair if you save too much time like those people who use many segments."

Nobody has said anything about fairness. It has simply been argued that it is more interesting to limit yourself to one segment per level. Given that well over a third of SDA's runs of save-anywhere games are OSPL runs, much of SDA seems to agree.

Quote:

so the end result is a category with run times between those of ss and normal segmented runs. doesn't sound very interesting to watch.

Fine, if you don't like the idea of people doing OSPL runs and don't want to give them any recognition or legitimacy, I don't have a problem with that, I just disagree. But you should've said that in the first place instead of squirming around with illogical and misleading arguments.

Quote:

frankly, i'd rather get to see all the cool speed tricks (segmented run) rather than sacrificing some of them for no real reason.

It would probably be relevant for me to point out that, as far as I know, no OSPL runs had to ditch any time-saving tricks as a result of the run being done OSPL instead of MS. Not saying it wouldn't happen, in fact it probably would if a formal category was created - but then that's not what I'm advocating.

Quote:

Sounds a bit like tough luck to me, straight from SDA rules is the following: "A segmented run implies a higher level of risk-taking and a lower tolerance for mistakes. Use as many segments as is optimal to achieve the fastest final time. We will not be more impressed if you use a small number of segments. In particular, do not feel like you must use roughly the same number of segments as a run you are attempting to obsolete. "

ARGH! This whole argument is about changing (or perhaps more accurately, tweaking) the rules. Saying that we shouldn't do so because the rules say so is a tad dumb.

Quote:

There are plenty of runs which have IL tables anyway so this problem of 'unfair' obsoletion that some people seem to be worried about, doesn't look like its going to happen anytime soon, if at all.

I have said it before, I will say it again: what possible relevance can IL tables have to an argument which is specifically about games which don't have them?

Edit: dex, I just realised I never replied to your proposal

Quote:

Edit: more thoughts, maybe rewrite the segments paragraph in the rules to something like:
Quote
Games that allow you to save your progress and continue later can be done using segments. You can retry segments as much as you want, in order to optimize them. Keep in mind that the purpose of segmentation is not to make life easier for you or to reduce the amount of time it takes you to produce a run. A segmented run implies a higher level of risk-taking and a lower tolerance for mistakes. Use as many segments as you feel are necessary. Bear in mind, however, that we will not be more impressed if you use a small amount of overly long, unoptimised segments. In particular, do not feel like you must use roughly the same number of segments as a run you are attempting to obsolete.
Subtle, but I feel it gets the point that you can segment however you want across.

I think this is definitely an improvement. However, I'm not sure about the use of the word 'necessary' - necessary for what? I really think that it would be preferable to put in a line specifically noting the fact that some runners choose to only use autosaves, or to only use one segment per level, and saying that this decision is generally respected and not counted against the run, especially if it does not require sacrificing any time-saving tricks. I don't think this would be out of place or arbitrary, given the significant number of runners already segmenting this way.

[quote="Arkarian"]there are few games that can be organized into levels that don't already allow il runs[/quote]

I would've thought the games for which one-segment-per-level is possible outnumber the games for which IL is possible. At any rate the numbers are comparable, they're certainly not a small minority.

[quote]cabbage: not sure what you're talking about, really.[/quote]

I don't know what you could fail to understand. I had two issues, and laid them out, I thought, perfectly clearly. But since you've decided to not respond to either of them and to reply to me as though I'm talking crap, I'll lay them out again:

1) You said that if someone beats a one-segment-per-level run with a mass segmented run, one segment per level is 'clearly not optimal', as an argument in favour of having mass segmented runs obsolete OSPL runs. However, firstly the fact that a OSPL run would be less optimal than a mass segmented run is obvious, and secondly it is irrelevant, since this entire argument is about whether to continue with the rule requiring runners to "use as many segments as is optimal to achieve the fastest final time", or to recognise runners' decisions to do OSPL runs and not treat that decision as a mistake. Saying 'we should accept runs that use more segments to optimise better because they use more segments to optimise better' is circular reasoning. Do you not agree? If not, why not? Don't simply say you don't know what I'm talking about this time.

2) You said that this category was unnecessary because runners could just do an IL run. However, this category is specifically for games without an IL category, so what you said simply isn't true. Again, do you accept this?

[quote]segmentation is used for a reason— to enable runners to pull off more difficult time-saving tricks.[/quote]

That is one reason for it.

[quote]what's being proposed in this thread is basically saying "oh you're allowed to segment [i]a little bit[/i] to save [i]some[/i] time, but it's just not fair if you save [i]too[/i] much time like those people who use many segments."[/quote]

Nobody has said anything about fairness. It has simply been argued that it is more interesting to limit yourself to one segment per level. Given that well over a third of SDA's runs of save-anywhere games are OSPL runs, much of SDA seems to agree.

[quote]so the end result is a category with run times between those of ss and normal segmented runs. doesn't sound very interesting to watch.[/quote]

Fine, if you don't like the idea of people doing OSPL runs and don't want to give them any recognition or legitimacy, I don't have a problem with that, I just disagree. But you should've said that in the first place instead of squirming around with illogical and misleading arguments.

[quote]frankly, i'd rather get to see [i]all[/i] the cool speed tricks (segmented run) rather than sacrificing some of them for no real reason.[/quote]

It would probably be relevant for me to point out that, as far as I know, no OSPL runs had to ditch any time-saving tricks as a result of the run being done OSPL instead of MS. Not saying it wouldn't happen, in fact it probably would if a formal category was created - but then that's not what I'm advocating.

[quote]Sounds a bit like tough luck to me, straight from SDA rules is the following: "A segmented run implies a higher level of risk-taking and a lower tolerance for mistakes. Use as many segments as is optimal to achieve the fastest final time. We will not be more impressed if you use a small number of segments. In particular, do not feel like you must use roughly the same number of segments as a run you are attempting to obsolete. "[/quote]

[i]ARGH![/i] This whole argument is about changing (or perhaps more accurately, tweaking) the rules. Saying that we shouldn't do so because the rules say so is a tad dumb.

[quote]There are plenty of runs which have IL tables anyway so this problem of 'unfair' obsoletion that some people seem to be worried about, doesn't look like its going to happen anytime soon, if at all.[/quote]

I have said it before, I will say it again: what possible relevance can IL tables have to an argument which is specifically about games which [i]don't have them[/i]?

Edit: dex, I just realised I never replied to your proposal

[quote]Edit: more thoughts, maybe rewrite the segments paragraph in the rules to something like:
Quote
Games that allow you to save your progress and continue later can be done using segments. You can retry segments as much as you want, in order to optimize them. Keep in mind that the purpose of segmentation is not to make life easier for you or to reduce the amount of time it takes you to produce a run. A segmented run implies a higher level of risk-taking and a lower tolerance for mistakes. Use as many segments as you feel are necessary. Bear in mind, however, that we will not be more impressed if you use a small amount of overly long, unoptimised segments. In particular, do not feel like you must use roughly the same number of segments as a run you are attempting to obsolete.
Subtle, but I feel it gets the point that you can segment however you want across.[/quote]

I think this is definitely an improvement. However, I'm not sure about the use of the word 'necessary' - necessary for what? I really think that it would be preferable to put in a line specifically noting the fact that some runners choose to only use autosaves, or to only use one segment per level, and saying that this decision is generally respected and not counted against the run, especially if it does not require sacrificing any time-saving tricks. I don't think this would be out of place or arbitrary, given the significant number of runners already segmenting this way.

arkarian

2009-07-12 09:07:56 pm

gamelogs.org

i like the idea of one segment per two levels. that just makes a whole lot of sense to me, and i'm sure many people would want to do such a run. it should get its own category.

oh, and three levels per segment should get a category too, since many games have multiple levels in a row that are related somehow.

do you see where i'm going with this? basically more categories causes more headaches. what ridd3r said is spot-on: either do a ss run and show us your skill, or do a segmented run and show us the fastest possible way to beat the game with the coolest tricks. not something in between.

InsipidMuckyWater

2009-07-12 09:35:58 pm

Visit my profile to see my runs!

^ Believe me, I'm not taking sides, but just for the record I think the problem with this criticism is that you're treating segment categories interchangeably, but one of the points behind OSPL is that the segment length [one level] is not arbitrary. In fact, they mean to argue that its appropriate (and honestly I don't think it's difficult to see why).

That being said, I personally am not convinced the addition of the OSPL category would actually benefit the site, so I guess I'd vote no if push came to shove, but again I am not particularly concerned.

Spider-Waffle

(user is banned)

Spider-Waffle

Edit history:

Spider-Waffle: 2009-07-12 11:13:13 pm

2009-07-12 10:19:26 pm

Don't think! feeeeeal

Quote:

so the end result is a category with run times between those of ss and normal segmented runs. doesn't sound very interesting to watch.

So you I guess you have issues with the way the quake, quake II, Unreal, the Warcraft III, the Splinter Cell: Chaos Theory run, the Jedi Knight segmented runs were done, or any of the IL runs for games which can save anywhere like doom, doom II, starcraft, starcraft: broodwar. Not to mention all the runs very close to OSPL that surely would have been made that way if the runners had the category to submit to. These runs don't interest you? You don't think the quake speedrunners were onto anything when they first assumed OSPL was the best way to run quake?

I'm guessing the quake speedrunning section of SDA already has OSPL along with it's ILs as OSPL? If it doesn't I'll beat some ILs tonight. I've already beaten a Starcraft IL OSPL with 3 segments getting under half the current record, I'll just need to record this, shouldn't take more than a half hour.

Quote:

i like the idea of one segment per two levels. that just makes a whole lot of sense to me, and i'm sure many people would want to do such a run.

Besides from being arbitrary, can you back up your statement that many will want to do such run with proof such as I can for OSPL by the fact that many if not most runners already prefer and have segmented their runs this way?

Quote:

AvP2: Primal Hunt, Baldur's Gate: TotSC, Doom II (two runs), Morrowind, Fable (I think it had save anywhere), Fallout 1&2, Half-Life, Max Payne 1&2, Portal, Quake (4 categories), Rise of Nations had saving IIRC, Serious Sam: TFE, Silent Hill 2 (I think there was save anywhere on the PC, though I'm not sure since I always used the regular saving system, could someone confirm? The SS is on PC), Uplink.

Hmm, that's a lot more than I thought, but most these are very short games and old games and/or done on an easier difficulty? Which of these have runs on the highest difficulty? Also noticed some of the runs didn't say their difficulty but other runs for that game did. I think it'd be nice if SDA also specified what the difficulty was relative the available difficulty options for that game. Like max payne, DOA, I have no idea where DOA ranks, is it 3/3, 2/3, 2/9, 10/11, I have no idea. This is off topic but I'm just curious about this.

[quote]so the end result is a category with run times between those of ss and normal segmented runs. doesn't sound very interesting to watch.[/quote]

So you I guess you have issues with the way the quake, quake II, Unreal, the Warcraft III, the Splinter Cell: Chaos Theory run, the Jedi Knight segmented runs were done, or any of the IL runs for games which can save anywhere like doom, doom II, starcraft, starcraft: broodwar.  Not to mention all the runs very close to OSPL that surely would have been made that way if the runners had the category to submit to.  These runs don't interest you?  You don't think the quake speedrunners were onto anything when they first assumed OSPL was the best way to run quake?

I'm guessing the quake speedrunning section of SDA already has OSPL along with it's ILs as OSPL?  If it doesn't I'll beat some ILs tonight.  I've already beaten a Starcraft IL OSPL with 3 segments getting under half the current record, I'll just need to record this, shouldn't take more than a half hour.

[quote]i like the idea of one segment per two levels. that just makes a whole lot of sense to me, and i'm sure many people would want to do such a run.[/quote]

Besides from being arbitrary, can you back up your statement that many will want to do such run with proof such as I can for OSPL by the fact that many if not most runners already prefer and have segmented their runs this way?

[quote]AvP2: Primal Hunt, Baldur's Gate: TotSC, Doom II (two runs), Morrowind, Fable (I think it had save anywhere), Fallout 1&2, Half-Life, Max Payne 1&2, Portal, Quake (4 categories), Rise of Nations had saving IIRC, Serious Sam: TFE, Silent Hill 2 (I think there was save anywhere on the PC, though I'm not sure since I always used the regular saving system, could someone confirm? The SS is on PC), Uplink.[/quote]

Hmm, that's a lot more than I thought, but most these are very short games and old games and/or done on an easier difficulty?  Which of these have runs on the highest difficulty?  Also noticed some of the runs didn't say their difficulty but other runs for that game did.  I think it'd be nice if SDA also specified what the difficulty was relative the available difficulty options for that game.  Like max payne, DOA, I have no idea where DOA ranks, is it 3/3, 2/3, 2/9, 10/11, I have no idea.  This is off topic but I'm just curious about this.

Spider-Waffle

(user is banned)

Spider-Waffle

Edit history:

Spider-Waffle: 2009-07-12 11:45:09 pm

2009-07-12 10:54:48 pm

Don't think! feeeeeal

Quote:

I think the type of runner that makes a lesser run on another category instead of beating an existing run on a more prominent category most likely didn't think they could beat the existing run anyway because they lacked some combination of skill, discipline, patience, time. So they likely wouldn't have ever contributed a run at all if the different category never existed.

And that would be better than them submitting a run inferior to one that already exists, in my opinion.

If this is the case I think the problem is in SDA's admissions or how they advertise the runs available to viewers, not the fact that the categories exist.

SDA could require runs on new categories be of similar quality to existing runs for that game. Also, SDA could give an editor's ranking of how the runs for a game rank within that game. Both of which I think are better solutions than reducing the number of available categories, especially when this means denying runners and viewers the category they most prefer.

If mediocre runs on separate categories are hurting SDA, then SDA shouldn't accept them, it's really that simple. The problem is that they're accepting the runs, not that the runs have a place to exist.

[quote][quote]I think the type of runner that makes a lesser run on another category instead of beating an existing run on a more prominent category most likely didn't think they could beat the existing run anyway because they lacked some combination of skill, discipline, patience, time.  So they likely wouldn't have ever contributed a run at all if the different category never existed.[/quote]

And that would be better than them submitting a run inferior to one that already exists, in my opinion.[/quote]

If this is the case I think the problem is in SDA's admissions or how they advertise the runs available to viewers, not the fact that the categories exist.

SDA could require runs on new categories be of similar quality to existing runs for that game.  Also, SDA could give an editor's ranking of how the runs for a game rank within that game.  Both of which I think are better solutions than reducing the number of available categories, especially when this means denying runners and viewers the category they most prefer.

If mediocre runs on separate categories are hurting SDA, then SDA shouldn't accept them, it's really that simple.  The problem is that they're accepting the runs, not that the runs have a place to exist.

Spider-Waffle

(user is banned)

Spider-Waffle

Edit history:

Spider-Waffle: 2009-07-12 11:51:56 pm

2009-07-12 11:41:36 pm

Don't think! feeeeeal

Quote:

Can you point out any example where more categories lead to the best run for that game being worse?

Quote:

That's a pretty unfair challenge. How could you ever directly show cause and effect here?

Well I thought this was the crux of why your against a OSPL cat? So you're basically just assuming this is true based off your intuition? I don't think it's fair to have the conclusion to your argument solely resting on your intuition about something. Just because your intuition about something tells it's true doesn't mean that it actually is. I would perter it if you went to more lengths to try and prove this before you reach your conclusion seeing as this seems to be the only counter-argument.

Here's an example and please don't debate this off topic: your intuition would tell you that bicycle riders wearing a bike helmet are less likely to sustain a head injury than bicycle riders not wearing a helmet. But the stats show that this isn't the case. This is largely because riders wearing a helmet are more reckless than riders who don't and bike helmets don't offer nearly enough protection to your head against car collisions, they're only rated for very low velocity impacts. So if the only reason (in this case it probably isn't the only reason) you want to enforce a bike helmet law is because you think it will protect bike riders, your assumption is wrong and if your whole argument is resting on this assumption your argument is quite weak.

[quote]
Can you point out any example where more categories lead to the best run for that game being worse?

[quote]That's a pretty unfair challenge. How could you ever directly show cause and effect here?[/quote]
[/quote]

Well I thought this was the crux of why your against a OSPL cat?  So you're basically just assuming this is true based off your intuition?  I don't think it's fair to have the conclusion to your argument solely resting on your intuition about something.  Just because your intuition about something tells it's true doesn't mean that it actually is.  I would perter it if you went to more lengths to try and prove this before you reach your conclusion seeing as this seems to be the only counter-argument.

Here's an example and please don't debate this off topic: your intuition would tell you that bicycle riders wearing a bike helmet are less likely to sustain a head injury than bicycle riders not wearing a helmet.  But the stats show that this isn't the case.  This is largely because riders wearing a helmet are more reckless than riders who don't and bike helmets don't offer nearly enough protection to your head against car collisions, they're only rated for very low velocity impacts.  So if the only reason (in this case it probably isn't the only reason) you want to enforce a bike helmet law is because you think it will protect bike riders, your assumption is wrong and if your whole argument is resting on this assumption your argument is quite weak.

Lag.Com

2009-07-13 03:58:41 am

sda loyalist

Spider, stop using analogies, please, they aren't helping your case. Roll Eyes

The reason Quake runners picked 'OSPL' to create their project runs (afaik) was because it made sense. Personally I'd rather see OSPL replace 'segmented' but that isn't going to happen because people think lower numbers are more important than impressive videos.

ExplodingCabbage

2009-07-13 05:35:16 am

Quote from Spider-Waffle:

I'm guessing the quake speedrunning section of SDA already has OSPL along with it's ILs as OSPL? If it doesn't I'll beat some ILs tonight. I've already beaten a Starcraft IL OSPL with 3 segments getting under half the current record, I'll just need to record this, shouldn't take more than a half hour.

Segmenting isn't allowed in IL runs. Which, of course, further supports the well-expressed point by InsipidMuckyWater that one segment per level is no arbitrary - SDA has already taken exactly this stance on IL runs.

Quote from Spider-Waffle:

Quote:

Can you point out any example where more categories lead to the best run for that game being worse?

Quote:

That's a pretty unfair challenge. How could you ever directly show cause and effect here?

Well I thought this was the crux of why your against a OSPL cat?

Well, no, although it's one of the main reasons. But bigger than this is the fact that having multiple similar categories confuses and probably ultimately turns away viewers. For example, have a look at the Metroid Prime page:
http://speeddemosarchive.com/MetroidPrime.html
There are 15 runs. Now, I know Metroid speedrunning goes back to the dawn of civilisation and I don't doubt these runs are all excellent - quality isn't the issue here. However, how am I, as a naive viewer who knows nothing about Metroid Prime speedrunning, supposed to choose one to watch? The times are all similar, and so are many of the categories, not to mention that every PAL run has its NTSC equivalent. The end result was that I simply didn't watch any of them.

The same problem will apply if you have OSPL and MS categories. If you get a solid submission to both categories, viewers won't know which to choose and at any rate they are unlikely to watch both if the times are similar (which they will be - for the vast majority of games, there will be no major route differences between a OSPL run and a MS run), which means in the end each run gets half as many viewers as it otherwise would (or less, if some turn away when they can't decide which to watch). As a viewer, I'd rather not have to worry about whether the run I was downloading was inferior to the other option.

As for the question of the impact of more categories on play quality...

Quote:

So you're basically just assuming this is true based off your intuition? I don't think it's fair to have the conclusion to your argument solely resting on your intuition about something. Just because your intuition about something tells it's true doesn't mean that it actually is. I would perter it if you went to more lengths to try and prove this before you reach your conclusion seeing as this seems to be the only counter-argument.

Okay, lots of points to make here:
1) Intuition may not be reliable, but in the absence of evidence it makes sense to go with intuition.
2) If we forget about the quality of the 'best' run for a given game, and instead focus on the average quality of all runs on the site, a decrease seems much more likely. If A does an OSPL run, and then B comes along and does a better MS run that isn't *enough* better to obsolete A's run, then A's run is kicking around dragging standards down (and getting viewers that ought's to be B's). But if we have no OSPL category, then A does an MS run, B comes along and does a better MS run and A's run gets obsoleted.
3) It's not the only counter argument. See my first post in this thread. I'd say the lack of necessity (especially if we can formally acknowledge the fact that runners often do OSPL or autosave-only runs in the rules) and the nuisance to viewers/inelegance of having multiple similar runs are the biggest arguments against, with the probable decrease in run quality just being another relevant point.

[quote="Spider-Waffle"]I'm guessing the quake speedrunning section of SDA already has OSPL along with it's ILs as OSPL?  If it doesn't I'll beat some ILs tonight.  I've already beaten a Starcraft IL OSPL with 3 segments getting under half the current record, I'll just need to record this, shouldn't take more than a half hour.[/quote]

Segmenting isn't allowed in IL runs. Which, of course, further supports the well-expressed point by InsipidMuckyWater that one segment per level is no arbitrary - SDA has already taken exactly this stance on IL runs.

[quote="Spider-Waffle"]
[quote]
Can you point out any example where more categories lead to the best run for that game being worse?

[quote]That's a pretty unfair challenge. How could you ever directly show cause and effect here?[/quote]
[/quote]

Well I thought this was the crux of why your against a OSPL cat? [/quote]

Well, no, although it's one of the main reasons. But bigger than this is the fact that having multiple similar categories confuses and probably ultimately turns away viewers. For example, have a look at the Metroid Prime page:
http://speeddemosarchive.com/MetroidPrime.html
There are 15 runs. Now, I know Metroid speedrunning goes back to the dawn of civilisation and I don't doubt these runs are all excellent - quality isn't the issue here. However, how am I, as a naive viewer who knows nothing about Metroid Prime speedrunning, supposed to choose one to watch? The times are all similar, and so are many of the categories, not to mention that every PAL run has its NTSC equivalent. The end result was that I simply didn't watch any of them.

The same problem will apply if you have OSPL and MS categories. If you get a solid submission to both categories, viewers won't know which to choose and at any rate they are unlikely to watch both if the times are similar (which they will be - for the vast majority of games, there will be no major route differences between a OSPL run and a MS run), which means in the end each run gets half as many viewers as it otherwise would (or less, if some turn away when they can't decide which to watch). As a viewer, I'd rather not have to worry about whether the run I was downloading was inferior to the other option.

As for the question of the impact of more categories on play quality...

[quote]So you're basically just assuming this is true based off your intuition?  I don't think it's fair to have the conclusion to your argument solely resting on your intuition about something.  Just because your intuition about something tells it's true doesn't mean that it actually is.  I would perter it if you went to more lengths to try and prove this before you reach your conclusion seeing as this seems to be the only counter-argument.[/quote]

Okay, lots of points to make here:
1) Intuition may not be reliable, but in the absence of evidence it makes sense to go with intuition.
2) If we forget about the quality of the 'best' run for a given game, and instead focus on the average quality of all runs on the site, a decrease seems much more likely. If A does an OSPL run, and then B comes along and does a better MS run that isn't *enough* better to obsolete A's run, then A's run is kicking around dragging standards down (and getting viewers that ought's to be B's). But if we have no OSPL category, then A does an MS run, B comes along and does a better MS run and A's run gets obsoleted.
3) It's not the only counter argument. See my first post in this thread. I'd say the lack of necessity (especially if we can formally acknowledge the fact that runners often do OSPL or autosave-only runs in the rules) and the nuisance to viewers/inelegance of having multiple similar runs are the biggest arguments against, with the probable decrease in run quality just being another relevant point.

ridd3r.

2009-07-13 05:43:18 am

we have lift off

Quote:

Here's an example and please don't debate this off topic

So you give an example which is meant to further your argument and we aren't allowed to argue the basis of this example when it is clearly ridiculous?

Quote:

but most these are very short games and old games and/or done on an easier difficulty?

You just go against the very point you try and make less than 2 hours later. Now you are the one assuming that because these games are done in a SS on games where you can save anywhere they are obviously on easy difficulty and short games? Now whos using their intuition? Max Payne 1 (which I did in a SS) was done on the hardest difficulty (DoA) and was 1.5 hours long, not exactly short for a SS. I don't see why it is so hard to tell what difficulty they are done on, most of the people interested in the run are going to be people who have played it. For those who haven't theres a slight hint from watching the run where on the title screen I clearly go to the bottom difficulty setting and it says the ultimate challenge.

Quote:

ARGH! This whole argument is about changing (or perhaps more accurately, tweaking) the rules. Saying that we shouldn't do so because the rules say so is a tad dumb.

I find arguing about a problem which hasn't and probably won't occur dumb to be honest. It seems pretty obvious by now that the general consensus is the category is not going to be added, as for tweaking the rules, if people did OSPL runs with that rule set in place and are continuing to do so whats the problem? By stating the rules I was trying to point out that the goal of SDA for segmented runs is to get the fastest time. Your "tweak" if it can be called that is trying to go against this completely in a given situation.

I would have thought that new runners would tend to follow in the old runs footsteps as they will not be familiar with the SDA process and want to guarantee their run gets accepted. If unsure they would ask someone here who would point them in the right direction, either way new or old members get to know the score pretty quickly. I do however think that the rules should pretty much remain as they are and cases like this are really unwritten rules, tweaking them to include what you want exactly would send a mixed message. Yes you should segment as much as you want and get the fastest time but only if a run of that game has not been done before? Also they may get confused and think they are limited to this rule set when doing a new run. Some things are just best left unsaid, or maybe very subtly like dex's proposal.

Quote:

Nobody has said anything about fairness. It has simply been argued that it is more interesting to limit yourself to one segment per level. Given that well over a third of SDA's runs of save-anywhere games are OSPL runs, much of SDA seems to agree.

You yourself are against this category right? Here is seems to me that you are all for it, no mere tweaking of the rules would lead to an application of OSPL runs across the board. Playing the devil's advocate is OK to a point but you seem to be missing peoples points a lot.
Also fairness has cropped up, you yourself have said that it would be unfair for you to obsolete a run with mass segmenting! What the other guy was saying was simply the flipside of this.

[quote] Here's an example and please don't debate this off topic [/quote]

So you give an example which is meant to further your argument and we aren't allowed to argue the basis of this example when it is clearly ridiculous?

[quote] but most these are very short games and old games and/or done on an easier difficulty? [/quote]

You just go against the very point you try and make less than 2 hours later. Now you are the one assuming that because these games are done in a SS on games where you can save anywhere they are obviously on easy difficulty and short games? Now whos using their intuition? Max Payne 1 (which I did in a SS) was done on the hardest difficulty (DoA) and was 1.5 hours long, not exactly short for a SS. I don't see why it is so hard to tell what difficulty they are done on, most of the people interested in the run are going to be people who have played it. For those who haven't theres a slight hint from watching the run where on the title screen I clearly go to the bottom difficulty setting and it says the ultimate challenge.

[quote] ARGH! This whole argument is about changing (or perhaps more accurately, tweaking) the rules. Saying that we shouldn't do so because the rules say so is a tad dumb. [/quote]

I find arguing about a problem which hasn't and probably won't occur dumb to be honest. It seems pretty obvious by now that the general consensus is the category is not going to be added, as for tweaking the rules, if people did OSPL runs with that rule set in place and are continuing to do so whats the problem? By stating the rules I was trying to point out that the goal of SDA for segmented runs is to get the fastest time. Your "tweak" if it can be called that is trying to go against this completely in a given situation.

I would have thought that new runners would tend to follow in the old runs footsteps as they will not be familiar with the SDA process and want to guarantee their run gets accepted. If unsure they would ask someone here who would point them in the right direction, either way new or old members get to know the score pretty quickly. I do however think that the rules should pretty much remain as they are and cases like this are really unwritten rules, tweaking them to include what you want exactly would send a mixed message. Yes you should segment as much as you want and get the fastest time but only if a run of that game has not been done before? Also they may get confused and think they are limited to this rule set when doing a new run. Some things are just best left unsaid, or maybe very subtly like dex's proposal.

[quote] Nobody has said anything about fairness. It has simply been argued that it is more interesting to limit yourself to one segment per level. Given that well over a third of SDA's runs of save-anywhere games are OSPL runs, much of SDA seems to agree. [/quote] 
You yourself are against this category right? Here is seems to me that you are all for it, no mere tweaking of the rules would lead to an application of OSPL runs across the board. Playing the devil's advocate is OK to a point but you seem to be missing peoples points a lot.
Also fairness has cropped up, you yourself have said that it would be unfair for you to obsolete a run with mass segmenting! What the other guy was saying was simply the flipside of this.

ExplodingCabbage

2009-07-13 02:57:06 pm

Quote from ridd3r.:

Quote:

ARGH! This whole argument is about changing (or perhaps more accurately, tweaking) the rules. Saying that we shouldn't do so because the rules say so is a tad dumb.

I find arguing about a problem which hasn't and probably won't occur dumb to be honest.

Quote:

if people did OSPL runs with that rule set in place and are continuing to do so whats the problem?

The problem is that those runs are officially worse than they would be if they used mass segmentation, which isn't fair, and that people who have just read the rules and seen that you're supposed to use the optimal number of segments to minimise your time won't understand the runners' segmentation decisions and will dismiss good runs as a result.

To give but two examples, the first time I watched the Unreal run, or dex's Deus Ex run that is presently on the site, I thought they were pretty poor because, as a newcomer having just read the rules and seen SpiderWaffle's Half Life run, I just saw 'segmented' and assumed therefore that the runs used as many segments as optimal to minimise the time subject to the save penalty, as the rules currently suggest runners should do. Yet, clearly both those runs had plenty of little mistakes, some individually over half a second, which could've easily been removed with an extra segment. I assumed the reason for these mistakes was simple poor standards and laziness on the part of the runners. It wasn't until I'd been around a few months that I realised the runners had made a decision not to follow the philosophy laid out in the rules and to apply their own segmentation systems (OSPL for the Unreal run, and I have no idea how dex makes his segmenting decisions). In fact, I have never seen a save-anywhere run not by SpiderWaffle where the runner has actually used the optimal number of segments to minimise their time.

I would say that if the vast, vast majority of runners are following different philosophies on segmentation to that which is given in the rules (do you not agree that this is so?), then we should rewrite those rules. I also think that people new to the site not understanding runners' segmentation decisions and consequently dismissing some of the best runs is a real problem, not a hypothetical one.

Quote:

By stating the rules I was trying to point out that the goal of SDA for segmented runs is to get the fastest time. Your "tweak" if it can be called that is trying to go against this completely in a given situation.

As has already been pointed out, doing one segment per level is comparable to IL runs and so doesn't really run counter to SDA's philosophy. And besides, as I argued above, nobody actually just aims to get the fastest time in their segmented runs except SpiderWaffle anyway, whatever the rules may say.

Quote:

I would have thought that new runners would tend to follow in the old runs footsteps as they will not be familiar with the SDA process and want to guarantee their run gets accepted. If unsure they would ask someone here who would point them in the right direction, either way new or old members get to know the score pretty quickly. I do however think that the rules should pretty much remain as they are and cases like this are really unwritten rules, tweaking them to include what you want exactly would send a mixed message. Yes you should segment as much as you want and get the fastest time but only if a run of that game has not been done before? Also they may get confused and think they are limited to this rule set when doing a new run. Some things are just best left unsaid, or maybe very subtly like dex's proposal.

I do take your point here. Any suggestions on the precise wording to use would be welcome.

Quote:

You yourself are against this category right?

Yes.

Quote:

you seem to be missing peoples points a lot.

Which? If I've genuinely missed the point somewhere I'll be happy to go back and reread and either concede the point or try to explain my dismissal of it better.

Quote:

Also fairness has cropped up, you yourself have said that it would be unfair for you to obsolete a run with mass segmenting! What the other guy was saying was simply the flipside of this.

No it isn't. In Arkarian's twisting of the argument, I and SpiderWaffle were asserting that MS runs are somehow inherently unfair, which we've never said and plainly isn't true. The unfairness comes from having OSPL and MS runs competing in the same category, and he (deliberately?) missed this point and presented a parody of our argument rather than dealing with the argument itself.

[quote="ridd3r."][quote] ARGH! This whole argument is about changing (or perhaps more accurately, tweaking) the rules. Saying that we shouldn't do so because the rules say so is a tad dumb. [/quote]

I find arguing about a problem which hasn't and probably won't occur dumb to be honest.[/quote]
[quote]if people did OSPL runs with that rule set in place and are continuing to do so whats the problem?[/quote]

The problem is that those runs are officially worse than they would be if they used mass segmentation, which isn't fair, and that people who have just read the rules and seen that you're supposed to use the optimal number of segments to minimise your time won't understand the runners' segmentation decisions and will dismiss good runs as a result.

To give but two examples, the first time I watched the Unreal run, or dex's Deus Ex run that is presently on the site, I thought they were pretty poor because, as a newcomer having just read the rules and seen SpiderWaffle's Half Life run, I just saw 'segmented' and assumed therefore that the runs used as many segments as optimal to minimise the time subject to the save penalty, as the rules currently suggest runners should do. Yet, clearly both those runs had plenty of little mistakes, some individually over half a second, which could've easily been removed with an extra segment. I assumed the reason for these mistakes was simple poor standards and laziness on the part of the runners. It wasn't until I'd been around a few months that I realised the runners had made a decision not to follow the philosophy laid out in the rules and to apply their own segmentation systems (OSPL for the Unreal run, and I have no idea how dex makes his segmenting decisions). In fact, I have never seen a save-anywhere run not by SpiderWaffle where the runner has actually used the optimal number of segments to minimise their time.

I would say that if the vast, vast majority of runners are following different philosophies on segmentation to that which is given in the rules (do you not agree that this is so?), then we should rewrite those rules. I also think that people new to the site not understanding runners' segmentation decisions and consequently dismissing some of the best runs is a real problem, not a hypothetical one.

[quote]By stating the rules I was trying to point out that the goal of SDA for segmented runs is to get the fastest time. Your "tweak" if it can be called that is trying to go against this completely in a given situation.[/quote]

As has already been pointed out, doing one segment per level is comparable to IL runs and so doesn't really run counter to SDA's philosophy. And besides, as I argued above, nobody [i]actually[/i] just aims to get the fastest time in their segmented runs except SpiderWaffle anyway, whatever the rules may say.

[quote]I would have thought that new runners would tend to follow in the old runs footsteps as they will not be familiar with the SDA process and want to guarantee their run gets accepted. If unsure they would ask someone here who would point them in the right direction, either way new or old members get to know the score pretty quickly. I do however think that the rules should pretty much remain as they are and cases like this are really unwritten rules, tweaking them to include what you want exactly would send a mixed message. Yes you should segment as much as you want and get the fastest time but only if a run of that game has not been done before? Also they may get confused and think they are limited to this rule set when doing a new run. Some things are just best left unsaid, or maybe very subtly like dex's proposal.[/quote]

I do take your point here. Any suggestions on the precise wording to use would be welcome.

[quote]You yourself are against this category right?[/quote]

Yes.

[quote]you seem to be missing peoples points a lot.[/quote]

Which? If I've genuinely missed the point somewhere I'll be happy to go back and reread and either concede the point or try to explain my dismissal of it better.

[quote]Also fairness has cropped up, you yourself have said that it would be unfair for you to obsolete a run with mass segmenting! What the other guy was saying was simply the flipside of this.[/quote]

No it isn't. In Arkarian's twisting of the argument, I and SpiderWaffle were asserting that MS runs are somehow inherently unfair, which we've never said and plainly isn't true. The unfairness comes from having OSPL and MS runs competing in the same category, and he (deliberately?) missed this point and presented a parody of our argument rather than dealing with the argument itself.

dex

2009-07-13 03:06:04 pm

Invisible avatar

Quote from ExplodingCabbage:

I would say that if the vast, vast majority of runners are following different philosophies on segmentation to that which is given in the rules (do you not agree that this is so?), then we should rewrite those rules.

Conversely, if the vast, vast majority of runners are already following the different philosophy, the way the rules are phrased clearly doesn't stop them from doing so. I think some things are best left in the subtext - if you laid out the segmentation in the rules more, people could get the idea that any segmentation is seen as a bad thing.

In Deus Ex, I segment only after map load, and I try to avoid segmenting whenever possible. By which I mean I've spent a few days running Versalife in a very long segment instead of keeping the segments separate, just because I knew I could do it without the additional save.

Also, the 'one segment per map/level' is a misnomer, as LL pointed out. That would let the runner run 3 levels consecutive without saving and then save thrice in a difficult segment, still keeping the average at one per map. Just a bit of pedantry ;).

Also, groobo used a lot of segments in the Diablo run.

[quote="ExplodingCabbage"]I would say that if the vast, vast majority of runners are following different philosophies on segmentation to that which is given in the rules (do you not agree that this is so?), then we should rewrite those rules.[/quote]
Conversely, if the vast, vast majority of runners are already following the different philosophy, the way the rules are phrased clearly doesn't stop them from doing so. I think some things are best left in the subtext - if you laid out the segmentation in the rules more, people could get the idea that any segmentation is seen as a bad thing.

In Deus Ex, I segment only after map load, and I try to avoid segmenting whenever possible. By which I mean I've spent a few days running Versalife in a very long segment instead of keeping the segments separate, just because I knew I could do it without the additional save.

Also, the 'one segment per map/level' is a misnomer, as LL pointed out. That would let the runner run 3 levels consecutive without saving and then save thrice in a difficult segment, still keeping the average at one per map. Just a bit of pedantry ;).

Also, groobo used a lot of segments in the Diablo run.

Spider-Waffle

(user is banned)

Spider-Waffle

Edit history:

Spider-Waffle: 2009-07-13 05:47:01 pm

2009-07-13 03:38:35 pm

Don't think! feeeeeal

Quote:

but most these are very short games and old games and/or done on an easier difficulty?

You just go against the very point you try and make less than 2 hours later. Now you are the one assuming that because these games are done in a SS on games where you can save anywhere they are obviously on easy difficulty and short games? Now whos using their intuition? Max Payne 1 (which I did in a SS) was done on the hardest difficulty (DoA) and was 1.5 hours long, not exactly short for a SS. I don't see why it is so hard to tell what difficulty they are done on, most of the people interested in the run are going to be people who have played it. For those who haven't theres a slight hint from watching the run where on the title screen I clearly go to the bottom difficulty setting and it says the ultimate challenge.

Key word, most... It's kind of funny I use the word most then you point out one example as if this means my statement wasn't true. I actually did look at game page for everyone of those runs. The max payne 1 run was one of the only ones on highest difficulty and wasn't a short game, of course I just found out DOA is the highest difficulty. Also as a viewer I'd like to know what DOA difficulty meant without having to DL and watch the run first. Also higher difficulties can be unlocked in many games so I still wouldn't know it was the highest by watching the run. The Silent hill 2 run doesn't say the difficulty but there's another run on that game page which it does give a difficulty for. The max payne 2 run doesn't say it's single segment in the list of the runs on the game page, only in when you scroll down the author's comments does it give it's segmentation category. Other runs like Rise of nations and uplink don't give difficulties; if these games just have one difficulty it'd be nice to know that as well.

[quote][quote]
but most these are very short games and old games and/or done on an easier difficulty?[/quote]

You just go against the very point you try and make less than 2 hours later. Now you are the one assuming that because these games are done in a SS on games where you can save anywhere they are obviously on easy difficulty and short games? Now whos using their intuition? Max Payne 1 (which I did in a SS) was done on the hardest difficulty (DoA) and was 1.5 hours long, not exactly short for a SS. I don't see why it is so hard to tell what difficulty they are done on, most of the people interested in the run are going to be people who have played it. For those who haven't theres a slight hint from watching the run where on the title screen I clearly go to the bottom difficulty setting and it says the ultimate challenge.[/quote]

Key word, most...  It's kind of funny I use the word most then you point out one example as if this means my statement wasn't true.  I actually did look at game page for everyone of those runs.  The max payne 1 run was one of the only ones on highest difficulty and wasn't a short game, of course I just found out DOA is the highest difficulty.  Also as a viewer I'd like to know what DOA difficulty meant without having to DL and watch the run first.  Also higher difficulties can be unlocked in many games so I still wouldn't know it was the highest by watching the run.  The Silent hill 2 run doesn't say the difficulty but there's another run on that game page which it does give a difficulty for.  The max payne 2 run doesn't say it's single segment in the list of the runs on the game page, only in when you scroll down the author's comments does it give it's segmentation category.  Other runs like Rise of nations and uplink don't give difficulties; if these games just have one difficulty it'd be nice to know that as well.

Spider-Waffle

(user is banned)

2009-07-13 03:44:10 pm

Don't think! feeeeeal

Does anyone else find it arbitrary that ILs must be OSPL and aren't allowed to segment more, and segmented full runs are allowed to segment more and don't have a OSPL rule?

If I want to make a multi-segment IL run I can't. If I want to make a OSPL full run I can't if there's already a good segmented run, or can but someone can beat it with multiple segments per level.

What's with these arbitrary, game-dependent category rules?

ShadowWraith

2009-07-13 04:01:04 pm

.

...and if I want to make a multi-segment single-segment run I can't. IL tables are well defined, just as single-segment and segmented run categories are. What's arbitrary about that?

dex

2009-07-13 04:03:28 pm

Invisible avatar

Quote from Spider-Waffle:

Does anyone else find it arbitrary that ILs must be OSPL and aren't allowed to segment more, and segmented full runs are allowed to segment more and don't have a OSPL rule?

Not me. ILs follow for games which have a natural way of selecting the level to play - either through the menu or through a supported console command. The decision to limit segmentation there was Radix's own - I assume because he decided playing the level in one go was the natural way to do it when the level choice was given. Also, most speedrunning communities till that point forbade segmenting in ILs. Finally, in some games there's no saving mid-level in IL runs, even though saving is available during normal gameplay (Max Payne 2, for example). On the other hand, standard segmented is aimed at using the save anywhere feature of the set game. Since different people have different ideas of segmentation, it's better to give the runners the choice on how they want to segment, without dividing their segmentation into TOTALLY arbitrary categories, from which the one-segment-per-level would be the less interesting one.

Quote from Spider-Waffle:

If I want to make a OSPL full run I can't if there's already a good segmented run

But if the run is already very good and you have no new tactics (which would presumably let you tackle the old run OSLP), why bother even if there's an OLSP category? Why would you want a run with the same routes and that's also segmented, but a little less, and is slower? That doesn't make any sense. That you can't beat a run in such a situation (no new routes, no new tricks, less segments) is true; the question is, why would the SDA viewers actually want a run that adds absolutely nothing of worth to the SDA archive?

Quote from Spider-Waffle:

or can but someone can beat it with multiple segments per level.

Except if the improvement was solely due to segmentation, it'd get rejected by the verifiers for being lame and boring.

Quote from Spider-Waffle:

What's with these arbitrary, game-dependent category rules?

Games are different; they are hence treated slightly differently. IL for the games where you can choose the levels, 100% for the games that track some stats etc. That rules strive to avoid being game-dependant doesn't mean they can avoid it always.

[quote="Spider-Waffle"]Does anyone else find it arbitrary that ILs must be OSPL and aren't allowed to segment more, and segmented full runs are allowed to segment more and don't have a OSPL rule?[/quote]
Not me. ILs follow for games which have a natural way of selecting the level to play - either through the menu or through a supported console command. The decision to limit segmentation there was Radix's own - I assume because he decided playing the level in one go was the natural way to do it when the level choice was given. Also, most speedrunning communities till that point forbade segmenting in ILs. Finally, in some games there's no saving mid-level in IL runs, even though saving is available during normal gameplay (Max Payne 2, for example). On the other hand, standard segmented is aimed at using the save anywhere feature of the set game. Since different people have different ideas of segmentation, it's better to give the runners the choice on how they want to segment, without dividing their segmentation into TOTALLY arbitrary categories, from which the one-segment-per-level would be the less interesting one.

[quote="Spider-Waffle"]If I want to make a OSPL full run I can't if there's already a good segmented run[/quote]
But if the run is already very good and you have no new tactics (which would presumably let you tackle the old run OSLP), why bother even if there's an OLSP category? Why would you want a run with the same routes and that's also segmented, but a little less, and is slower? That doesn't make any sense. That you can't beat a run in such a situation (no new routes, no new tricks, less segments) is true; the question is, why would the SDA viewers actually want a run that adds absolutely nothing of worth to the SDA archive?

[quote="Spider-Waffle"]or can but someone can beat it with multiple segments per level.[/quote]
Except if the improvement was solely due to segmentation, it'd get rejected by the verifiers for being lame and boring.

[quote="Spider-Waffle"]What's with these arbitrary, game-dependent category rules?[/quote]
Games are different; they are hence treated slightly differently. IL for the games where you can choose the levels, 100% for the games that track some stats etc. That rules strive to avoid being game-dependant doesn't mean they can avoid it always.