Fire Emblem Genesis

screw playing, i want to argue about it on the internet


You are not connected. Please login or register

Data Transfers

View previous topic View next topic Go down  Message [Page 1 of 1]

1 Data Transfers on Wed Oct 21, 2009 5:25 am

Mekkah

avatar
Admin
I think there should be an FE10 tier list that ignores transfers. Besides the fact that I think the list looks ugly with all the transfers in it, it's dubious whether someone is going to take the time to play through POR just to boost FE10 characters. You could make a comparison with gaiden chapters (which all have opportunity costs but also add value to a playthrough in the form of items, new chars and EXP) and I think it's pretty easy to see that playing the entire game of POR costs much more time than using an inferior char, even if it's the hailed "Easy Mode Ike".

Of course, I believe you can use a transfer as many times as you wish, so over the long term it might be beneficial. But there's also the fact that not everyone has access to POR.

All of this combined is, imo, impossible to properly factor in as opportunity costs for transfer characters. If you want to try it, I say go ahead...but I'd say a more simple version would be a good idea as well.

Thoughts?

View user profile http://fegenesis.forummotion.com

2 Re: Data Transfers on Wed Oct 21, 2009 6:40 am

Mekkah wrote:Thoughts?

I was with you until about halfway through the second sentence. There are only two reasons for extricating transfers from the tier list that I consider to be serious:

1) aesthetics
2) opportunity cost becomes unwieldy due to how the normals were ranked

Whether "someone" plays through PoR is not relevant to any aspect of tiering.

Firstly, tier lists as psychology experiments are doomed to collapse inwards on themselves in a race to the bottom, because everyone has a different personal definition of an "average" player, and there is no way to quantify it at all. As soon as you pivot from talking about potential to talking about likelyhood, you've just pissed in your own Cheerios. And don't tell me that tier lists serve some higher purpose: they are incapable of being useful character guides, due to the format. I guarantee you an epic argument between vykan and myself when he brings this up (again).

Secondly, your actions in PoR have nothing but a positive impact on RD's efficiency and turn counts, which is what the tier list is based on to begin with. We are also not ranking real life time taken to beat this game. Where is the wedge here for PoR to be excluded, when it doesn't impair the only things that we're ranking in the first place? This is a situation that has no real equivalent elsewhere in the series, excepting perhaps a discussion of Lyn Mode.


So I basically agree with the 15% of your post that isn't going to make tiering self-destruct. There is a valid administrative reason to split the list in twain, and a valid aesthetic one, but everything other than that is candy canes and time bombs all the way to infinity.

View user profile

3 Re: Data Transfers on Wed Oct 21, 2009 7:19 am

Mekkah

avatar
Admin
So I basically agree with the 15% of your post that isn't going to make tiering self-destruct. There is a valid administrative reason to split the list in twain, and a valid aesthetic one, but everything other than that is candy canes and time bombs all the way to infinity.

Then we could be done talking. If you agree with my proposal, though for different reasons, I have no real reason to oppose your post. But some things I want to clarify:

Firstly, tier lists as psychology experiments are doomed to collapse inwards on themselves in a race to the bottom, because everyone has a different personal definition of an "average" player, and there is no way to quantify it at all. As soon as you pivot from talking about potential to talking about likelyhood, you've just pissed in your own Cheerios.

I agree there is no point in thinking about "what the average player does" so I'll reform the bit that makes it look like I said there was. The question is "is it more or less efficient overall to play POR for transfer data for RD, or to just play Radiant Dawn?".

We are also not ranking real life time taken to beat this game.

Well, you don't have to use real time per sé. It also takes much more turns total to play through POR+RDwithtransfers than just RDwithouttransfers.

This is a situation that has no real equivalent elsewhere in the series, excepting perhaps a discussion of Lyn Mode.

I compared it to gaidens in the original post, which are much of the same thing: you spend some more in a voluntairy setting to improve performance outside of it. Though Lyn Mode and Gaidens, unlike POR, are not part of the game that is being debated.

View user profile http://fegenesis.forummotion.com

4 Re: Data Transfers on Wed Oct 21, 2009 7:39 am

Mekkah wrote:The question is "is it more or less efficient overall to play POR for transfer data for RD, or to just play Radiant Dawn?".
I think that this question has an invalid premise. See below.

Well, you don't have to use real time per sé. It also takes much more turns total to play through POR+RDwithtransfers than just RDwithouttransfers.
Not from where I'm standing, it doesn't. This is an RD tier list, not a PoR + RD tier list, so turns spent playing PoR are turns that aren't being measured in the first place. Turns spent playing PoR have the same effect as Ike's performance in Smash Bros.

Now, a PoR + RD tier list might be interesting to do, aka tracking Ike's performance in *both* games, transfers included, but that's not what we have here.

I compared it to gaidens in the original post, which are much of the same thing: you spend some more in a voluntairy setting to improve performance outside of it.
I don't think that they are the same thing, seeing as how they exist in the game that you're debating, and have a direct impact on turns counts at the very least.

Though Lyn Mode and Gaidens, unlike POR, are not part of the game that is being debated.
I don't understand what you are saying right there. PoR is not part of the game being debated here, either.

View user profile

5 Re: Data Transfers on Wed Oct 21, 2009 8:12 am

I dunno, I prefer the idea of separate lists for Transfers. I mean, it's pretty much what we have already, just more organized. I mean, not everybody has a fixed disc that takes Easy>Hard Transfers (coughmecough)

And just cuz I'm bored:

TRANSFERS

Easy Mode
Ike

Top (4)
Reyson
Haar
Volug
Mia
Titania

High (12)
Sothe
Gatrie
Leanne
Nailah
Zihark
Ulki
Janaff
Nolan
Shinon
Rafiel
Nephenee

Upper Middle (13)
Jill
Elincia
Micaiah
Black Knight
Tibarn
Naesala
Oscar
Laura
Boyd
Mordecai
Ranulf
Marcia
Aran
Mist

Middle (15)
Heather
Brom
Tauroneo
Rolf
Soren
Rhys
Tanith
Giffca
Caineghis/Giffca
Muarim
Skrimir
Kieran
Makalov
Geoffrey
Calill

Lower Middle (15)
Edward
Lucia
Volke
Stefan
Tormod
Leonardo
Nealuchi
Nasir
Ilyana
Ena
Sigrun
Danved
Vika

Low (10)
Bastian
Kurthnaga
Lethe
Renning
Gareth
Lehran
Oliver
Kyza
Astrid
Sanaki
Pelleas

Bottom (4)
Meg
Fiona
Lyre


okay that just made me more bored. Anybody want to do the No Transfers list?
[I still don't buy Giffca>Cain or even Giffca=Cain, for any reasoning at all. Formshift>>>>>Slightly better transform stats]

View user profile

6 Re: Data Transfers on Wed Oct 21, 2009 8:35 am

bblader1 wrote:[I still don't buy Giffca>Cain or even Giffca=Cain, for any reasoning at all. Formshift>>>>>Slightly better transform stats]
That's Red Fox's baby to defend, but I beleive that the issue in that match-up is that Giff's chance of getting a significant SPD lead makes for a non-trivial difference in their combat performance. He only loses 4 Player Phases to the Formshift difference.

Sounds like a topic for a new thread to me.

View user profile

7 Re: Data Transfers on Wed Oct 21, 2009 8:40 am

Mekkah

avatar
Admin
Not from where I'm standing, it doesn't. This is an RD tier list, not a PoR + RD tier list, so turns spent playing PoR are turns that aren't being measured in the first place. Turns spent playing PoR have the same effect as Ike's performance in Smash Bros.

If you want to fully ignore happenings in POR because it is a different game, then that's pretty much that. I personally think that's having your cake and eat it though: you want to take into account the bonuses Ike got from playing through the game, but not the antics you have to go through just to get that.

Turns spent playing PoR have the same effect as Ike's performance in Smash Bros.

Only one of these is the way to get the +2s for RD Ike, so that's why only one of these has to be taken into account somehow.

I don't understand what you are saying right there. PoR is not part of the game being debated here, either.

reverse that* Lyn Mode and Gaidens are part of the game at hand, POR is not, but both are skipable steps.

B2BD, let's keep on topic alright.

View user profile http://fegenesis.forummotion.com

8 Re: Data Transfers on Wed Oct 21, 2009 9:24 am

Mekkah wrote:If you want to fully ignore happenings in POR because it is a different game, then that's pretty much that. I personally think that's having your cake and eat it though: you want to take into account the bonuses Ike got from playing through the game, but not the antics you have to go through just to get that.
This song's not about me, it doesn't have anything to do with what I want, it has to do with what the existing list measures. We seem to be ranking efficiency and speedy completion of RD, not effort and irl time.

Only one of these is the way to get the +2s for RD Ike, so that's why only one of these has to be taken into account somehow.
I don't think that something should be taken into account just for the sake of it. We clearly take the benefits of the transfer into account already. As for the costs, they exist, but the goals don't appear to care. If there's a button that's all gravy and no grief, so be it. We aren't obligated to puritanize it.

reverse that* Lyn Mode and Gaidens are part of the game at hand, POR is not, but both are skipable steps.
But that's the fundamentally different point between them. Lyn and Gaidens have an in-game cost that can be measured relative to tier goals.

If you had, for example, a HHM tier list based on ranks, Lyn mode is PoR (has zero direct effect on your HHM rank) but the Gaidens may actually harm your rank. The other parallel is that I only need to do Lyn mode once, for an infinite number of HHM runs. A PoR run to make transfers for RD is completely decoupled from the RD run itself, because, as you said, you can use the end result as many times as you want.



And on top of all of this, there's one over-arching result from putting PoR turn-counts on the table, and that's to either:

1) auto-shit-tier every transfer character, or
2) remove them from the list entirely

... both of which kill debate opportunities. I don't see the problem if people want to argue about it, other than the aforementioned logistical problems with ranking the T's and the Normie's on the same list but with different standards.

View user profile

9 Re: Data Transfers on Wed Oct 21, 2009 11:10 am

Yeah, I don't see the point of separating them.

Here's what you get:

List 1:
Normal characters only

List 2:
Normal characters + transfer versions coexisting.


You may notice that List 2 is what we have now, so its purely aesthetics if you want a pretty list that doesn't include transfers.

You can't make a list where everyone that can have a transfer from PoR has one, since you aren't likely to get more than 10 units to 20/20 outside of EM or massive abuse in PoR. Our general idea has been to simply assume a small number of characters receive a transfer at a time so Titania (T) compares mostly to normal characters. Even in the case where we compare Titania(T) to Mia(T) on the list, they don't have to coexist on the same playthrough, so Mia(T)'s team could have either Titania(T) or Titania(N) on it. As a result, you could easily make a HM PoR game and transfer that file. It's not like its hard to get a few characters to 20/20. I got 8 there (Ike, Mia, Mist, Rhys, Ilyana, Elincia, Marcia, Astrid). I'm sure even with a turncount lower than mine you could pull off 5 characters pretty easily.

If we go with bblader1's list, you run into the issue where everyone is assumed to have a transfer, and that's just not practical.


If somebody wants to simply copy all the Ns from the current list and make a new list without any Ts while leaving the current list alone, go ahead, but I don't see much point. It doesn't result in any different comparisons than what we have now. The (N)s are only competing with (T)s in a direct comparison (Like whether or not to put Mak(T) over Tanith(N)). Against the rest of the (N)s, the (T)s are not assumed to exist (Like Tanith (N) vs. Kieran(N) does not assume any other (T)s exist, as Mak(T) would reduce Kieran's utility from 2-3 and 3-9 greatly).

View user profile

10 Re: Data Transfers on Wed Oct 21, 2009 2:34 pm

The only problem I see here is that it looks "messy" (Which, no offense, is a somewhat childish complaint imo). (T) characters have no bearing on (N) characters placements, so if you can't transfer or don't want to discuss them you don't have to. If this forum has collapsible spoiler tags, I could remove the (T) characters, put the new list in one, and that would likely solve the problem.

The only possible problem I could see is if someone ignoring the (T) list argues someone up on the other list that passes a (T) character, but I'm sure things will work out perfectly fine with that.

View user profile

11 Re: Data Transfers on Wed Oct 21, 2009 2:37 pm

There should definitely be 2 lists.

For transfers to occur, you have to account for the following:

-If you’ve even played POR, if you’ve beaten it, possibly multiple times to get fixed mode and bands.
-If you’ve raised the correct units for the transfer. If you use 20 units in RD who are capable of getting decent transfers, maybe only 8 of them will actually get them.
-The units who you got to 20/20 have to have all the correct stats capped. Just because there’s a 60% chance of capping str doesn’t mean it will happen, and assuming reset abuse is pretty taboo.
-There’s reason to assume you’re aiming for efficiency in a POR playthrough. Thus you cannot assume you’re going to save stat boosters, bands, etc for characters you want to achieve transfers with.

I think there might be a couple of other factors as well. If you converted those points into some sort of probability function, the chance of any individual character getting a transfer is pretty slim.

It’s too awkward to compare abstract chances with each other. Instead, we treat transfers in a binary manner (0% chance for every stat in 1 list, 100% chance for specific stats in another).

Come to think of it, transfers should be treated like average stats when we do actually account for them. Like if Boyd has a 35% chance of +2 str, we should add 0.7 to his average.

Edit: There's also the issue of getting transfers through wiisaves. While that does eliminate some of the assumptions, it also adds one about purchased hardware, and allows for greater transfer yields, possibly to the point of cheating (ie +2 to every stat for every character).

View user profile

12 Re: Data Transfers on Wed Oct 21, 2009 2:49 pm

Vykan12 wrote:For transfers to occur, you have to account for the following:
Oh not likely.

-If you’ve even played POR, if you’ve beaten it, possibly multiple times to get fixed mode and bands.
Doesn't matter, doesn't matter, doesn't matter. This is a Hard Mode tier list for RD, which requires a cleared file to begin with, resulting in ninja magic before transfers even come into play.

-If you’ve raised the correct units for the transfer. If you use 20 units in RD who are capable of getting decent transfers, maybe only 8 of them will actually get them.
Transfer characters are currently being considered in a void, not as the whole army, so the number of people who can make it out are immaterial.

-The units who you got to 20/20 have to have all the correct stats capped. Just because there’s a 60% chance of capping str doesn’t mean it will happen, and assuming reset abuse is pretty taboo.
Yeah, and Fixed mode turns a lot of those 60% chances into 100% with room to spare.

-There’s reason to assume you’re aiming for efficiency in a POR playthrough. Thus you cannot assume you’re going to save stat boosters, bands, etc for characters you want to achieve transfers with.
There is actually no reason to assume efficiency in a PoR playthrough, since this list does not rank PoR turn counts or PoR efficiency. There is no reason to assume no-resets in Random Mode, either, but it's been agreed that RNG-abuse doesn't contribute to the discussion.

I think there might be a couple of other factors as well. If you converted those points into some sort of probability function, the chance of any individual character getting a transfer is pretty slim.
I disagree. The stat-up that you see on RFoF's list have a 100% chance of happening. Literally 100%. Not five nines.

Edit: There's also the issue of getting transfers through wiisaves. While that does eliminate some of the assumptions, it also adds one about purchased hardware, and allows for greater transfer yields, possibly to the point of cheating (ie +2 to every stat for every character).
I think that we can agree as reasonable people that haxing +2 for every stat for every character doesn't really result in a more intresting discussion than reasonable transfers based on Fixed mode.



vyklan, I don't understand why you put niggling details on such a pedestal that you allow them to kill interesting debate in the crib. Tier lists serve no other purpose, after all, so...



Last edited by Interceptor on Wed Oct 21, 2009 2:52 pm; edited 1 time in total

View user profile

13 Re: Data Transfers on Wed Oct 21, 2009 2:51 pm

We went over this with a lot of detail in the RD board, so first I'm going to say it's kind of annoying that two people who weren't even there for that are now complaining about it.

It's assuming fixed and normal/easy (Or Hard, I guess. I only say E/N for an easier time reaching 20/20, but if (T) units are compared in a void, it doesn't make much difference) mode, and I did calculations for every single PoR unit to see what they could reasonably cap with bands, etc., included.

-If you’ve even played POR, if you’ve beaten it, possibly multiple times to get fixed mode and bands.
As previously stated, irrelevant. If you haven't played PoR or whatever you don't have to bother with the (T) units on the list. And you only need to beat it once for Fixed mode and bands.

-If you’ve raised the correct units for the transfer. If you use 20 units in RD who are capable of getting decent transfers, maybe only 8 of them will actually get them.
Already accounted for. When comparing (T) units, we compare them mostly to a team of (N) units.

-The units who you got to 20/20 have to have all the correct stats capped. Just because there’s a 60% chance of capping str doesn’t mean it will happen, and assuming reset abuse is pretty taboo.
Read above.

-There’s reason to assume you’re aiming for efficiency in a POR playthrough. Thus you cannot assume you’re going to save stat boosters, bands, etc for characters you want to achieve transfers with.
Honestly, this is a non-issue in PoR because it's just too easy. Not many of the units are even assumed to get stat boosters as it is, and if you're using a team with that unit for the transfer team...Well, where's the issue? It's not as if someone who caps a stat naturally is going to need a booster for that stat.

View user profile

14 Re: Data Transfers on Wed Oct 21, 2009 3:08 pm

I didn’t even notice that the transfer and non-transfer counterparts were in the same list. There goes most of my objections.

There is actually no reason to assume efficiency in a PoR playthrough, since this list does not rank PoR turn counts or PoR efficiency.

Forgive me, but it sounds absurd for a player to abuse like hell in one game to increase efficient play in another. Why would the player have such a contradictory playstyle between games?

The stat-up that you see on RFoF's list have a 100% chance of happening. Literally 100%. Not five nines.

Unless I made a calculation error, Titania’s chances of capping skill is 66.55% even if we give her a skl band and a secret book. That’s nowhere near 100%, and regardless 100% is asymptotic in most cases.

View user profile

15 Re: Data Transfers on Wed Oct 21, 2009 3:11 pm

Vykan12 wrote:Unless I made a calculation error, Titania’s chances of capping skill is 66.55% even if we give her a skl band and a secret book. That’s nowhere near 100%, and regardless 100% is asymptotic in most cases.
With the calculations I did, I determined that Titania can get Skl and Spd with a Secret Book and the correct use of bands/weapons/etc (And the demand for SB was low) with a 100% chance in Fixed mode (or as close as possible before enemies are taken into account). If you need me to show you, I'm sure I can. If it turns out I made a mistake, I'll just make the proper adjustment on the list.

By the way, I added a collapsible spoiler tag without the (T) units on the list.

View user profile

16 Re: Data Transfers on Wed Oct 21, 2009 3:14 pm

Ok, and what happens if a unit needs 4+ stat boosters and constant knight ward usage to get their full listed transfers on fixed mode?

But whatever, I suppose we can neglect that sort of thing.

View user profile

17 Re: Data Transfers on Wed Oct 21, 2009 3:18 pm

Vykan12 wrote:Forgive me, but it sounds absurd for a player to abuse like hell in one game to increase efficient play in another. Why would the player have such a contradictory playstyle between games?
I don't see how an imaginary player's "style" is at all germane to this issue. The rankings in this RD list are based on efficient play, not some phantom's random inclinations.

Unless I made a calculation error, Titania’s chances of capping skill is 66.55% even if we give her a skl band and a secret book. That’s nowhere near 100%, and regardless 100% is asymptotic in most cases.
You made a logic error. Are you unaware of how Fixed mode works?

View user profile

18 Re: Data Transfers on Wed Oct 21, 2009 3:19 pm

Vykan12 wrote:Ok, and what happens if a unit needs 4+ stat boosters and constant knight ward usage to get their full listed transfers on fixed mode?

But whatever, I suppose we can neglect that sort of thing.
No one on the list needs 4+ stat boosters to get what is currently assumed. Some, like Gatrie, need a lot of KW usage, but with trading it around in the base and everything else I don't see that to be such a huge issue.

The most boosters anything anyone has listed needs is 1 or 2 (I don't remember if I added any that need 2) and I did a careful analysis on how that should work which was met with only agreement or neutrality (is that a word?). The only scenario where someone would even need 3 is if Ike gets Res since he needs both Talismans, but I determined that to likely be reasonable as well despite it not sounding so.

I can't tell if your last sentence is sarcasm or not.

View user profile

Sponsored content


View previous topic View next topic Back to top  Message [Page 1 of 1]

Permissions in this forum:
You cannot reply to topics in this forum