Vandal Hearts

2010-01-13 10:12:15 pm

Waiting hurts my soul...

I kind of haven't touched this since last month, but I had some time so I have a fairly good strategy going. Definitely better than my test run since I'm getting through it 1 turn faster (4 turns). I need to worry about getting poisoned from two enemies. I need to advance Sara and Ash for this map, so I might as well advance Huxley and Darius at the same time, which will most likely help in the following map.

I'm not sure when I'll get this done now though, as I haven't had much time at all.

ZenicReverie

2010-01-25 12:47:49 pm

Waiting hurts my soul...

Just thought I'd mention how much this stage (4-5) sucks... 8 chances that a block will mess up the run and 2 chances Poison will. What are people's thoughts on keeping a sub-par segment due to a status ailment? One of those poisons I might be able to get by with and at least execute the plan (albeit about 20 - 30 seconds slower to show the Poison damage over two turns). Everything else will throw it too far off track to save.

I keep going back and forth on how streamlined this run should be since I'll probably be doing nearly the same thing on my next attempt through when I use the leveling exploit.

mikwuyma

2010-01-25 12:58:11 pm

My feelings on The Demon Rush

I'm okay with the poison damage. Is 4-5 the train, or is it rescuing Kira?

ZenicReverie

2010-01-25 02:08:20 pm

Waiting hurts my soul...

Quote from mikwuyma:

I'm okay with the poison damage. Is 4-5 the train, or is it rescuing Kira?

Rescuing Kira.

The train was done in 7 or 8 turns in 2 segments so I could manipulate a block from Zohar.

For this one I'm moving and positioning my guys in a weird way to manipulate the AI. The four blocks in the beginning (Eleni vs a skeleton, Clint vs a skeleton) can't happen or else the AI controlling the Dark Wizard off to the right of starting position won't move to attack Sara for whatever reason. The next turn the Dark Wizard has a high chance (seems to be 2 or 3/5) to block a frontal attack. Then the enemy moves to attack and can't block the next three attacks or I can't do enough damage to kill them off. Also, one of my guys can't be poisoned or the plan falls apart (I'd need to heal him for an extra 15 seconds in addition to the 20 - 30 lost in poison).

ZenicReverie

Edit history:

ZenicReverie: 2010-02-11 08:24:09 am

2010-02-10 12:20:39 am

Waiting hurts my soul...

4-5 done, I'll rewatch the segment to make sure I didn't mess up too badly, but I know of one where instead of flipping through the spell list by pressing up, I pressed down to get to the bottom of the list. Probably costs 2 - 3 seconds, but it was too late to reverse by the time I was consciously aware of it. I also think I miss-stepped once which costs about 2 seconds again, but no poison, and everything else working out is enough for me to move on.

Next up is 5-1. Apparently I've gotten through it in 5 turns, so that's what I'm aiming for, but now I need to figure out the minimum number of people to move. I need Sara to get up to level 21 so she gets the thunder flash spell, which is as simple as casting a spell on Zohar, or I may need to use a healing circle. The two Armors and Clint are not yet 20, I'm not sure if I should bother leveling them up though. I don't think they're that useful beyond this point except as damage targets, which extra movement might be useful for.

Edit: Still haven't tried out 5-1. I got Brain Lord, so I'm more interested in playing that at the moment. I wanted to update though, I just watched the recording for 4-5 again, and it's better than I thought; the spell list flub only costs 1 second, and doesn't look too bad if you're not paying close attention, and the miss-step wasn't in this recording. There is a slight pause at the home free point (except for one last bit of leveling up stats luck that already happened and I'm about to test out) before ending my turn.

4-5 done, I'll rewatch the segment to make sure I didn't mess up too badly, but I know of one where instead of flipping through the spell list by pressing up, I pressed down to get to the bottom of the list. Probably costs 2 - 3 seconds, but it was too late to reverse by the time I was consciously aware of it. I also think I miss-stepped once which costs about 2 seconds again, but no poison, and everything else working out is enough for me to move on.

Next up is 5-1. Apparently I've gotten through it in 5 turns, so that's what I'm aiming for, but now I need to figure out the minimum number of people to move. I need Sara to get up to level 21 so she gets the thunder flash spell, which is as simple as casting a spell on Zohar, or I may need to use a healing circle. The two Armors and Clint are not yet 20, I'm not sure if I should bother leveling them up though. I don't think they're that useful beyond this point except as damage targets, which extra movement might be useful for.

Edit: Still haven't tried out 5-1. I got Brain Lord, so I'm more interested in playing that at the moment. I wanted to update though, I just watched the recording for 4-5 again, and it's better than I thought; the spell list flub only costs 1 second, and doesn't look too bad if you're not paying close attention, and the miss-step wasn't in this recording. There is a slight pause at the home free point (except for one last bit of leveling up stats luck that already happened and I'm about to test out) before ending my turn.

ZenicReverie

2010-02-13 11:15:47 am

Waiting hurts my soul...

I don't know how, but I managed to shave off another turn from this stage. There was a pretty tense moment when I first found this run where I almost reset because if the AI was smart it would have been game over, but for whatever reason the AI doesn't end me and I go on to win. This happens every time, so there doesn't appear to be any manipulation necessary. However, there is some manipulation at the beginning of 5-1 where I need both mages to cast thunder ball on the group to the left. Sometimes they use their single target spell and I just can't make that work in a timely manner.

Over the past two stages (first time I've been able to save properly) I've gained another 11 minutes over my test run. I'm thinking another 2 minutes in 5-1 with tighter execution. 5-2 is short and I've tested out a few strategies that will probably work, so that will hopefully be a quick one. As of the end of 5-1, I'm looking at a total time under 4 hours.

ZenicReverie

Edit history:

ZenicReverie: 2010-02-15 05:47:25 pm

2010-02-15 12:43:20 am

Waiting hurts my soul...

And, 5-2 is down. I figured it'd be fairly easily since it's probably one of the shortest battles outside of chapter 1. Gained 4 minutes over the test run.

5-3 has a lot going on, but I have a strategy down for it that I think is pretty solid. Now just to pull it off. So many guys that can block, and one hit that could poison me. Sadly neither Zohar nor Eleni are 25 yet, so no Salamander going into the dry bones cave to fight the salamander, although I'll probably get it in the middle of the stage; then I get to hope to not encounter the same bug that caused me not to get the wyrm fang at the end of the stage. I've never seen it before, nor have I read it happening anywhere, so I'm hoping it's something easily figured out like needing to kill a certain enemy other than the salamander.

So, 4:01 starting 5-3. I have to go to a town, enter and leave the tavern (the first time I don't have to talk to everyone) and then I get news a little girl needs help. I leave town and move on to the stage. After this is 5-4, where I just need to leave the town and head to the cave. 5-5 happens right after I'm done with the previous stage. I head back to town and bam I'm in 5-5. 5-6 is the stage I need to baby sit the NPC as she activates switches, and where I pick up my second wyrm fang.

Then comes 6-1 where I kill Kane, 6-2 which is like a filler level where enemies blocking are more annoyances than deadly. 6-3 is where I kill Xeno; 6-4 kill Dolf.

All this means at least 8 more segments until the end. I may end up breaking down some levels. The potential candidates I can think of are 5-5, 6-1, and possibly 5-6 because those are the longer stages.

Edit: I'm updating the current est time to 7h, but realistically, if I keep improving things at the same rate this run will most likely go sub-6. Keep your fingers crossed.

And, 5-2 is down. I figured it'd be fairly easily since it's probably one of the shortest battles outside of chapter 1. Gained 4 minutes over the test run.

5-3 has a lot going on, but I have a strategy down for it that I think is pretty solid. Now just to pull it off. So many guys that can block, and one hit that could poison me. Sadly neither Zohar nor Eleni are 25 yet, so no Salamander going into the dry bones cave to fight the salamander, although I'll probably get it in the middle of the stage; then I get to hope to not encounter the same bug that caused me not to get the wyrm fang at the end of the stage. I've never seen it before, nor have I read it happening anywhere, so I'm hoping it's something easily figured out like needing to kill a certain enemy other than the salamander.

So, 4:01 starting 5-3. I have to go to a town, enter and leave the tavern (the first time I don't have to talk to everyone) and then I get news a little girl needs help. I leave town and move on to the stage. After this is 5-4, where I just need to leave the town and head to the cave. 5-5 happens right after I'm done with the previous stage. I head back to town and bam I'm in 5-5. 5-6 is the stage I need to baby sit the NPC as she activates switches, and where I pick up my second wyrm fang.

Then comes 6-1 where I kill Kane, 6-2 which is like a filler level where enemies blocking are more annoyances than deadly. 6-3 is where I kill Xeno; 6-4 kill Dolf.

All this means at least 8 more segments until the end. I may end up breaking down some levels. The potential candidates I can think of are 5-5, 6-1, and possibly 5-6 because those are the longer stages.

Edit: I'm updating the current est time to 7h, but realistically, if I keep improving things at the same rate this run will most likely go sub-6. Keep your fingers crossed.

rekenner

2010-02-19 06:17:14 am

Hm. I saw this from the FF:MQ thread. A friend of mine is a ... master at this game, and he's done a run through of the game in about 6.5h with becoming a Vandalier. I'll prod him into reading through this topic and getting him to comment over the weekend, to help you tighten this up.

(Though, his initial comment is that Vandalier Ash *might* speed the game up, but...)

ZenicReverie

Edit history:

ZenicReverie: 2010-02-19 09:59:00 am

2010-02-19 08:02:58 am

Waiting hurts my soul...

Quote from rekenner:

Hm. I saw this from the FF:MQ thread. A friend of mine is a ... master at this game, and he's done a run through of the game in about 6.5h with becoming a Vandalier. I'll prod him into reading through this topic and getting him to comment over the weekend, to help you tighten this up.

(Though, his initial comment is that Vandalier Ash *might* speed the game up, but...)

Well if he's at 6.5 hours with getting Vandalier, then this run will be faster. I take it he's not doing the leveling exploit either? How far along is he? This thread and some videos on youtube have been around for a while, where have you guys been? Roll Eyes

It'd be good to get some feedback on the game.

I say this run will be faster because in my head I'm estimating sub-6, although that all depends on execution. The 'current est' in the topic is basically my current time up to where I am plus whatever time it took me in my previous test run, which I've been pretty consistently surpassing every segment by 10+ minutes in the last chapter (obviously the longer the levels the greater the improvement). With 8 timing points/segments left that's another 1h20m that will potentially come off. So I guess a better estimate is 5h40m when everything is said and done, but I'd rather just go along with it and see where the time lands for this.

Vandalier will definitely be slower though. You only get Vandalier during chapter 6, and chapter 6 takes all of 1(estimate) - 1.5hours(last test run). The time spent in the training levels and the time spent getting the items will definitely cancel out the time saved by having the Vandalier.

I'm not sure if this is posted anywhere but my party is Huxley and Sara = Ninja, Grog and Dolan = Armor, Diego and Darius = Archer, Kira and Amon = Airman, Clint = Swordsman. It also could be I might be able to help your friend out with some strategies of his own.

Edit:

On the topic of the run, I've got a good strategy for 5-3 now, I just need to execute it and the mage AI needs to cooperate already. To Vandal Hearts: Just decide if you're going to use Harmful Wave or Thunder Ball, because this wishy washy attitude is making it hard to plan things out.

The first mage that moves can cast Harmful Wave and it doesn't throw things off too much, it just means Sara doesn't gain a level from her heal. The second though will target Dolan which kills him, and Thunder Ball would have hurt Amon, which brings his HP low enough that the Skeleton attacks him. Instead the Skeleton attacks Ash, and his counter attack doesn't bring the Skeleton's HP low enough to die to the Phase Shift he should be dying to, which means I need to spend someone's turn attacking him, which I can't spare. So what causes the AI to choose one spell over another? Might as well be button presses for all I can figure, because choosing the same commands doesn't decide it.

[quote="rekenner"]
Hm. I saw this from the FF:MQ thread. A friend of mine is a ... master at this game, and he's done a run through of the game in about 6.5h [b]with[/b] becoming a Vandalier. I'll prod him into reading through this topic and getting him to comment over the weekend, to help you tighten this up.

(Though, his initial comment is that Vandalier Ash *might* speed the game up, but...)
[/quote]
Well if he's at 6.5 hours with getting Vandalier, then this run will be faster. I take it he's not doing the leveling exploit either? How far along is he? This thread and some videos on youtube have been around for a while, where have you guys been? ::) ;) It'd be good to get some feedback on the game.

I say this run will be faster because in my head I'm estimating sub-6, although that all depends on execution. The 'current est' in the topic is basically my current time up to where I am plus whatever time it took me in my previous test run, which I've been pretty consistently surpassing every segment by 10+ minutes in the last chapter (obviously the longer the levels the greater the improvement). With 8 timing points/segments left that's another 1h20m that will potentially come off. So I guess a better estimate is 5h40m when everything is said and done, but I'd rather just go along with it and see where the time lands for this.

Vandalier will definitely be slower though. You only get Vandalier during chapter 6, and chapter 6 takes all of 1(estimate) - 1.5hours(last test run). The time spent in the training levels and the time spent getting the items will definitely cancel out the time saved by having the Vandalier.

I'm not sure if this is posted anywhere but my party is Huxley and Sara = Ninja, Grog and Dolan = Armor, Diego and Darius = Archer, Kira and Amon = Airman, Clint = Swordsman. It also could be I might be able to help your friend out with some strategies of his own.

Edit:

On the topic of the run, I've got a good strategy for 5-3 now, I just need to execute it and the mage AI needs to cooperate already. To Vandal Hearts: Just decide if you're going to use Harmful Wave or Thunder Ball, because this wishy washy attitude is making it hard to plan things out.

The first mage that moves can cast Harmful Wave and it doesn't throw things off too much, it just means Sara doesn't gain a level from her heal. The second though will target Dolan which kills him, and Thunder Ball would have hurt Amon, which brings his HP low enough that the Skeleton attacks him. Instead the Skeleton attacks Ash, and his counter attack doesn't bring the Skeleton's HP low enough to die to the Phase Shift he should be dying to, which means I need to spend someone's turn attacking him, which I can't spare. So what causes the AI to choose one spell over another? Might as well be button presses for all I can figure, because choosing the same commands doesn't decide it.

rekenner

Edit history:

rekenner: 2010-02-19 12:47:58 pm

2010-02-19 10:53:39 am

Well, I just made an account, finally, due to Rane.
So.
And - Again, this is from a brief convo with him this morning, that's with Vandaliers, but I think without leveling exploit (though I was wrong), tho he did say you're doing it in an inefficient way.

And, his 6.5h run was a psuedo-speed run, I don't think he's actually going to try to speed run the game any time soon. He's slaughtered the game in other ways (He had a theory, if you were to never lose a character in the entire game, including the gorram civilians in the zombie fight, you'd never have to sell anything to buy the best equips as they come in. He was right.) - so, I think he'd be good at finding the squishy bits in your playing.

ZenicReverie

Edit history:

ZenicReverie: 2010-02-19 02:18:28 pm

2010-02-19 01:53:27 pm

Waiting hurts my soul...

Quote from rekenner:

Well, I just made an account, finally, due to Rane.
So.
And - Again, this is from a brief convo with him this morning, that's with Vandaliers, but I think without leveling exploit (though I was wrong), tho he did say you're doing it in an inefficient way.

Doing what in an inefficient way? If you mean the leveling exploit, then I'm currently not doing that, so what is the efficient way? I'm considering this:

Sara casting Mystic Shield on herself while off the healing square. Then once MP dips to the threshold where I won't be able to cast next turn, I step on the square after the Mystic Shield. I then step off after re-filling MP and repeat.

Also, that paragraph doesn't make a whole lot of sense.

Quote from rekenner:

And, his 6.5h run was a psuedo-speed run, I don't think he's actually going to try to speed run the game any time soon. He's slaughtered the game in other ways (He had a theory, if you were to never lose a character in the entire game, including the gorram civilians in the zombie fight, you'd never have to sell anything to buy the best equips as they come in. He was right.) - so, I think he'd be good at finding the squishy bits in your playing.

Why wouldn't you sell your equipment?

What exactly do you mean by 'psuedo'? Seems pretty quick if it only took an hour or so to get through the 6 Dojo stages. Also, I can see the squishy bits in my playing. It'd be my planning he could help me with.

I'd also really like to know if he's ever ran into the problem of not getting the wyrmfang from the salamander.

[quote="rekenner"]
Well, I just made an account, finally, due to Rane.
So.
And - Again, this is from a brief convo with him this morning, that's with Vandaliers, but I think without leveling exploit (though I was wrong), tho he did say you're doing it in an inefficient way.
[/quote]
Doing what in an inefficient way? If you mean the leveling exploit, then I'm currently not doing that, so what is the efficient way? I'm considering this:

Sara casting Mystic Shield on herself while off the healing square. Then once MP dips to the threshold where I won't be able to cast next turn, I step on the square after the Mystic Shield. I then step off after re-filling MP and repeat.

Also, that paragraph doesn't make a whole lot of sense.

[quote="rekenner"]
And, his 6.5h run was a psuedo-speed run, I don't think he's actually going to try to speed run the game any time soon. He's slaughtered the game in other ways (He had a theory, if you were to never lose a character in the entire game, including the gorram civilians in the zombie fight, you'd never have to sell anything to buy the best equips as they come in. He was right.) - so, I think he'd be good at finding the squishy bits in your playing.
[/quote]
Why wouldn't you sell your equipment?

What exactly do you mean by 'psuedo'? Seems pretty quick if it only took an hour or so to get through the 6 Dojo stages. Also, I can see the squishy bits in my playing. It'd be my planning he could help me with.

I'd also really like to know if he's ever ran into the problem of not getting the wyrmfang from the salamander.

rekenner

2010-02-19 06:02:23 pm

Well, he was trying to play through the game quick, but this was before he'd heard of the concept of a speed run.
And why not sell your equipment? ... *shrug*. Testing a theory.

aetherspoon

2010-02-19 06:12:01 pm

I'd be the friend he was referring to. Smiley

I wish I would have done this run more recently - I did it back when I first decided to use an emulator to play a PS1 game, so that was... a really long time ago and well before I had heard about doing a modern-style speedrun (or having video recording abilities, sadly).

My run was around 6.5h by the end of the game, Ash as a Vandalier, using the leveling exploit on just Sara (it was way too slow for my tastes to go with both healers doing it) and using the exploit until I was a high enough level to oneshot everything in the game. I didn't collect anything but the required items for the Vandalier process though.
The only thing that I can think of was that the run might not have included the final set of battles (the last two), as I believe I was measuring time with the save. Not sure, it has been years. Smiley

The reason why I think going non-Vandalier is less efficient is due to the loss of a third offensive spellcaster. Sure, Ash can oneshot things pretty easily as a Vandalier, but if you level up high enough so can everyone else. The real important part is the triple casting of Salamander after everyone closes in on your group, which is enough to kill most non-boss non-mage enemies in the game.

Your run might end up being more efficient; it is pretty close in my mind, but I haven't tried optimizing a run for VH without going Vandalier. Exploit leveling both Hux and Sara is very inefficient though.

Rekenner is right though, I was really just testing a theory re:selling equipment. I tend to do that - one wouldn't believe some of the crazy theories I've tried with VH and ShF2. Tongue

I'd be the friend he was referring to.  :)
I wish I would have done this run more recently - I did it back when I first decided to use an emulator to play a PS1 game, so that was...  a really long time ago and well before I had heard about doing a modern-style speedrun (or having video recording abilities, sadly).

My run was around 6.5h by the end of the game, Ash as a Vandalier, using the leveling exploit on just Sara (it was way too slow for my tastes to go with both healers doing it) and using the exploit until I was a high enough level to oneshot everything in the game.  I didn't collect anything but the required items for the Vandalier process though.
The only thing that I can think of was that the run might not have included the final set of battles (the last two), as I believe I was measuring time with the save.  Not sure, it has been years.  :)

The reason why I think going non-Vandalier is less efficient is due to the loss of a third offensive spellcaster.  Sure, Ash can oneshot things pretty easily as a Vandalier, but if you level up high enough so can everyone else.  The real important part is the triple casting of Salamander after everyone closes in on your group, which is enough to kill most non-boss non-mage enemies in the game.

Your run might end up being more efficient; it is pretty close in my mind, but I haven't tried optimizing a run for VH without going Vandalier.  Exploit leveling both Hux and Sara is very inefficient though.

Rekenner is right though, I was really just testing a theory re:selling equipment.  I tend to do that - one wouldn't believe some of the crazy theories I've tried with VH and ShF2.  :P

ZenicReverie

Edit history:

ZenicReverie: 2010-02-19 08:26:55 pm

2010-02-19 08:11:41 pm

Waiting hurts my soul...

Quote from aetherspoon:

I'd be the friend he was referring to. Smiley

I wish I would have done this run more recently - I did it back when I first decided to use an emulator to play a PS1 game, so that was... a really long time ago and well before I had heard about doing a modern-style speedrun (or having video recording abilities, sadly).

Did you use save stats for making sure an enemy didn't block? I've been wondering if things are truly random or if it's calculated.

Quote from aetherspoon:

My run was around 6.5h by the end of the game, Ash as a Vandalier, using the leveling exploit on just Sara (it was way too slow for my tastes to go with both healers doing it) and using the exploit until I was a high enough level to oneshot everything in the game. I didn't collect anything but the required items for the Vandalier process though.
The only thing that I can think of was that the run might not have included the final set of battles (the last two), as I believe I was measuring time with the save. Not sure, it has been years. Smiley

So, 6.5 without counting the Xeno or Dolf fight, so that's about another 10 - 20 minutes? I'm not sure what you mean by both healers doing it. The exploit is to use one healer and then have everyone else use an item or spell on the healer.

Quote from aetherspoon:

The reason why I think going non-Vandalier is less efficient is due to the loss of a third offensive spellcaster. Sure, Ash can oneshot things pretty easily as a Vandalier, but if you level up high enough so can everyone else. The real important part is the triple casting of Salamander after everyone closes in on your group, which is enough to kill most non-boss non-mage enemies in the game.

Okay I have a problem with this though.

1) It takes 6 extra battles before you can be a Vandalier. Let's say, they average 5 minutes. That's 30 more minutes to the run. It's been a while since I've bothered with Vandalier, but I think it'd be longer, but let's just say 30 because that's probably close to the minimum.
2) To get the hidden items also takes extra time. But, let's not count that.
3) Vandalier is only available for Chapter 6, which is 4 maps. It takes about 1 hour to clear Chapter 6.

All this means that just to match the time with a Vandalier, you'd need to finish the last four maps in 30 minutes. I'm just not sure that's possible. There's a very small amount of time to be saved, if any. Probably 10 - 15 minutes max if the Vandalier can really break all the maps by spamming plasma wave. And, if it takes more than 30 minutes to get the Vandalier, then that time is further reduced.

Quote from aetherspoon:

Your run might end up being more efficient; it is pretty close in my mind, but I haven't tried optimizing a run for VH without going Vandalier.

Have you taken a look at my videos? You can comment on them if you have any ideas, although I'm past those points so it's a little late to make changes, but when I do the exploit run it would come in handy.

Quote from aetherspoon:

Exploit leveling both Hux and Sara is very inefficient though.

I never suggested exploit leveling both Huxley and Sara. I have no idea where you got that idea from.

Quote from aetherspoon:

Rekenner is right though, I was really just testing a theory re:selling equipment. I tend to do that - one wouldn't believe some of the crazy theories I've tried with VH and ShF2. Tongue

That's cool. What else have you tried in Vandal Hearts?

[quote="aetherspoon"]
I'd be the friend he was referring to.  :)
I wish I would have done this run more recently - I did it back when I first decided to use an emulator to play a PS1 game, so that was...  a really long time ago and well before I had heard about doing a modern-style speedrun (or having video recording abilities, sadly).
[/quote]
Did you use save stats for making sure an enemy didn't block? I've been wondering if things are truly random or if it's calculated.

[quote="aetherspoon"]
My run was around 6.5h by the end of the game, Ash as a Vandalier, using the leveling exploit on just Sara (it was way too slow for my tastes to go with both healers doing it) and using the exploit until I was a high enough level to oneshot everything in the game.  I didn't collect anything but the required items for the Vandalier process though.
The only thing that I can think of was that the run might not have included the final set of battles (the last two), as I believe I was measuring time with the save.  Not sure, it has been years.  :)
[/quote]
So, 6.5 without counting the Xeno or Dolf fight, so that's about another 10 - 20 minutes? I'm not sure what you mean by both healers doing it. The exploit is to use one healer and then have everyone else use an item or spell on the healer.

[quote="aetherspoon"]
The reason why I think going non-Vandalier is less efficient is due to the loss of a third offensive spellcaster.  Sure, Ash can oneshot things pretty easily as a Vandalier, but if you level up high enough so can everyone else.  The real important part is the triple casting of Salamander after everyone closes in on your group, which is enough to kill most non-boss non-mage enemies in the game.
[/quote]
Okay I have a problem with this though.

1) It takes 6 extra battles before you can be a Vandalier. Let's say, they average 5 minutes. That's 30 more minutes to the run. It's been a while since I've bothered with Vandalier, but I think it'd be longer, but let's just say 30 because that's probably close to the minimum.
2) To get the hidden items also takes extra time. But, let's not count that.
3) Vandalier is only available for Chapter 6, which is 4 maps. It takes about 1 hour to clear Chapter 6.

All this means that just to match the time with a Vandalier, you'd need to finish the last four maps in 30 minutes. I'm just not sure that's possible. There's a very small amount of time to be saved, if any. Probably 10 - 15 minutes max if the Vandalier can really break all the maps by spamming plasma wave. And, if it takes more than 30 minutes to get the Vandalier, then that time is further reduced.

[quote="aetherspoon"]
Your run might end up being more efficient; it is pretty close in my mind, but I haven't tried optimizing a run for VH without going Vandalier.
[/quote]
Have you taken a look at my videos? You can comment on them if you have any ideas, although I'm past those points so it's a little late to make changes, but when I do the exploit run it would come in handy.

[quote="aetherspoon"]
Exploit leveling both Hux and Sara is very inefficient though.
[/quote]
I never suggested exploit leveling both Huxley and Sara. I have no idea where you got that idea from.

[quote="aetherspoon"]
Rekenner is right though, I was really just testing a theory re:selling equipment.  I tend to do that - one wouldn't believe some of the crazy theories I've tried with VH and ShF2.  :P
[/quote]
That's cool. What else have you tried in Vandal Hearts?

aetherspoon

2010-02-19 09:27:34 pm

Quote from ZenicReverie:

Did you use save stats for making sure an enemy didn't block? I've been wondering if things are truly random or if it's calculated.

Save states on that one. Random block/poison drove me batty and all I've done is RNG muddling.

Quote from ZenicReverie:

1) It takes 6 extra battles before you can be a Vandalier. Let's say, they average 5 minutes. That's 30 more minutes to the run. It's been a while since I've bothered with Vandalier, but I think it'd be longer, but let's just say 30 because that's probably close to the minimum.
2) To get the hidden items also takes extra time. But, let's not count that.
3) Vandalier is only available for Chapter 6, which is 4 maps. It takes about 1 hour to clear Chapter 6.

All this means that just to match the time with a Vandalier, you'd need to finish the last four maps in 30 minutes. I'm just not sure that's possible. There's a very small amount of time to be saved, if any. Probably 10 - 15 minutes max if the Vandalier can really break all the maps by spamming plasma wave. And, if it takes more than 30 minutes to get the Vandalier, then that time is further reduced.

I think the last four maps in thirty minutes might be doable thanks to PWave - that's why I said it'd be close. One of the bonus maps is significantly longer than average (although having only fliers helps a bit), one significantly shorter than average, so I'd say that would be a reasonable guess. Probably not enough of a difference to try it, since it'd change the entire run.

Quote from ZenicReverie:

Have you taken a look at my videos? You can comment on them if you have any ideas, although I'm past those points so it's a little late to make changes, but when I do the exploit run it would come in handy.

Not yet, I'll do so when I have more time on the weekend. All I've done so far is a quick glance and comment to rekenner about things.

Quote from ZenicReverie:

I never suggested exploit leveling both Huxley and Sara. I have no idea where you got that idea from.

Sorry about that, misinterpreted something said earlier (or misread and my brain misinterpreted the rest, take your pick).

Quote from ZenicReverie:

That's cool. What else have you tried in Vandal Hearts?

Trying to figure out the most optimal way time-wise to solve that peasant battle, trying to figure out the exact differences between the same enemies of different classes, trying to figure out at what level different characters start one-shotting everything (I never did finish figuring that one out)... that's all that I remember at the moment. I'll try to see if I have any more of my notes.

[quote="ZenicReverie"]
Did you use save stats for making sure an enemy didn't block? I've been wondering if things are truly random or if it's calculated.
[/quote]
Save states on that one.  Random block/poison drove me batty and all I've done is RNG muddling.

[quote="ZenicReverie"]
1) It takes 6 extra battles before you can be a Vandalier. Let's say, they average 5 minutes. That's 30 more minutes to the run. It's been a while since I've bothered with Vandalier, but I think it'd be longer, but let's just say 30 because that's probably close to the minimum.
2) To get the hidden items also takes extra time. But, let's not count that.
3) Vandalier is only available for Chapter 6, which is 4 maps. It takes about 1 hour to clear Chapter 6.

All this means that just to match the time with a Vandalier, you'd need to finish the last four maps in 30 minutes. I'm just not sure that's possible. There's a very small amount of time to be saved, if any. Probably 10 - 15 minutes max if the Vandalier can really break all the maps by spamming plasma wave. And, if it takes more than 30 minutes to get the Vandalier, then that time is further reduced.[/quote]

I think the last four maps in thirty minutes might be doable thanks to PWave - that's why I said it'd be close.  One of the bonus maps is significantly longer than average (although having only fliers helps a bit), one significantly shorter than average, so I'd say that would be a reasonable guess.  Probably not enough of a difference to try it, since it'd change the entire run.

[quote="ZenicReverie"]
Have you taken a look at my videos? You can comment on them if you have any ideas, although I'm past those points so it's a little late to make changes, but when I do the exploit run it would come in handy.
[/quote]

Not yet, I'll do so when I have more time on the weekend.  All I've done so far is a quick glance and comment to rekenner about things.

[quote="ZenicReverie"]
I never suggested exploit leveling both Huxley and Sara. I have no idea where you got that idea from.
[/quote]
Sorry about that, misinterpreted something said earlier (or misread and my brain misinterpreted the rest, take your pick).

[quote="ZenicReverie"]
That's cool. What else have you tried in Vandal Hearts?
[/quote]

Trying to figure out the most optimal way time-wise to solve that peasant battle, trying to figure out the exact differences between the same enemies of different classes, trying to figure out at what level different characters start one-shotting everything (I never did finish figuring that one out)...  that's all that I remember at the moment.  I'll try to see if I have any more of my notes.

ZenicReverie

2010-02-19 10:07:24 pm

Waiting hurts my soul...

Quote from aetherspoon:

Save states on that one. Random block/poison drove me batty and all I've done is RNG muddling.

Yeah, it's pretty crazy, but did you determine that you could save state right before the enemies turn and it'd be different each time? I mean is it truly random, because it sure does seem that way.

Quote from aetherspoon:

I think the last four maps in thirty minutes might be doable thanks to PWave - that's why I said it'd be close. One of the bonus maps is significantly longer than average (although having only fliers helps a bit), one significantly shorter than average, so I'd say that would be a reasonable guess. Probably not enough of a difference to try it, since it'd change the entire run.

I remember one being all treasure chests, which would be easy to just pick the correct chest. Also, I guess all of the Dojo battles would be easier if you underleveled Ash while exploiting everyone else.

Quote from aetherspoon:

Not yet, I'll do so when I have more time on the weekend. All I've done so far is a quick glance and comment to rekenner about things.

Let me know if you have any questions. Also, if there are any stages you remember doing something well and I don't have it on youtube, I can put it up for you or describe it at least.

Quote from aetherspoon:

Sorry about that, misinterpreted something said earlier (or misread and my brain misinterpreted the rest, take your pick).

no problem.

Quote from aetherspoon:

Trying to figure out the most optimal way time-wise to solve that peasant battle, trying to figure out the exact differences between the same enemies of different classes, trying to figure out at what level different characters start one-shotting everything (I never did finish figuring that one out)... that's all that I remember at the moment. I'll try to see if I have any more of my notes.

In speedrunning I've noticed that level bonuses can be different based on the RNG. This includes HP, Atk, Def, and possibly Agl. MP seems to be set based no level. So, figuring out the level wouldn't be as helpful as figuring out the minimum Atk needed. There have been times I'll gain a level, but get low Atk and an attack that had previously killed an enemy would leave them with 1 or 2 HP. Yet, this only seems to last a stage or two, so maybe there's some equalizing code to make sure stats stay close to a baseline.

Do you remember how you did on the peasant battle? I got it down to 10 turns and no peasants dead. I think it was about 10 minutes.

[quote="aetherspoon"]
Save states on that one.  Random block/poison drove me batty and all I've done is RNG muddling.
[/quote]
Yeah, it's pretty crazy, but did you determine that you could save state right before the enemies turn and it'd be different each time? I mean is it truly random, because it sure does seem that way.

[quote="aetherspoon"]
I think the last four maps in thirty minutes might be doable thanks to PWave - that's why I said it'd be close.  One of the bonus maps is significantly longer than average (although having only fliers helps a bit), one significantly shorter than average, so I'd say that would be a reasonable guess.  Probably not enough of a difference to try it, since it'd change the entire run.
[/quote]
I remember one being all treasure chests, which would be easy to just pick the correct chest. Also, I guess all of the Dojo battles would be easier if you underleveled Ash while exploiting everyone else.

[quote="aetherspoon"]
Not yet, I'll do so when I have more time on the weekend.  All I've done so far is a quick glance and comment to rekenner about things.
[/quote]
Let me know if you have any questions. Also, if there are any stages you remember doing something well and I don't have it on youtube, I can put it up for you or describe it at least.

[quote="aetherspoon"]
Sorry about that, misinterpreted something said earlier (or misread and my brain misinterpreted the rest, take your pick).
[/quote]
no problem.

[quote="aetherspoon"]
Trying to figure out the most optimal way time-wise to solve that peasant battle, trying to figure out the exact differences between the same enemies of different classes, trying to figure out at what level different characters start one-shotting everything (I never did finish figuring that one out)...  that's all that I remember at the moment.  I'll try to see if I have any more of my notes.
[/quote]
In speedrunning I've noticed that level bonuses can be different based on the RNG. This includes HP, Atk, Def, and possibly Agl. MP seems to be set based no level. So, figuring out the level wouldn't be as helpful as figuring out the minimum Atk needed. There have been times I'll gain a level, but get low Atk and an attack that had previously killed an enemy would leave them with 1 or 2 HP. Yet, this only seems to last a stage or two, so maybe there's some equalizing code to make sure stats stay close to a baseline.

Do you remember how you did on the peasant battle? I got it down to 10 turns and no peasants dead. I think it was about 10 minutes.

aetherspoon

2010-02-20 08:27:18 am

I've had a chance to look at your videos now. I don't suppose you happen to have one of 2-1? I have a couple of theories on that stage and I'd like to see how you did it. 3-2 is another potential point where different oddball strategies could have a dramatically different effect on speed.

Nice job on 4-1, by the way. That was substantially faster than my version (about the same number of turns, but less time per turn).

A few things I'm curious about:

1) You mentioned stats being random on level up. The unusual thing is that I had come to the opposite conclusion from just using save states (they were consistent) - meaning the random seed for level up stats must change when you save/load, which I didn't do that test trip through VH. If I remember right, I even came to the conclusion that it stayed the same every level. Out of curiosity, if you level up the same character twice in one battle without saving in-between, do they have the same level up stats for both levels, or did my limited test just happen to turn out that way? If so, that might be an interesting way to optimize the run, but not really feasible on an actual console. Smiley

2) On the other hand, AI tactics changed practically every save state, like you thought. My states were immediately before the AI's turn and I could reload to not only change the block/status chance, but also the AI tactics. The AI seemed to weight different strategies where one particular tactic was the most common, but it would randomly change up and do something completely different. The interesting part is that the AI didn't just send random orders to a unit; they were all focused into a single strategy and that strategy seemed to change. How much RNG manipulation have you been doing on this?

3) On stage 3-7, did you have Ash die? That's the only level in the game that Ash can die without instantly losing (not counting the stage that he isn't in), and I don't think most people realize that. Not sure if it would make a difference in the speed run that you're currently working on, but it makes a huge difference in the exploit/Vandalier run (since I kept Ash at around level 14ish, meaning I had to spend a bit of time manipulating the AI to not attack him).

4) On 6-1... any particular reason why you're keeping all of your characters alive through P-Wave? I mean, yeah, it might help a bit with the battle, but you're eating both the time for P-Wave and the time for HolyH2O. I haven't tried speeding this battle without a Vandalier though, so I don't know how well the lone-Ash strategy works without one.

5) What's your ideas regarding promotion paths? With using the exploit, the movement speeds of each of the characters matters far more than anything else, so going all ninja over mage might even be the best choice.

I've had a chance to look at your videos now.  I don't suppose you happen to have one of 2-1?  I have a couple of theories on that stage and I'd like to see how you did it.  3-2 is another potential point where different oddball strategies could have a dramatically different effect on speed.

Nice job on 4-1, by the way.  That was substantially faster than my version (about the same number of turns, but less time per turn).

A few things I'm curious about:

1) You mentioned stats being random on level up.  The unusual thing is that I had come to the opposite conclusion from just using save states (they were consistent) - meaning the random seed for level up stats must change when you save/load, which I didn't do that test trip through VH.  If I remember right, I even came to the conclusion that it stayed the same every level.  Out of curiosity, if you level up the same character twice in one battle without saving in-between, do they have the same level up stats for both levels, or did my limited test just happen to turn out that way?  If so, that might be an interesting way to optimize the run, but not really feasible on an actual console.  :)

2) On the other hand, AI tactics changed practically every save state, like you thought.  My states were immediately before the AI's turn and I could reload to not only change the block/status chance, but also the AI tactics.  The AI seemed to weight different strategies where one particular tactic was the most common, but it would randomly change up and do something completely different.  The interesting part is that the AI didn't just send random orders to a unit; they were all focused into a single strategy and that strategy seemed to change.  How much RNG manipulation have you been doing on this?

3) On stage 3-7, did you have Ash die?  That's the only level in the game that Ash can die without instantly losing (not counting the stage that he isn't in), and I don't think most people realize that.  Not sure if it would make a difference in the speed run that you're currently working on, but it makes a huge difference in the exploit/Vandalier run (since I kept Ash at around level 14ish, meaning I had to spend a bit of time manipulating the AI to not attack him).

4) On 6-1...  any particular reason why you're keeping all of your characters alive through P-Wave?  I mean, yeah, it might help a bit with the battle, but you're eating both the time for P-Wave and the time for HolyH2O.  I haven't tried speeding this battle without a Vandalier though, so I don't know how well the lone-Ash strategy works without one.

5) What's your ideas regarding promotion paths?  With using the exploit, the movement speeds of each of the characters matters far more than anything else, so going all ninja over mage might even be the best choice.

ZenicReverie

2010-02-20 01:03:12 pm

Waiting hurts my soul...

Quote from aetherspoon:

I've had a chance to look at your videos now. I don't suppose you happen to have one of 2-1? I have a couple of theories on that stage and I'd like to see how you did it. 3-2 is another potential point where different oddball strategies could have a dramatically different effect on speed.

I can upload both. 3-2 is basically just wait 5 turns, not sure how that can be sped up very much.

Quote from aetherspoon:

Nice job on 4-1, by the way. That was substantially faster than my version (about the same number of turns, but less time per turn).

Thanks, it won't be as easy when I do the exploit run because my guys won't die as quickly. But they'll be moving farther, so maybe I'll just get everyone to the exit and plow through anyone that's in the way.

Quote from aetherspoon:

1) You mentioned stats being random on level up. The unusual thing is that I had come to the opposite conclusion from just using save states (they were consistent) - meaning the random seed for level up stats must change when you save/load, which I didn't do that test trip through VH. If I remember right, I even came to the conclusion that it stayed the same every level. Out of curiosity, if you level up the same character twice in one battle without saving in-between, do they have the same level up stats for both levels, or did my limited test just happen to turn out that way? If so, that might be an interesting way to optimize the run, but not really feasible on an actual console. Smiley

I haven't done any serious testing on the stat increases, but it seems less like you get stat increases and more like when you level up it chooses a random stat from a range. As an example one level the Atk stat can be from 90-92 and the next level can be from 93-95 instead of each level you get a 1 - 3 increase. It's actually a good way to keep random stats from affecting the entire game too much. It's still random though between saving and loading, so maybe it uses a different RNG from the AI.

Quote from aetherspoon:

2) On the other hand, AI tactics changed practically every save state, like you thought. My states were immediately before the AI's turn and I could reload to not only change the block/status chance, but also the AI tactics. The AI seemed to weight different strategies where one particular tactic was the most common, but it would randomly change up and do something completely different. The interesting part is that the AI didn't just send random orders to a unit; they were all focused into a single strategy and that strategy seemed to change. How much RNG manipulation have you been doing on this?

Yeah, this seems truly random. I don't know how the RNG works though, so I don't know how to manipulate it. However, I am playing through each level a few times to see what different strategies the AI uses, and then choosing which gives me the biggest advantage in terms of speed, although there are times when I exchange optimum luck with the chance to save my sanity.

Quote from aetherspoon:

3) On stage 3-7, did you have Ash die? That's the only level in the game that Ash can die without instantly losing (not counting the stage that he isn't in), and I don't think most people realize that. Not sure if it would make a difference in the speed run that you're currently working on, but it makes a huge difference in the exploit/Vandalier run (since I kept Ash at around level 14ish, meaning I had to spend a bit of time manipulating the AI to not attack him).

Nope, Ash doesn't die. Darius is the only one that dies because he's so under-leveled and doesn't have the first class change.

Quote from aetherspoon:

4) On 6-1... any particular reason why you're keeping all of your characters alive through P-Wave? I mean, yeah, it might help a bit with the battle, but you're eating both the time for P-Wave and the time for HolyH2O. I haven't tried speeding this battle without a Vandalier though, so I don't know how well the lone-Ash strategy works without one.

Well, that strategy isn't really set in stone. It's just what I used in the test run. Strategies have been changing throughout the run, so that in turn makes other strategies obsolete or others available. I don't think Ash alone would work very well without Vandalier, and if I remember correctly the Ninjas didn't have the healing power to keep people alive. I'm going to have to test that level out though. The armors die immediately to the first P-Wave in any case. I just need to figure out if I can draw Kain down from his perch or the minimum number of people I need to take him and his guards out.

Quote from aetherspoon:

5) What's your ideas regarding promotion paths? With using the exploit, the movement speeds of each of the characters matters far more than anything else, so going all ninja over mage might even be the best choice.

I'm not sure why mage wouldn't be better over ninja. I mean with mages you get Salamander to take out groups of enemies in one spell. I'll look into that more when I do the exploit run, but for this run I decided on 2 airmen, 2 archers, 2 armors, 1 swordsman, 2 ninja, 2 mages. For the exploit run, I'm thinking 3 swordsman and 3 or 4 airmen, but I really need to figure out what level to bring everyone up to and how useful people are going to be in each of the stages. It'll probably good to have some people under-leveled to be targets.

[quote="aetherspoon"]
I've had a chance to look at your videos now.  I don't suppose you happen to have one of 2-1?  I have a couple of theories on that stage and I'd like to see how you did it.  3-2 is another potential point where different oddball strategies could have a dramatically different effect on speed.
[/quote]
I can upload both. 3-2 is basically just wait 5 turns, not sure how that can be sped up very much.

[quote="aetherspoon"]
Nice job on 4-1, by the way.  That was substantially faster than my version (about the same number of turns, but less time per turn).
[/quote]
Thanks, it won't be as easy when I do the exploit run because my guys won't die as quickly. But they'll be moving farther, so maybe I'll just get everyone to the exit and plow through anyone that's in the way.

[quote="aetherspoon"]
1) You mentioned stats being random on level up.  The unusual thing is that I had come to the opposite conclusion from just using save states (they were consistent) - meaning the random seed for level up stats must change when you save/load, which I didn't do that test trip through VH.  If I remember right, I even came to the conclusion that it stayed the same every level.  Out of curiosity, if you level up the same character twice in one battle without saving in-between, do they have the same level up stats for both levels, or did my limited test just happen to turn out that way?  If so, that might be an interesting way to optimize the run, but not really feasible on an actual console.  :)
[/quote]
I haven't done any serious testing on the stat increases, but it seems less like you get stat increases and more like when you level up it chooses a random stat from a range. As an example one level the Atk stat can be from 90-92 and the next level can be from 93-95 instead of each level you get a 1 - 3 increase. It's actually a good way to keep random stats from affecting the entire game too much. It's still random though between saving and loading, so maybe it uses a different RNG from the AI.

[quote="aetherspoon"]
2) On the other hand, AI tactics changed practically every save state, like you thought.  My states were immediately before the AI's turn and I could reload to not only change the block/status chance, but also the AI tactics.  The AI seemed to weight different strategies where one particular tactic was the most common, but it would randomly change up and do something completely different.  The interesting part is that the AI didn't just send random orders to a unit; they were all focused into a single strategy and that strategy seemed to change.  How much RNG manipulation have you been doing on this?
[/quote]
Yeah, this seems truly random. I don't know how the RNG works though, so I don't know how to manipulate it. However, I am playing through each level a few times to see what different strategies the AI uses, and then choosing which gives me the biggest advantage in terms of speed, although there are times when I exchange optimum luck with the chance to save my sanity.

[quote="aetherspoon"]
3) On stage 3-7, did you have Ash die?  That's the only level in the game that Ash can die without instantly losing (not counting the stage that he isn't in), and I don't think most people realize that.  Not sure if it would make a difference in the speed run that you're currently working on, but it makes a huge difference in the exploit/Vandalier run (since I kept Ash at around level 14ish, meaning I had to spend a bit of time manipulating the AI to not attack him).
[/quote]
Nope, Ash doesn't die. Darius is the only one that dies because he's so under-leveled and doesn't have the first class change.

[quote="aetherspoon"]
4) On 6-1...  any particular reason why you're keeping all of your characters alive through P-Wave?  I mean, yeah, it might help a bit with the battle, but you're eating both the time for P-Wave and the time for HolyH2O.  I haven't tried speeding this battle without a Vandalier though, so I don't know how well the lone-Ash strategy works without one.
[/quote]
Well, that strategy isn't really set in stone. It's just what I used in the test run. Strategies have been changing throughout the run, so that in turn makes other strategies obsolete or others available. I don't think Ash alone would work very well without Vandalier, and if I remember correctly the Ninjas didn't have the healing power to keep people alive. I'm going to have to test that level out though. The armors die immediately to the first P-Wave in any case. I just need to figure out if I can draw Kain down from his perch or the minimum number of people I need to take him and his guards out.

[quote="aetherspoon"]
5) What's your ideas regarding promotion paths?  With using the exploit, the movement speeds of each of the characters matters far more than anything else, so going all ninja over mage might even be the best choice.
[/quote]
I'm not sure why mage wouldn't be better over ninja. I mean with mages you get Salamander to take out groups of enemies in one spell. I'll look into that more when I do the exploit run, but for this run I decided on 2 airmen, 2 archers, 2 armors, 1 swordsman, 2 ninja, 2 mages. For the exploit run, I'm thinking 3 swordsman and 3 or 4 airmen, but I really need to figure out what level to bring everyone up to and how useful people are going to be in each of the stages. It'll probably good to have some people under-leveled to be targets.

ZenicReverie

2010-02-21 12:00:12 am

Waiting hurts my soul...

Some more samples requested by aetherspoon and mikwuyma

When zombie villagers attack.

[flash=425,340]

Okay, now wait five turns and the enemy will give up.

[flash=425,340]

Ash: "Just a tiny bit longer..." Me: "Uhh, thanks for the suggestion Ash, but I think I'll just go now"

[flash=425,340]

This stage is split into two segments for luck manipulation. Xeno has raised the dead as a filler level to introduce the connection between him a Zohar. Although it's hard to read as the dialogue scrolls by too quickly.

[flash=425,340]

aetherspoon

2010-02-21 10:24:41 am

Interesting take on 2-1. I wonder if it would actually be faster to have a low level version of Eleni sitting back and blocking an exit (low level so she won't kill in one shot) just so all of the villagers won't move.

For 3-2, is that faster than just sending a character to each corner and having the flames attack the character instead? The enemy turns would be a lot shorter, in my mind, but then you're having to move characters, so I'm not really sure.

ZenicReverie

2010-02-21 01:28:58 pm

Waiting hurts my soul...

Quote from aetherspoon:

Interesting take on 2-1. I wonder if it would actually be faster to have a low level version of Eleni sitting back and blocking an exit (low level so she won't kill in one shot) just so all of the villagers won't move.

That might be interesting. Although I need her to help kill off one of those statues, and get her to level 10 in the next level to get the class change before the swamp. If I could get away with block one of the exit on the left side of the first house, then it might be faster. Although, I'm not sure since attacking and counter-attacking takes a bit of time. Huxley won't help either, as I want him to be a Ninja during the next level.

Quote from aetherspoon:

For 3-2, is that faster than just sending a character to each corner and having the flames attack the character instead? The enemy turns would be a lot shorter, in my mind, but then you're having to move characters, so I'm not really sure.

The purpose of this setup is to get Zohar as much exp as possible. Enemy turns won't be shorter because now you're dealing with 4 - 6 attacks every turn instead of having two turns where I wipe them out with phase shift. Each attack takes 7 - 9 seconds depending on the animations involved while phase shift takes approximately 11 - 12 seconds of animation plus 1 - 2 seconds for each enemy damaged.

I need Zohar to be level 16 so w/ Sara's Mage Oil he can cast Phase Shift three times next level. Then by two stages later he will be level 18 where he can cast it three times all by himself. A lot of these strategies have been worked out for the long term strategy, although some things haven't worked out (the armors still each have a mage gem, and Ash is still holding the mood ring picked up in Chapter 2.).

[quote="aetherspoon"]
Interesting take on 2-1.  I wonder if it would actually be faster to have a low level version of Eleni sitting back and blocking an exit (low level so she won't kill in one shot) just so all of the villagers won't move.
[/quote]
That might be interesting. Although I need her to help kill off one of those statues, and get her to level 10 in the next level to get the class change before the swamp. If I could get away with block one of the exit on the left side of the first house, then it might be faster. Although, I'm not sure since attacking and counter-attacking takes a bit of time. Huxley won't help either, as I want him to be a Ninja during the next level.

[quote="aetherspoon"]
For 3-2, is that faster than just sending a character to each corner and having the flames attack the character instead?  The enemy turns would be a lot shorter, in my mind, but then you're having to move characters, so I'm not really sure.
[/quote]
The purpose of this setup is to get Zohar as much exp as possible. Enemy turns won't be shorter because now you're dealing with 4 - 6 attacks every turn instead of having two turns where I wipe them out with phase shift. Each attack takes 7 - 9 seconds depending on the animations involved while phase shift takes approximately 11 - 12 seconds of animation plus 1 - 2 seconds for each enemy damaged.

I need Zohar to be level 16 so w/ Sara's Mage Oil he can cast Phase Shift three times next level. Then by two stages later he will be level 18 where he can cast it three times all by himself. A lot of these strategies have been worked out for the long term strategy, although some things haven't worked out (the armors still each have a mage gem, and Ash is still holding the mood ring picked up in Chapter 2.).

aetherspoon

2010-02-22 05:13:19 pm

Ah, but when you do the run using the exploit, you don't need to worry about Zohar's level as much, correct?

ZenicReverie

2010-02-22 07:04:16 pm

Waiting hurts my soul...

Quote from aetherspoon:

Ah, but when you do the run using the exploit, you don't need to worry about Zohar's level as much, correct?

True. It's something I'll keep in mind for that.

ZenicReverie

Edit history:

ZenicReverie: 2010-03-12 09:43:23 am

2010-03-12 09:24:32 am

Waiting hurts my soul...

So, I've got a fairly good segment of 5-3. The only thing that's holding me back from moving on is that if one of the AIs chooses a different spell, then it's an instant 10 seconds saved minus whatever input mess ups I have; so I'll keep trying to get that kind of luck, but if I don't have anything by the end of the week then I'm moving on to 5-4. This segment saves 10 minutes again over the test run, which is really nice in such a short segment. I don't imagine I'll see the same kind of savings for the next segment though, which is even shorter.

So, from further observation, it appears that the enemy actually randomly (maybe random with some decision making, like if the AoE I chose can only target one guy then choose the direct damage spell instead) chooses a spell, and then strategically decides where to target that spell based on the normal deciding factors (killing a guy, most damage, direction of the guy [not sure if that affects spells, but it seems to]). These might even be weighted somehow, but for the most part, the strategic decisions of the AI can be manipulated consistently, while the random elements cannot.

Edit:
I'm going to upload this stage and a couple of the worst stages of this run from chapter 1 to get feedback. I'm pretty sure I'll just rerun 1-1 thru 1-5 to get a better time since I'm sure 3+ minutes can come off, of course, I'm often wrong on those estimates. For timing, I think rounding up another minute on the last save of the reruns, then rounding up a minute on the save of 1-5 of the current run, taking the difference and subtracting from the final time would be easiest. That way the actual timing won't be off by more than a minute, and it'll provide a penalty against me I believe. Although, now that I'm thinking about it, the math might be off. I'll have to run some samples. Anyway, it's officially been one year since I started recording, so happy birthday Vandal Hearts run.

If there are any requests for further segments, let me know by tonight so I can burn them on the same disc. Thanks.

So, I've got a fairly good segment of 5-3. The only thing that's holding me back from moving on is that if one of the AIs chooses a different spell, then it's an instant 10 seconds saved minus whatever input mess ups I have; so I'll keep trying to get that kind of luck, but if I don't have anything by the end of the week then I'm moving on to 5-4. This segment saves 10 minutes again over the test run, which is really nice in such a short segment. I don't imagine I'll see the same kind of savings for the next segment though, which is even shorter.

So, from further observation, it appears that the enemy actually randomly (maybe random with some decision making, like if the AoE I chose can only target one guy then choose the direct damage spell instead) chooses a spell, and then strategically decides where to target that spell based on the normal deciding factors (killing a guy, most damage, direction of the guy [not sure if that affects spells, but it seems to]). These might even be weighted somehow, but for the most part, the strategic decisions of the AI can be manipulated consistently, while the random elements cannot.

Edit:
I'm going to upload this stage and a couple of the worst stages of this run from chapter 1 to get feedback. I'm pretty sure I'll just rerun 1-1 thru 1-5 to get a better time since I'm sure 3+ minutes can come off, of course, I'm often wrong on those estimates. For timing, I think rounding up another minute on the last save of the reruns, then rounding up a minute on the save of 1-5 of the current run, taking the difference and subtracting from the final time would be easiest. That way the actual timing won't be off by more than a minute, and it'll provide a penalty against me I believe. Although, now that I'm thinking about it, the math might be off. I'll have to run some samples. Anyway, it's officially been one year since I started recording, so happy birthday Vandal Hearts run.

If there are any requests for further segments, let me know by tonight so I can burn them on the same disc. Thanks.

ZenicReverie

Edit history:

ZenicReverie: 2010-03-15 07:56:33 am

2010-03-13 12:44:02 am

Waiting hurts my soul...

Alright people, here are some more samples.

These two are the worst two segments in my opinion. Like I said, I'm really not satisfied with these two, so I'll probably redo chapter 1 from 1-1 up to stage 1-5 for good measure. Only thing that would stop me from redoing the segments is someone saying I really don't need to. By someone I mean mostly Mike, but feel free to give your opinion on it.

[flash=425,340]

And this is the current segment I hope to improve with a little more luck manipulation, but I think it's pretty good for the luck I did get.

[flash=425,340]

Edit: So, no luck getting better luck plus the same execution; I'll be moving on now. 5-4 should be fairly easy (how many times have I said that?), but there is a fair amount of enemy blocking if I remember correctly. 5-5 is the next segment I'm least looking forward to, and like I said before I don't expect a whole lot of time saved over the test run here.

Alright people, here are some more samples.

These two are the worst two segments in my opinion. Like I said, I'm really not satisfied with these two, so I'll probably redo chapter 1 from 1-1 up to stage 1-5 for good measure. Only thing that would stop me from redoing the segments is someone saying I really don't need to. By someone I mean mostly Mike, but feel free to give your opinion on it.
[spoiler="Stage 1-2"][flash=425,340]http://www.youtube.com/v/6AOXc1UkqaY&hl=en&fs=1[/flash][/spoiler]
[spoiler="Stage 1-5"][flash=425,340]http://www.youtube.com/v/n1nazCb4C60&hl=en&fs=1[/flash][/spoiler]

And this is the current segment I hope to improve with a little more luck manipulation, but I think it's pretty good for the luck I did get.
[spoiler="Stage 5-3"][flash=425,340]http://www.youtube.com/v/GFrlxvGzXX0&hl=en&fs=1[/flash][/spoiler]

Edit: So, no luck getting better luck plus the same execution; I'll be moving on now. 5-4 should be fairly easy (how many times have I said that?), but there is a fair amount of enemy blocking if I remember correctly. 5-5 is the next segment I'm least looking forward to, and like I said before I don't expect a whole lot of time saved over the test run here.