D&D 3E/3.5 Your take on Mirror Image, 3.0 or 3.5

[N'raac]

Several illusory duplicates of you pop into being, making it difficult for enemies to know which target to attack. The figments stay near you and disappear when struck.
Mirror image creates 1d4 images plus one image per three caster levels (maximum eight images total). These figments separate from you and remain in a cluster, each within 5 feet of at least one other figment or you. You can move into and through a mirror image. When you and the mirror image separate, observers can’t use vision or hearing to tell which one is you and which the image. The figments may also move through each other. The figments mimic your actions, pretending to cast spells when you cast a spell, drink potions when you drink a potion, levitate when you levitate, and so on.
Enemies attempting to attack you or cast spells at you must select from among indistinguishable targets. Generally, roll randomly to see whether the selected target is real or a figment. Any successful attack against an image destroys it. An image’s AC is 10 + your size modifier + your Dex modifier. Figments seem to react normally to area

Yes, if the caster is holding a weapon, so would the images. Would the be impaled? Only if the DM decides to functionally disallow the spell by being a jerk.

Given the spell indicates a targeting is generally achieved by a random roll, it seems like a GM who rules a structure where that is not the case is functionally weakening the spell. So is that GM also a jerk?

Why are they behind the attacker?

Because that was the only way to spread out eight images into their own separate squares, considering the 5’ requirement, the terrain, and the location of the wizard, his allies and the opponents? Because the attacker cleverly worked out that, if they are arranged in a grid of 3x3, the wizard would logically wish to be in the back and advanced

That means he passed through two images, which the spell allows, and is now sharing a space with one of them, now that I consider it. This seems to contradict the “no sharing squares” rule. If the attacker can’t pass through a figment’s square, the wizard should arrange the images to force the enemy to carve through them first, augmenting the spell.

Unless the caster was standing right next to the attacker when the spell was cast, and the caster then failed to move (even a 5 foot step), there's no requirement that any of the images (or the original) be within striking distance of the attacker. The attacker would have to decide where to step up to, and where to strike.

It seems pretty likely the attacker would seldom have the ability to move within striking range of every image, especially if our wizard keeps backpedaling.

Okay, so we're assuming that the images form a line, rather than the "cluster" described in the spell. With you so far...

The terrain in my example forces that. However, it does beg the question who decides how they “cluster”. Does the caster make that choice? The GM? Most spells give the caster the ability to place his spell, and most spells with mobile effects indicate how that movement is determined. Funny how this one doesn’t – almost as if the designer perceives no need to address which squares the images are in.

By your illustration, they're all within 5 feet of another image. They're just not within five feet of the caster. If the attacker has the Spellcraft to recognize the anomaly then they'll know that the caster, standing alone, has to be real.

Good point - I had not considered the potential for the caster to separate from the cluster. The description does technically say they remain in a cluster, not that the wizard does.

Can the wizard control them to move one way while he slips away another? Perhaps he casts Invisibility, all images vanish with him, he slips away and then dismisses the invisibility. Can he make the images go the other way so they reappear in line of sight and he does not?

Oh wait, “the figments stay near you”. So how near is that? Seems quite ill defined if the intent is that they fan out within a large area rather than the “Personal” range of the spell.

Are you asking if the caster can move 50 feet with a five foot step? I think the question answers itself. No. Or am I misunderstanding something?

Then once any action identifies him, it should become pretty tough to become indistinguishable from all the images in many cases.

Okay, my question answered. Your example presumes an arrangement (a chorus line) that the spell description doesn't include. It talks about a "cluster", and never says that the caster has direct control, other than the chance to switch places with images. Also, I'm guessing that you're envisioning a wall at the left side of our text frame, so images couldn't shuffle past?

That frames it nicely. You are quite correct that the spell does not indicate who controls placement of the images. So why would we presume it is not the caster? Can the GM decide they all fan out in a manner that makes it easier for the opponents to reach the caster? Does the caster get to decide, and choose to his advantage? SOMEONE has to place them, and there is no rule stating how that is to be done. Again, almost as if the designer sees no need for such a rule.

While moving, I can merge and split off. That’s even a 5’ step. It’s hard to believe this confounds the viewer if I’m the only one with limited movement.

Do you recall where I pointed out that the "roll randomly" part was the "general rule", not the universal/only rule? You're trying to invent scenarios where that general rule doesn't work. Which is why it's a general rule, not the universal/only rule.

Yet no other rule is provided, nor any guidance for when the general rule does not apply. Again, like the designer perceives no need to identify the exceptions.

Here’s a thought – what does the FAQ say (duh!) I’ve provided the whole FAQ on the spell for completeness.

Are the multiple figments from a mirror image spell legal targets for cleaving? That is, if you have the Cleave feat and you hit an image and destroy it, can you then attack another target within reach (such as another figment
from the spell or perhaps the spell user)? What about Whirlwind Attack? Can you use this feat to attack all the images around the spell user? What about spells that allow multiple targets, such as magic missile? Can you aim magic missiles at different images?

For all intents and purposes, the figments from a foe’s mirror image spell are your foes. You aim your spells and your
attacks at the figments just as though they were real creatures. Any spell you can aim at a creature you can aim at an image. When you use a spell that allows you to select multiple creatures as targets, such as magic missile, you can choose
multiple images as targets. If you have the Cleave or Great Cleave feat, destroying an image with a melee attack triggers the feat (and your cleaving attack might well strike the spell user instead of another image). Likewise, you can use Whirlwind Attack to strike at any image you can reach. A Whirlwind Attack almost certainly will allow you to strike once at the spell user.

Is there a way to decide which squares the figments from a mirror image spell occupy? Or do the images distribute themselves randomly? If it’s the latter, how does the DM decide where they go?\

Although the spell description says the images from a mirror image spell always stay within 5 feet of either the user or another image, it’s easiest to assume that all the images occupy the same space the spell user occupies. Any attack that can reach the user’s space can affect an image.

The mirror image spell description says the images have an Armor Class of 10 + size modifier + Dexterity modifier. Can you improve this with spells the spellcaster casts on herself, such as shield or mage armor? If so, why doesn’t the spell description say the images have the caster’s Armor Class? What happens if the caster has cover from her surroundings? Will cover improve the images’ ACs? What about concealment? Will fog or foliage produce a miss chance for a foe that aims an attack at an image? What about magical concealment, such as a blur or displacement spell?

The images from a mirror image spell don’t use the caster’s Armor Class. Use the formula in the spell description to calculate each image’s Armor Class (10 + caster’s size modifier + caster’s Dexterity modifier). Use the caster’s current Dexterity modifier for each image’s Armor Class, no matter how the caster happened to get that modifier. Any Armor Class improvements the caster might have from equipment she carries or wears, or from magic operating on her person, don’t apply to the images. For example, a Medium user with a Dexterity score of 16, a shield spell, and a suit of +2 leather armor has an Armor Class of 21 (10 +4 shield, +4 armor, and +3 Dexterity), but her images have an Armor Class of 13 (10 +3 Dexterity).

It’s easiest to assume the images share the user’s location on the battlefield, and gain any cover bonuses that might apply to the spell user in that location. If the character in the previous example were behind cover, she would have an Armor Class of 25 and her images would have an Armor Class of 17. If the user has concealment from her surroundings, the images have the same concealment. The images also look just like the caster, and they share purely visual effects such as the blur or displacement spell. If the mirror image user is also using either of these effects, an attack aimed at an image has the same miss chance an attack aimed at the caster has.

What happens if a mirror image user is incorporeal? Are the user’s images also incorporeal? Do attacks aimed at the images have the incorporeal miss chance? If the incorporeal user moves through a wall, can the images move through
the wall, too? What happens if the user goes to another plane? Do the images go along? What if the mirror image user employs a blink spell?

Incorporeal spell casters create corporeal effects. So the figments from an incorporeal user’s mirror image spell are themselves corporeal. Attacks aimed at the images have no incorporeal miss chance. The images, however, appear like the caster and move as the caster moves. If an incorporeal user moves through a wall, its mirror images also appear to move through the wall. If a mirror image user moves to another plane, the images go along. If the user employs a blink spell, the images blink right along with the user, and any attack aimed at an image has the same miss chance (50%) it has if aimed at the caster.

What happens when an attacker accidentally uses a touch spell against a figment from a mirror image spell? You can hold the charge with a touch spell, right? So if you touch an image (which really isn’t there), is the touch spell discharged? Does the touch spell user get a chance to disbelieve the image and avoid discharging the spell?

D&D FAQ v.3.5 85 Update Version: 6/30/08
As noted in an earlier question, the figments from a mirror image spell function just like creatures for the purpose of aiming spells. If a foe using a touch spell touches an image, the spell is harmlessly discharged (though the image is destroyed). There is no chance to disbelieve a mirror image spell—if there was, the spell would have a saving throw entry and the entry would read “Will disbelief.”

So the recommended assumption, repeated more than once, is that the caster and images share one location. When the caster moves, the images move along with him, so their reshuffling is not moving several squares, but shuffling within a smaller space. That will be a disadvantage if the opponent has Great Cleave or WWA –as the FAQ makes clear!

A wizard, holding a pike, casts Mirror Image. Are at least some of the images impaled? If they were in separate squares they wouldn't have to be, but all crowded into one? A lot harder to envision them not being impaled, isn't it?

Much easier to envision the images passing through one another, and the caster, repeatedly, as the spell indicates.

A caster with his entourage is standing on a ledge, facing an attacker who can fly. Attacker swings his spiked chain at the caster. How does he "hit" only one image? He can't have images dancing aside, there's no place for them to dance to.

The same way he hits only one target under any other scenario. An attack hits a single target.

Alternately, what we're left with is that always unsatisfactory explanation of "Because it's magic".

Well, it IS magic, isn’t it? We also abstract a lot of attack mechanics, and that’s part of the explanation as well.

But I'm curious: When the spell says the images "separate from you", how/why do you read that as "The images don't separate from you"?

I don’t. I read “the images separate and continue to move around and through you always staying within 5’”. I then read that consistent with the view that the person is not a 5’x5’ mass, but stands within the square, generally moving around, possibly reaching into any of the eight adjacent squares, on an ongoing basis, rather than standing stock still between actions.

A lot of your other points are irrelevant if we follow the FAQ. You then scientifically analyze spell issues, followed by suggesting we not do so. And at the end, we agree that we should

let the spell work "the way it's supposed to". Don't try to dissect it. Don't try to logic it apart. Play it as written, and get on with the game.

To me, the FAQ indicates the way the spell is supposed to work. Done?

To address how I dealt with issus like on a rock or on a pike.

Presuming the caster could be standing on a pike the images would NOT be standing on pikes, they would be 5 feet from the caster down and to the side. On a rock same thing. The pike would NOT be duplicated.

As far as the in mid-air thing. The combat area is a 3 dimensional space and the spell says that the images need only appear within five feet of the caster or another image. You could have one Above you, one behind you one in front of you, so on and so forth. They would appear to be walking in mid air if you walked.

So how does the spell know whether the images should touch the ground, float above it or be buried halfway or all the way in it? Seems like all the images are supposed to look like they could be the caster, which not standing on the ground may screw up.

 

[Greenfield]

I'm going to do two things here.

First, I'll note that the FAQ says "it's simplest to assume" that they're all in one square. In stating that that's the simplest assumption, it makes it clear that it's an assumption, not necessarily the rule. Like the random dice roll, it's just the easiest way to handle it, and easy is fine.

There are lots of times when we've seen the Sage and the FAQ give answers that run counter to the written rules. I see this as one of them.

The second thing I'm going to do is something I've done before: Concede that you are free to treat the FAQ as gospel, and run your game any way you like. I trust that you'll extend me the sme courtesy, and accdept that I'm free to read the RAW as written, treat that as gospel, and run my games the way I like.


As a final note, by the RAW the images aren't constantly passing through and switching places with the caster. That only takes place when the caster moves. i.e. on his turn. So while you keep saying that, it's not part of the spell.

 

[GaimMastr]

Duration

Why would it be a waste? Because it takes a full round action to search one five foot square.
That will tell you if that one image is real or not, nothing more. If you're searching the wrong square you only learn about that one square. Unless, of course, the image appears to avoid your groping hands, just as the real caster would. If that happens, you learn nothing.

Alternately, if all the images are in one square, constantly shifting and switching places, all you've done is verify something you already knew, that one of those images is the caster. But unless you've grabbed hold (something you can't do as part of a Search), you've got nothing. The image you identified can get lost in the shuffle. Earlier editions included the note that, once you identified an image as the caster you could continue to attack that image for the rest of the round. That text is notably missing from the 3.5 SRD.

Most people still play as if the text were there, but by RAW, identifying which image is real gives you nothing.

And whatever you do manage to learn becomes obsolete next round.

The duration of the spell is one minute per level, so even though you may have discovered the real caster, you are still fighting with figments until the spell ends (1 min/lev). The caster could have taken a 5ft step after you discovered them and because the images move around, you'll again not know which one is real again, next turn. And I also deal with searching a little more leniently I allow a person in combat to spend a full round action to search the 3 areas to the front and front-left/front-right with one check. When not rushed or threatened they can search the squares all around them including the one they are on as a standard modification to the rules. I have have too many players search every square on their way into a dungeon taking literally a whole session to make it to the next room cause they wanted to search every square.

 

[GaimMastr]

Easy

Given the spell indicates a targeting is generally achieved by a random roll, it seems like a GM who rules a structure where that is not the case is functionally weakening the spell. So is that GM also a jerk?



Because that was the only way to spread out eight images into their own separate squares, considering the 5’ requirement, the terrain, and the location of the wizard, his allies and the opponents? Because the attacker cleverly worked out that, if they are arranged in a grid of 3x3, the wizard would logically wish to be in the back and advanced

That means he passed through two images, which the spell allows, and is now sharing a space with one of them, now that I consider it. This seems to contradict the “no sharing squares” rule. If the attacker can’t pass through a figment’s square, the wizard should arrange the images to force the enemy to carve through them first, augmenting the spell.



It seems pretty likely the attacker would seldom have the ability to move within striking range of every image, especially if our wizard keeps backpedaling.



The terrain in my example forces that. However, it does beg the question who decides how they “cluster”. Does the caster make that choice? The GM? Most spells give the caster the ability to place his spell, and most spells with mobile effects indicate how that movement is determined. Funny how this one doesn’t – almost as if the designer perceives no need to address which squares the images are in.



Good point - I had not considered the potential for the caster to separate from the cluster. The description does technically say they remain in a cluster, not that the wizard does.

Can the wizard control them to move one way while he slips away another? Perhaps he casts Invisibility, all images vanish with him, he slips away and then dismisses the invisibility. Can he make the images go the other way so they reappear in line of sight and he does not?

Oh wait, “the figments stay near you”. So how near is that? Seems quite ill defined if the intent is that they fan out within a large area rather than the “Personal” range of the spell.



Then once any action identifies him, it should become pretty tough to become indistinguishable from all the images in many cases.



That frames it nicely. You are quite correct that the spell does not indicate who controls placement of the images. So why would we presume it is not the caster? Can the GM decide they all fan out in a manner that makes it easier for the opponents to reach the caster? Does the caster get to decide, and choose to his advantage? SOMEONE has to place them, and there is no rule stating how that is to be done. Again, almost as if the designer sees no need for such a rule.

While moving, I can merge and split off. That’s even a 5’ step. It’s hard to believe this confounds the viewer if I’m the only one with limited movement.



Yet no other rule is provided, nor any guidance for when the general rule does not apply. Again, like the designer perceives no need to identify the exceptions.

Here’s a thought – what does the FAQ say (duh!) I’ve provided the whole FAQ on the spell for completeness.



So the recommended assumption, repeated more than once, is that the caster and images share one location. When the caster moves, the images move along with him, so their reshuffling is not moving several squares, but shuffling within a smaller space. That will be a disadvantage if the opponent has Great Cleave or WWA –as the FAQ makes clear!



Much easier to envision the images passing through one another, and the caster, repeatedly, as the spell indicates.



The same way he hits only one target under any other scenario. An attack hits a single target.



Well, it IS magic, isn’t it? We also abstract a lot of attack mechanics, and that’s part of the explanation as well.



I don’t. I read “the images separate and continue to move around and through you always staying within 5’”. I then read that consistent with the view that the person is not a 5’x5’ mass, but stands within the square, generally moving around, possibly reaching into any of the eight adjacent squares, on an ongoing basis, rather than standing stock still between actions.

A lot of your other points are irrelevant if we follow the FAQ. You then scientifically analyze spell issues, followed by suggesting we not do so. And at the end, we agree that we should



To me, the FAQ indicates the way the spell is supposed to work. Done?



So how does the spell know whether the images should touch the ground, float above it or be buried halfway or all the way in it? Seems like all the images are supposed to look like they could be the caster, which not standing on the ground may screw up.

They can be anywhere as long as they are within 5 feet of the caster or another image. Lets say the caster is floating 10 feet off the ground, he could have 2 below him, above him. All of them above him, or anywhere he chooses. Their locations are determined upon casting of the spell. You could have one half in an object if you wanted. After all it's only a figment. But you'd still potentially think it's a caster and have to roll the chance to swing at it. From the attackers perspective just seeing multiple of anyone is disorienting in and of itself. You ever go back to the TV section at Wally world and try to focus on one screen when all 30 of them have the same picture on them???

 

[pemerton]

Why would it be a waste? Because it takes a full round action to search one five foot square.
That will tell you if that one image is real or not, nothing more. If you're searching the wrong square you only learn about that one square. Unless, of course, the image appears to avoid your groping hands, just as the real caster would. If that happens, you learn nothing.

Alternately, if all the images are in one square, constantly shifting and switching places, all you've done is verify something you already knew, that one of those images is the caster. But unless you've grabbed hold (something you can't do as part of a Search), you've got nothing. The image you identified can get lost in the shuffle.

<snip>

And whatever you do manage to learn becomes obsolete next round.

When you search, you can identify which of the several visible wizards is leaving footprints. Or which one has a scent. Or which one is breathing. (Or are we now ruling that the spell also creates illustory footprints, and has olfactory and tacticle components despite neither being mentioned in its description?)

Once you have found the real wizard, you can tell your friends when to attack. If you are adjacent (or have the appropriate threatening reach), you can also take OAs against the real wizard if s/he casts a spell.

As for "getting lost in the shuffle", there is no shuffle until the wizard's next turn, if s/he moves: "While moving, you can merge with and split off from figments so that enemies who have learned which image is real are again confounded." If the wizard hasn't moved since your search (eg because s/he hasn't had a turn yet, or is trapped in a web, or any other of the myriad possible reasons) then nothing about the spell suggests you lose your information about which is the real wizard.

 

[GaimMastr]

When you search, you can identify which of the several visible wizards is leaving footprints. Or which one has a scent. Or which one is breathing. (Or are we now ruling that the spell also creates illustory footprints, and has olfactory and tacticle components despite neither being mentioned in its description?)

Once you have found the real wizard, you can tell your friends when to attack. If you are adjacent (or have the appropriate threatening reach), you can also take OAs against the real wizard if s/he casts a spell.

As for "getting lost in the shuffle", there is no shuffle until the wizard's next turn, if s/he moves: "While moving, you can merge with and split off from figments so that enemies who have learned which image is real are again confounded." If the wizard hasn't moved since your search (eg because s/he hasn't had a turn yet, or is trapped in a web, or any other of the myriad possible reasons) then nothing about the spell suggests you lose your information about which is the real wizard.

Jut because you have information in one tun doesn't mean that information is relevant 6 seconds later. By assuming that a Character would assume the same way you would in a meta-game sense would be poor roleplaying depending on the PC's abilities.

We should refer to the description of the school of the spell and any relevant sub-school information when making decisions like this.

[h=4]Illusion[/h]Illusion spells deceive the senses or minds of others. They cause people to see things that are not there, not see things that are there, hear phantom noises, or remember things that never happened.
[h=5]Figment[/h]A figment spell creates a false sensation. Those who perceive the figment perceive the same thing, not their own slightly different versions of the figment. (It is not a personalized mental impression.) Figments cannot make something seem to be something else. A figment that includes audible effects cannot duplicate intelligible speech unless the spell description specifically says it can. If intelligible speech is possible, it must be in a language you can speak. If you try to duplicate a language you cannot speak, the image produces gibberish. Likewise, you cannot make a visual copy of something unless you know what it looks like.


So it is reasonable to rule that an image is creating it's own "smell".

 

[pemerton]

Jut because you have information in one tun doesn't mean that information is relevant 6 seconds later. By assuming that a Character would assume the same way you would in a meta-game sense would be poor roleplaying depending on the PC's abilities.

<snip>

So it is reasonable to rule that an image is creating it's own "smell".

I don't see how either of these things is remotely reasonable.

The spell says nothing about olfactory deception. In fact it refers repeatedly to "images", mentions that if you're blinded you can't be fooled, and has a special clause to deal with hearing. Nothing about smell at all.

As for your suggestion that it would be improper metagaming to tell my friends whose turns fall between mine and the wizards which "image" to strike - why is that metagaming? That's just playing the game in accordance with its turn rules and action economy. It's not different from me casting Entanglement on my turn, and then all my friends getting to belt up on the DEX-penalised target until it gets to take a turn and move outside the area of effect.

 

[GaimMastr]

That's what a DM is for...

I don't see how either of these things is remotely reasonable.

The spell says nothing about olfactory deception. In fact it refers repeatedly to "images", mentions that if you're blinded you can't be fooled, and has a special clause to deal with hearing. Nothing about smell at all.

As for your suggestion that it would be improper metagaming to tell my friends whose turns fall between mine and the wizards which "image" to strike - why is that metagaming? That's just playing the game in accordance with its turn rules and action economy. It's not different from me casting Entanglement on my turn, and then all my friends getting to belt up on the DEX-penalised target until it gets to take a turn and move outside the area of effect.

Likewise the spell does not exclude it either. Devils advocate. I'm just saying that I can see it going either way depending on the group.

When a rule fails to quantify a quandry you look at it's "parent" rules, much like when you use an ability check in place of a skill check. In this case we look at the definition of what an illusion is and it is a school of spells that "deceive the senses". That would be sight, sound, taste, touch, smell, and thought. We can further refine it to the "Figment" sub-school if necessary. The whole point of this sub-argument is to prove why and how smell can play a part. If you were to close your eyes then you are unaffected by the spell, this includes smell. But if you leave your eyes open you are subject to it.

 

[N'raac]

I'm going to do two things here.

First, I'll note that the FAQ says "it's simplest to assume" that they're all in one square. In stating that that's the simplest assumption, it makes it clear that it's an assumption, not necessarily the rule. Like the random dice roll, it's just the easiest way to handle it, and easy is fine.

There are lots of times when we've seen the Sage and the FAQ give answers that run counter to the written rules. I see this as one of them.

Well and good, and my comments do note this is a recommended assumption, not a rule as written. The absence of rulings for many issues which do not arise when we assume the images cluster in the 25 or so square feet surrounding the caster implies to me that this is the intention of the RAW, however.

The figments “stay near you”, which is undefined but doesn’t strike me as 15’ or 20’ away. There is no indication of how to determine where, exactly, they appear or shuffle to, how and where they move, or how rapidly they can move. A random roll makes no sense if I am only in striking distance of three out of nine images (including the real wizard). All of these gaps make it difficult for me to conclude the intent of the spell’s designer was that the images spread out, each in its own square, and rush back and forth whenever the Wizard takes a move action of a 5’ step.

Also, not directed at the images’ location, if “observers can’t use vision or hearing to tell which one is you”, searching one or more squares to differentiate images from caster seems to violate that specific RAW as most PC’s rely on vision and hearing to search.

The second thing I'm going to do is something I've done before: Concede that you are free to treat the FAQ as gospel, and run your game any way you like. I trust that you'll extend me the sme courtesy, and accdept that I'm free to read the RAW as written, treat that as gospel, and run my games the way I like.

You can treat anything you want as the rules and run the game in accordance with that. However, I do not read the spell description as you do, so I do not consider your interpretation to be RAW. It is simply your interpretation.

As a final note, by the RAW the images aren't constantly passing through and switching places with the caster. That only takes place when the caster moves. i.e. on his turn. So while you keep saying that, it's not part of the spell.

It does not say “when he takes a move action or a 5’ step”, nor does it state that an attacker can know the location of the actual target until the caster’s next turn. I can see that interpretation being made. I can also see the term “moving” being its actual meaning, and my view is that characters are always moving about in combat, and do not freeze in place between actions. Creatures with STR 0 or DEX 0 “cannot move at all”. Moving is not a defined term – “move action” is, as is 5’ step. That, to me, is more consistent with the inability to determine the caster from the image. A somatic component is a movement of the hand, so a character casting such a spell must be “moving” despite not travelling any distance.

They can be anywhere as long as they are within 5 feet of the caster or another image. Lets say the caster is floating 10 feet off the ground, he could have 2 below him, above him. All of them above him, or anywhere he chooses. Their locations are determined upon casting of the spell. You could have one half in an object if you wanted. After all it's only a figment. But you'd still potentially think it's a caster and have to roll the chance to swing at it. From the attackers perspective just seeing multiple of anyone is disorienting in and of itself. You ever go back to the TV section at Wally world and try to focus on one screen when all 30 of them have the same picture on them???

Emphasis added. How do we determine the caster chooses, and what are the parameters for that choice? I’m inclined to agree with your approach, if we accept the images spread out to different squares at all. The spell is silent on who controls the movement of the images, or how their movement is determined, so caster control makes the most sense. But it seems odd he can control the movements of up to 8 images with no actions when figments like Silent Image require ongoing concentration.

If we accept the images may fly, be embedded in objects, etc. then I think the attacker needs to get to choose which to attack (whether the flying ones as he thinks that’s where the Wizard would be, or the ones not embedded in the ground since the wizard has only cast one spell since we attacked), rather than being guided by random roll.

That random roll still seems wrong to me if the wizard calls up the images – don’t I know where he was when he called them up? If one or more of the images end up further away than I think he could have moved after casting the spell (or if he had to back off to avoid an AoO for casting it), can’t I reasonably choose to ignore those images?

Fanning them out also means they should disappear pretty quick if the wizard starts casting – each of them mimics his movements, so doesn’t each one attract an AoO from anyone in proximity to that image?

Overall, it seems unlikely, at least to me , that the many obvious questions which arise if each image occupies its own square would not be addressed by the spell description. It seems far more likely, and consistent with the FAQ, that they were never intended to spread out so broadly.

 

[Jackinthegreen]

You can treat anything you want as the rules and run the game in accordance with that. However, I do not read the spell description as you do, so I do not consider your interpretation to be RAW. It is simply your interpretation.

And why do you feel that your interpretation should be the only one used? Because your own experiences, opinions, and interpretations trump others'? Such hubris!

If there can be multiple different, but possibly correct, interpretations of a rule then saying one of them is "RAW" is meaningless because the others could very well be "RAW" too. Then again, I often find any kind of RAW discussion is meaningless. Effectively, what RAW really means is "The Rules As I Interpret Them And You Can't Prove I'm Wrong, Nyeah." (Taken from The Ten Commandments of Practical Optimization.) The game can't be played strictly RAW because it was meant for humans with common sense to play, not robot rules lawyers. Do you also rule that monks have a -4 on attack rolls with unarmed strikes because by RAW they aren't proficient in them? I should hope not!

What is the correct way of playing this? The way that works best for the table, which might not be the way that's strictly presented in the book if the interpretation of the way the book presents it can even be agreed upon in the first place. How do we figure that out? By looking at it from different points of view, figuring out what the pros and cons are of each, and then offering those possibilities to the group to see what works best for them if the group really feels it's worth going over.