Which weapons are best?

[Tech-Priest]

So after looking through the codex with so many choices for troops I was thinking about paying the Crowe tax and using purifiers as my troop selection...

 

My next question is how do you think it's best to load out a 10 man squad?

 

With the stat lines the squad is more melée and I think the halberds are the way to go for the bonus to initiative the a hammer for the squad leader...

 

But I do want to have a little flexibility so I'm wondering which guns would be best to take? Is psybolt ammo worth it for the squad?

 

Should I dedicate certain squads more shooty and just take swords for them or is balance the better option?

 

 

The list I'm considering is mainly PAGK (purifiers) a couple dreadknights, a storm raven and if points allow maybe a couple dreadnoughts set up for shootyness

[HERO]

Squads of 10 should take Psybolts.

 

Squads of 5 should take 2x Psycannons and sit in a Rhino.

 

Squads of 10 should take 4x Psycannons for a AT role, or 4x Incinerators for a up-close anti-horde role.

 

Squads of 10 w/ only 2x Psycannons should play like GSS without fear of getting into CC. The same with 2x Incinerators.

 

All Purifiers should have Halberds save for the heavy weapons, and Justicar who should have a MC Daemonhammer.

[Joasht]

Sometimes, I wonder about the number of Psycannons. Psycannons are great and all, but unless you remain stationary (not always the case) you sacrifice your close combat lethality for two (good) shots and you pay points for that. If you buy four Psycannons and a Hammer, this leaves you with five Halberd guys - a measly ten attacks (assuming not charging) at I6 that would only kill 2.5 MEQs on average.

 

I'd think that you may want to purposely use less, like two Psycannons as suggested, for an assault unit. Psybolts would still be a rather decent option as they are cheap for the benefit, but this means this squad would be one zooming forward in a Rhino, ready to beat faces in, as opposed to the four-Psycannon squad that would play more like an "old" GK squad, keeping the enemy at 24" as much as possible, except that in the advent the enemy does reach them, they can still fight back.

 

And there is also the question of the Incinerator. Is it worth splashing on Incinerator in there? For an assault unit it might be worth either buying one or replacing a Psycannon with an Incinerator because against non-vehicle targets, the Incinerator would probably kill more - assuming you get close enough to use it properly.

[Gaz1858]

With the stat lines the squad is more melée and I think the halberds are the way to go for the bonus to initiative the a hammer for the squad leader...

 

This is the way to go in my view, striking at I6 is worth more than an extra attack. Also its cheaper points wise. The DH is always good incase you get in combat with a walker or something.

 

As has been said for a 10 man squad take Psybolts.

 

As for the special weapons options it depends on how you play them. The units in the Stormraven should have Incinerators and the rest Psycannons.

 

Thats how im planning on running it

[Tech-Priest]

Any thoughts on the weapons for the dreadknights plan on taking 2 but if I load them to the gills they get kinda pricey...

 

The Gatling psylencer's 12 shots seems worth it to me and on paper the great sword sounds useful but is it worth the points? And not sure about how the doomfist work does it get the bonuses of dccws? (probably need to wait on the FAQ for that one)

 

Seems to me he is got the oomph to handle quite a bit if I load him up... Soften things up at range and take the fight to the big uglies if needed... But two cost me a good chunk of points if it's not a large game

[Ashe Darke]

Sometimes, I wonder about the number of Psycannons. Psycannons are great and all, but unless you remain stationary (not always the case) you sacrifice your close combat lethality for two (good) shots and you pay points for that. If you buy four Psycannons and a Hammer, this leaves you with five Halberd guys - a measly ten attacks (assuming not charging) at I6 that would only kill 2.5 MEQs on average.

 

I'd think that you may want to purposely use less, like two Psycannons as suggested, for an assault unit. Psybolts would still be a rather decent option as they are cheap for the benefit, but this means this squad would be one zooming forward in a Rhino, ready to beat faces in, as opposed to the four-Psycannon squad that would play more like an "old" GK squad, keeping the enemy at 24" as much as possible, except that in the advent the enemy does reach them, they can still fight back.

 

And there is also the question of the Incinerator. Is it worth splashing on Incinerator in there? For an assault unit it might be worth either buying one or replacing a Psycannon with an Incinerator because against non-vehicle targets, the Incinerator would probably kill more - assuming you get close enough to use it properly.

This is why I'm not running psycannons on my SS. I need all the attacks I can get. In my 1750+ list there is a 10 man Purifier squad with 4 psycannons to sick back and control the mid-field. The way how I see them is a shooty unit that can also do OK in combat should I need them to. They also don't take up a Heavy Slot which has been used up by Dreads.

 

But yeah in a unit of combat orientated purifiers I wouldn't go more than 2.

[Inquisitor Fox]

Think of it this way... a psycannon is an autocannon that can be fired on the move with half range. Ask an Imperial Guard player if they'd take being able to shoot their autocannon at half range on the move instead of not at all. ;)

 

Even on the move, it is an effective anti-infantry and anti-vehicle weapon. Stationary it becomes even better. Psilencers and Incinerators are nice, but my money is still on the psycannon. One Strike Squad of mine will have 2, the other will have 2 Incinerators (mostly because that's the models I have...), and my Terminator units will all be bearing Psycannons, as does my HQ ;)

 

It's fun having basic troops being able to blow up tanks with side and rear shots, as well as holding their ground and being stationary which allows them to reliably threaten a Land Raider with up to 8 rending shots :D

[Gaz1858]

Any thoughts on the weapons for the dreadknights plan on taking 2 but if I load them to the gills they get kinda pricey...

 

I dont think any of the NDK's guns are worth the points. Maybe the Incinerator but thats it. The Psilencer doesnt have an AP and is only S4, the Psycannon can drift and miss. Im taking one with a teleporter and Greatsword.

 

And not sure about how the doomfist work does it get the bonuses of dccws? (probably need to wait on the FAQ for that one)

 

If you have 2 Doomfists you get +1 Attack at S10 with 2d6 armour penetration. I prefer taking a Greatsword. The re-rolls are worth more than the +1A.

[Brother Valerius]

The Gatling psylencer's 12 shots seems worth it to me and on paper the great sword sounds useful but is it worth the points? And not sure about how the doomfist work does it get the bonuses of dccws? (probably need to wait on the FAQ for that one)

 

Depends on your opponent. The intent of the rules is obvious, and a reasonable opponent will give you the bonuses despite the letter of the law being against it. So if they want to argue RAW you don't have any ground to stand on, but it's hard to imagine that people you play against regularly will take that stance.

[Fred Johnson the 3rd jr]

Squads of 10 should take 4x Psycannons for a AT role, or 4x Incinerators for a up-close anti-horde role.

 

What ridiculous hordes do you plan on fighting with 4 incincerators plus Cleansing Flame? :D

4 sounds like complete overkill to me. May as well do a mixed squad of 2 incinerators and 2 psycannons.

[Marshal van Trapp]

i'm planning on running as many models with regular NFW and SB as i can in my crowe'd purifier squads (yes i'm calling them crowe'd) then a 10man DH interceptor squad, and a 10 man terminator squad (which i will break up into 2 combat squads as needed) my heavy weapons will be reserved for my Purgation squad where they belong!

[Aidoneus]

my heavy weapons will be reserved for my Purgation squad where they belong!

Granted, I've only played 1 game with the new 'dex, but if it turns out to be anything like the old 'dex, you'll need all 3 of those slots for anti-tank stuff. Psycannons are nice and all, especially now that they're S7, but they're still useless against AV 13 and 14, and at long range. What you need are Dreadnoughts for anti-tank (ranged or close, depending on loadout), Dreadknights for anti-anything big and tough, and/or Land Raiders for transport as well as anti-tank at range.

 

For the Purifiers, I tend to agree with those who have said that they should keep the special weapons to a minimum if you want them in close combat. Giving up those shiny force weapons, not to mention I6 ones, is not the way to go if you plan to be in combat. A single Incinerator for large hordes is a possibility, but otherwise leave 'em with Storm Bolters. If, on the other hand, you want them to hang back in a shooty role... well, why not just take a Strike Squad?

 

For the Dreadknight, the only ranged weapon I'm particularly impressed with is the Heavy Incinerator. A single large blast Autocannon is only good at killing clumped, lightly-armoured infantry out of cover... and for the few times that will ever come up, that's what Storm Bolters are for. And the Psilencer, even the Gattling version, suffers from the exact same problem, only it's piteously underpowered as well. The Heavy Incinerator is nice because it negates cover at range, which means you can fling it at troops hunkered down in cover, even if they're screened by other units. Really though, the Dreadknight doesn't actually need a ranged weapon, as most of the time it'll be busy smashing something's face in up close.

[Inquisitor Fox]

Granted, I've only played 1 game with the new 'dex, but if it turns out to be anything like the old 'dex, you'll need all 3 of those slots for anti-tank stuff. Psycannons are nice and all, especially now that they're S7, but they're still useless against AV 13 and 14, and at long range. What you need are Dreadnoughts for anti-tank (ranged or close, depending on loadout), Dreadknights for anti-anything big and tough, and/or Land Raiders for transport as well as anti-tank at range.

The new Psycannon gets double the shots when stationary, but more importantly becomes rending. This combined with it's high shot output make it actually more likely to affect a Land Raider than a Lascannon :)

 

We've got the anti-tank, it's the range of more than a Stormbolter that's the issue.

[Brother Valerius]

my heavy weapons will be reserved for my Purgation squad where they belong!

Psycannons are nice and all, especially now that they're S7, but they're still useless against AV 13 and 14, and at long range.

 

You'd think so, but no. A single psycannon firing in the heavy mode has a 68.36% chance to pen AV14 at least once, compared to the 58.33% chance a melta weapon has. And a psycannon can do it from 24", whereas melta can do it from 12" at best. Granted, a psycannon isn't AP1, and has limited range, but useless it's not.

 

Edit: ninja'd!

[Aidoneus]

You'd think so, but no. A single psycannon firing in the heavy mode has a 68.36% chance to pen AV14 at least once, compared to the 58.33% chance a melta weapon has. And a psycannon can do it from 24", whereas melta can do it from 12" at best. Granted, a psycannon isn't AP1, and has limited range, but useless it's not.

Not sure I quite understand your math:

 

Psycannon: 4 shots, BS 4, need to rend, anything but a 1 on the d3...

 

4*(2/3)*(1/6)*(2/3)=(8/27)= 29.630% chance to pen AV14 (Edit: Actually, you average .29630 pens per turn)

 

Meltagun: 1 shot, BS 4, need at least 7 on 2d6...

 

1*(2/3)*(21/36)=(21/54)= 38.889% chance to pen AV14 (Edit: Actually, you average .38889 pens per turn)

 

When you factor in chance to kill, these change to .09877 and .19444 kills per turn respectively. The meltagun is therefore much better at destroying AV14 vehicles. That is, assuming I did the math right. It's entirely possible that I messed something up. Can someone double check me on this?

 

 

Also, keep in mind that having the chance to shoot the Psycannon as a 24" heavy weapon is hard to do. Even the 24" mobile version on terminators, or similarly-ranged multi-melta on dreadnoughts, is way easier to ensure a shot with. Granted, you'll have multiple units with psycannons, so your chances aren't all that bad that you'll get the shot off. But when you consider how much more effective those squads would be shooting all those lovely storm bolters at infantry, or charging into close combat with all those force weapons, it seems a shame to stand still and shoot a tank. That's the real reason you ought to allocate the heavy anti-tank duties to your HS; it frees up your infantry to do what they do best; mill enemy infantry into the dirt.

[Brother Valerius]

You're doing your probability math wrong (definitely in one respect, and I'm pretty sure in another). First, you can't multiply a probability by 4 to get the probability of 4 shots. You raise (1-probability) to the fourth power. 1-probability is the chance you miss, and raised to the fourth power it is the chance you missed every time (or did not ever pen, in this case). Then subtract this value from 1. That is the probability you will have hit (penned) at least once. That is one part you're definitely doing wrong.

 

The other part, which I'm a little less certain of but still fairly certain, is that you're factoring rending in wrong. The probability of something is (number of events with desired result)/(number of events in the possibility space). By doing the 2/3 chance like you are, you're treating it as if it happens every time which skews the result (multiplying probabilities represents the probability of independent events, when in fact the 2/3 roll is dependent on the result of the 1/6 roll). There are 8 events in the space, all equally likely: 1,2,3,4,5,6+1,6+2,6+3. 6+2 and 6+3 are the magic numbers, so there's a 25% chance to get that result.

 

I probably did err in not including chance to hit, since that probability hits the psycannon harder since it's firing 4 shots. Even so, it's still better chance to pen than melta: 38.89% chance to pen with melta, 51.77% chance to pen with a psycannon.

 

You also erred with factoring in chance to kill, similar to the way you did with the rend. I made a probability table (there's probably a much easier way, and I probably learned it in combinatorics class, but that was a long time ago :lol:), and the chance to kill AV14 is 31.48% with melta, 20.44% with a psycannon firing at 4 shots. It appears that the AP1 goes further than I initially thought.

 

You (correctly) point out that it's hard to get the chance to fire the psycannon in heavy mode, but as a multi-melta is heavy and has to be within 12" (and a meltagun has to be within 6"!), I think the two weapons are evenly matched in that respect.

[Aidoneus]

I wasn't calculating the probability of at least one pen, it's true. I therefore should not have expressed the final results as percentages. I was calculating the average number of pens per turn, which should actually be a larger number. For that purpose, simply multiplying by 4 is correct, since you have 4 independent opportunities to pen each turn.

 

Even if you wanted to calculate the probability of getting at least one pen, I still don't think exponents would come into play. The easiest way to do that is to take each shot individually and calculate the probability that it does not pen. You then add those 4 probabilities together, and subtract that sum from 1. Adding is proper here because you need all 4 shots to fail to pen.

 

Rending is also done properly, because the initial d6 and subsequent d3 are indeed independent. First you need to roll the initial 6, which happens 1/6 times. Then, and only then, do you need a 2+ on the d3. But you only get to roll that d3 1/6 of the time. It's sort of like how you need to roll to hit before you roll to wound. Another way to think about it is that you roll both a d6 and a d3 every time, and pen only when the d6 rolls a 6 and the d3 rolls a 2+.

 

Chance to kill should be correct, because you simply divide the psycannon # by 3 (1/3 of pens kill) and the melta # by 2 (1/2 of melta pens kill).

 

As for the distance argument, keep in mind that melta are nigh-universally mounted in transports or on bikes or vehicles. That means the effective range of a meltagun is generally 18-20" (counting disembarking), and a multi-melta is 30" (dreadnought) to 36" (bike/speeder). Sure, to get half-range you need to subtract 6" from the meltagun and 12" from the multi-melta, and I'll readily admit that that makes things harder. But keep in mind that the melta will be moving every turn, and so can continue to keep up with things moving away from it, whereas psycannons need to stay still every turn to get those 4 shots. And of course, lascannons and psy-autocannons have so much range it's almost never an issue.

[Inquisitor Fox]

And here I just ran numbers the other day on whether a twin-linked lascannon or twin-linked "psycannon" would be better on my Storm Raven... :wub:

 

What I found is that the "psycannon" outdid the lascannon in number of pens as well as number of glances in one "round" of firing (specifically both of them twin linked and at BS4). For the curiuous it was 14.82% chance of a glancing hit with a lascannon, and 14.82% chance of a penetrating hit with a lascannon, versus 19.77% chance of a glancing hit with a "psycannon" and a 39.50% chance of a penetrating hit with a "psycannon" . I then compared this to the TL autocannons with psybolts and a typhoon missile launcher just to get a better idea of what equipping Razorbacks or Dreads would really mean.

 

* The method here was a simple one: # of shots * % chance to hit (including the twin-linked) for number of hits, then multiply that by the .1667 chance for each d6 result (mostly... the rending one had another branch on the probability tree at that point, where it was a .0556 chance to get a AV14 glance on a d6, and a .1111 chance to get an AV14 pen)

 

The long and short of it is this though: In terms of all around weapon, the psycannon is king. It can be fired stationary for full effect, but can still be fired on the move. It can harm Any unit in the game, up to and including a Land Raider, at a fairly decent range. It's threat bubble pairs excellently with stormbolters. Yes, it probably is not as effective at destroying a Land Raider in a single volley as a melta-weapon, but it is largely significant to a melta weapon against infantry. In short, it is the do everything weapon our do everything troopers really need. I don't think you can ever go wrong by including a psycannon or two in every squad if you have the points and another weapon does not fit their intended roll in your forces better.

 

EDIT: I realize that rending and Monoliths don't really play together, but I argue that if you're shooting the Monolith and not the 'Necrons' you're doing it wrong... :whoops:

[BrotherWasted]

Shoot two Psycannons out of a Rhino. There's your four shots at 30" Range.

 

On a rending hit, you are guarantee to glance the land raider. On AV13, you are guaranteed to penetrate when you trigger rending (7+6+1=14).

 

When in doubt, hit the damn thing with Meltabombs, a Dreadnought/Dreadknight or a Daemonhammer.

[Brother Valerius]

I wasn't calculating the probability of at least one pen, it's true. I therefore should not have expressed the final results as percentages. I was calculating the average number of pens per turn, which should actually be a larger number. For that purpose, simply multiplying by 4 is correct, since you have 4 independent opportunities to pen each turn.

 

Even if you wanted to calculate the probability of getting at least one pen, I still don't think exponents would come into play. The easiest way to do that is to take each shot individually and calculate the probability that it does not pen. You then add those 4 probabilities together, and subtract that sum from 1. Adding is proper here because you need all 4 shots to fail to pen.

 

Rending is also done properly, because the initial d6 and subsequent d3 are indeed independent. First you need to roll the initial 6, which happens 1/6 times. Then, and only then, do you need a 2+ on the d3. But you only get to roll that d3 1/6 of the time. It's sort of like how you need to roll to hit before you roll to wound. Another way to think about it is that you roll both a d6 and a d3 every time, and pen only when the d6 rolls a 6 and the d3 rolls a 2+.

 

Chance to kill should be correct, because you simply divide the psycannon # by 3 (1/3 of pens kill) and the melta # by 2 (1/2 of melta pens kill).

 

No, you don't simply add probabilities. You can't. Probabilities have to be multiplied, not added. The probability of flipping a coin heads once in two tries isn't 1, it's 3/4 (you can quickly verify this by flipping a coin twice and having it come up tails both times, which would be impossible if probabilities were added). The probability of failing with all 4 shots isn't x + x + x + x, it's x * x * x * x. No offense, but if you're adding probabilities like you are, you need to re-learn how to do probabilities because that is an extremely incorrect way to do things.

 

The rending is done improperly (after doing some thinking, I realized why) because the events are not independent. You are incorrect in this. The outcome of the 1/6 roll DOES influence the second in that it only happens 1/6 of the time. By calculating it as if you do it every time, you add 10 extra outcomes (1-5 is already one outcome each, so splitting it into 3 dice results in addition creates only two extra) to the possibility space. This drives the fraction of successful events down, even though the outcomes you added to the possibility space are not actual outcomes that can happen.

 

Your chance to kill is also incorrect, for the same reasons (as I already said). You're enlarging the possibility space with outcomes that can never happen, which ruins your probability that you calculate. There's also a minor error in that 1/6 of melta glances kill, which I don't believe you took into account (though it is a very small error).

 

Edit: I apologize if I come off sounding harsh towards you, because that is not at all the intent. I just can't think of a less blunt way to word what I'm trying to say.

[Aidoneus]

No, you don't simply add probabilities. You can't. Probabilities have to be multiplied, not added. The probability of flipping a coin heads once in two tries isn't 1, it's 3/4 (you can quickly verify this by flipping a coin twice and having it come up tails both times, which would be impossible if probabilities were added). The probability of failing with all 4 shots isn't x + x + x + x, it's x * x * x * x. No offense, but if you're adding probabilities like you are, you need to re-learn how to do probabilities because that is an extremely incorrect way to do things.

Urg, you're right. Sort of. You actually do add probabilities sometimes, just not here. You multiply the independent probabilities of failing to pen with each shot, then subtract the product from 1. Going with your coin example, the chance to flip heads at least once in 2 flips is 3/4 because 1-(1/2*1/2)=3/4. So yes, I admit that you were right about how to calculate the chance to pen at least once (at least, on this level. I still disagree about other parts, like rending). Still doesn't affect my original calculation though, since I still correctly figured out the average number of pens per turn with each weapon.

 

The rending is done improperly (after doing some thinking, I realized why) because the events are not independent. You are incorrect in this. The outcome of the 1/6 roll DOES influence the second in that it only happens 1/6 of the time. By calculating it as if you do it every time, you add 10 extra outcomes (1-5 is already one outcome each, so splitting it into 3 dice rolls in addition creates only two extra) to the possibility space. This drives the fraction of successful events down, even though the outcomes you added to the possibility space are not actual outcomes that can happen.

This is just plain wrong. The chance of rolling a 1 on the original d6 is not the same as the chance of rolling a 6 on it and then rolling a 2 on the d3. Isn't that obvious? The chance of rolling the 6 is the same as of rolling the 1, and then after you've gotten the 6, you need the further result of a 2 or 3 on the d3. The dice rolls are completely independent of each other in that the outcome of one does not affect the probabilities on the other, and so multiplication is correct.

 

Edit: For what it's worth, the probability of penning at least 1 time in a given turn with a psycannon is 26.497%. {prob miss+prob hit but don't pen all ^4 subtracted from 1 = 1-((1/3)+(2/3)((5/6)+(1/6)(1/3)))^4 }The probability of penning at least once with a meltagun is the same as the average number of pens per turn with a meltagun (.38889), because it is incapable of getting more than 1 pen per turn. Therefore, if you want to compare probabilities of at least one pen rather than the average number of pens, the meltagun is very slightly further ahead than before.

[Zagman]

Squads of 10 should take Psybolts.

 

Squads of 5 should take 2x Psycannons and sit in a Rhino.

 

Squads of 10 should take 4x Psycannons for a AT role, or 4x Incinerators for a up-close anti-horde role.

 

Squads of 10 w/ only 2x Psycannons should play like GSS without fear of getting into CC. The same with 2x Incinerators.

 

All Purifiers should have Halberds save for the heavy weapons, and Justicar who should have a MC Daemonhammer.

 

Precisely. I completely agree.

[Brother Valerius]

Urg, you're right. Sort of. You actually do add probabilities sometimes, just not here. Still no exponents though. You multiply the independent probabilities of failing to pen with each shot, then subtract the product from 1. Going with your coin example, the chance to flip heads at least once in 2 flips is 3/4 because 1-(1/2*1/2)=3/4. Still doesn't affect my original calculation though, since I still correctly figured out the average number of pens per turn with each weapon.

 

Exponents are certainly used, I'm just simplifying a bit further than you. Let chance to pen be x, and let y be 1-x. That means the chance to fail to pen every time is y*y*y*y, and the chance to pen at least once is 1-y*y*y*y... or 1-y^4, as I originally stated. I'm still a bit unclear as to if you calculated the average thingy right, but I'll leave that one since I honestly cannot recall myself how to do it.

 

This is just plain wrong. The chance of rolling a 1 on the original d6 is not the same as the chance of rolling a 6 on it and then rolling a 2 on the d3. Isn't that obvious? The chance of rolling the 6 is the same as of rolling the 1, and then after you've gotten the 6, you need the further result of a 2 or 3 on the d3. The dice rolls are completely independent of each other in that the outcome of one does not affect the probabilities on the other, and so multiplication is correct.

 

Because the D3 roll is contingent upon getting a roll of 6, the only time the D3 enters the possibility space is when you roll a 6. Multiplying 2/3 by 1/6 introduces the D3 roll into the table for every result, not just 6. The rolls are independent in the sense you suggest, but they are not truly independent because the second roll doesn't always happen. If the second roll always happened, straight multiplication would be proper, but it doesn't so it isn't.

 

Remember, probability is defined as (number of events with desired outcome)/(number of events in the space). Your table of dice roll permutations looks like this: 1-1, 1-2, 1-3, 2-1, 2-2, 2-3... most of those outcomes don't actually exist, which makes your probability seem smaller than it is by inflating the space larger than is possible.

[Gaz1858]

BOOOOOOOOOOOM

 

 

That was the sound of my head exploding

[Resv]

Back to the original subject!

 

The easiest way to approach Purifiers is to think about what you want them to accomplish. Honestly, give it some serious thought! Hero's examples above are not all that great mainly because you are paying far to much for a single unit that die just like any other basic Marine. Instead of buying up every single option under the sun these are point that might be better served buying up more bodies. I do understand that 5 Purifiers can be very effective in close combat HOWEVER they don't last very long at all against shooting. This is why it is nice to have some ablative bodies in the unit to make sure they can get into a position where they can do the most damage.

 

I usually take 10x Purifiers 8x Halberds 2x Psycannons and a Rhino. This makes for a not overly expensive unit that has done very well for me so far. Why only 2x Psycannons? You can only shoot two out of the top of a Rhino, seems logical. Any more than that and they are just going to waste until your transport gets shot out from under you. This also allows for more Halberds which will give your unit a bigger punch in CC which is where Purifiers REALLY shine. The Master Crafted NDH is also a bit much if you ask me. I understand why taking one is useful but what are those attacks going to accomplish? It is awesome that you can bust up armor maybe take on a MC but how often does that happen? Its nice to have the ability but I would much rather save up my points to take another Psy-Rifle Dread or something that can add even more utility to my army.

 

Just about every unit in the new Gray Knight Codex can get really expensive REALLY quickly and Purifiers are no exception. When you start buying up kit like it is going out of style than you might be hamstringing other important parts of your army. This is why I advocate for keeping your Purifiers, as well as all of your units, on the cheap side until you can start tailoring them to make up for the weaker elements in your army.