The Talos is back, baby, and he's got some brand new tricks.

The standout among them, to my mind at least, is the chainflail. No longer need your Talos be stranded at the floor of his considerable combat potential by a poorly timed '1', for you can roll two dice and take the best!

Either that, or you can equip the beast with an additional close combat weapon, for a bonus attack. Or hey, why not both?

Upgrade	None	CF	CCW	CF+CCW
Exp Attacks	3.5	4.472	4.5	5.472
Attacks/pt	0.0350	0.0407	0.0391	0.0438
Charge EA	4.5	5.472	5.5	6.472
CAtt/pt	0.0450	0.0497	0.0478	0.0518

It turns out that the most effective is also the most efficient, in that combining the two options gives more attacks per point than the other, and both upgrade options provide more attacks per point than the un-upgraded golem. The chainflail edges the extra close combat weapon out slightly, and although its expected value (mean) is very slightly lower, its median result is higher (5 vs 4.5).

Personally, I favour the chainflail, as it provides better insurance against rolling 1, 2 or 3 attacks, but if I had the points I would take both. For the cost, I would always make room for at least the chainflails. I really hate rolling just one attack!

There is also the ichor injector, for some insta-death lovin'. If you find your Talos up against a lot of Nobz, tyranid warriors or Thunderwolf cavalry, this little toy will let you chew through them quicker than your dear opponent can reach for a barf bag. On the other hand, its pretty useless against most everything else, so unless you know your foe in advance I would prefer the chainflails. If you do take it, upgrade with the extra ccw so that you won't get stuck on one attack.

And aside from the hand-to-claw upgrades, there are a trio of gun options available. Personally I don't dig the pods; although the blast is nice, and the S5 nice, I can't see it doing anything that the splinter cannons don't already. The liquifier could come in handy if hordes are foreseen, as being a monstrous creature he can fire both the cannons and the liquifier prior to charging. Again, take the extra ccw, as the liquifier takes the chain-flail's slot.

Finally, there are two antitank options, in the heat lance and haywire blaster. As our metal friend has lost the dinky little rule restricting him to one hit against vehicles, he has suddenly become quite a beastly anti-tank assault unit. One of these anti-tank weapons on top will let him harass enemy vehicles from a distance, and give him a good shot at immobilising a vehicle prior to the assault. In which case it becomes very easy meat.

As may have become clear, I have been a long time fan of the Talos, and almost always make room for him in my heavy support slots. Do you use a Talos? If so, which new tricks will you be taking?

Djinni-ous

Previously I looked at the slightly mad idea of taking the Djinn Blade, and started wondering how it stacked up against the other common weapons. Now, I'll have a look at how much flailing needs to happen before an opposing independent character can be taken out of action, thus charging the soul trap and transforming the limp rhubarb of S3 to the brutally sharp knife of S6.

So what is the probability of taking out an independent character or monstrous creature with repeated application of limp rhubarb?

The table below summarises the probability of inflicting sufficient unsaved wounds to remove a character, of 2 or 3 starting wounds.



Clearly the Husk Blade is the best at nailing multiwound characters, since this is what it was designed for. Of interest is that the Djinn blade has a higher probability of doing enough damage to kill compared to the agoniser, against T4, and not against T5+. Again, the purpose of the agoniser is to reliably wound high toughness opponents, so this is no surprise. It suggests that there is a narrow window around T3/T4 where the Djinn blade is optimal against Rank and File, and serviceable against independent characters. Notwithstanding the concomitant risk.

For the curious, even though the Djinn Blade has a lower expected return against T4 opponents than the agoniser, it has a higher chance to inflict 2+ wounds than the agoniser in the same circumstances. It also has a higher chance to bomb out completely, hence the lower expected return.

Of course, the comparison is hardly flattering to the imprisoned genie. An agoniser is generally better against a broad range of opponents, and a Husk Blade is much better at offing independent characters and monstrous creatures. Why would you bother with trapping your foes essence in a blade of not so much utility?

	Empowered
	Djinn Blade	Husk Blade
Drug	vs T <>	vs T <>
Fleet/+1WS	4.26	3.33
+1S	4.26	3.33
+1A	4.81	3.89
reroll wounds	4.97	3.89
Fcharge	4.26	3.33

That's why. An expected return of over 4 wounds per charge round, against any foe with T<5, or <6 if furious charge or +1S drugs are enabled.

Of course, if you are willing to spend a few more points, there is the third way. Equip with your choice of agoniser or Husk blade, and take the Djinn Blade along for the ride. As it stands, it is a wargear item, meaning that you can equip one other special weapon and still have a pistol for the bonus attack. Use the other weapon to charge the soul trap, then switch to the Djinn Blade to start reaping the other ranks.

And if you do bet the Devil your head, hope your shadowfield pulls through!

Djinni in a Bottle?

I'm sure I'm not the only budding Archon maniacal enough to consider combining the only item that lets you attack yourself, with the only item that lets you boost your strength past the instant death threshold. Am I?

The Soul Trap is already pretty well known, coming as it does on the stock Archon model. All you have to do is nail an independent character or monstrous ceature, capture its soul, and voila! Double strength. This lends itself naturally to being paired with the Husk Blade, also on the out-of-the-box Archon, to insta-death those pesky targets in one hit.

But, the psychotic within me is drawn to one of the seeming red-headed stepchilds of the wargear list, the Djinn blade. Oh the potential, an extra two power weapon attacks, at S6 with the trapped soul of an enemy poseur! Hell, throw in a pair of pain tokens and a favourable combat drug result, and suddenly its 8 attacks with a S8 I8 power weapon, also known as dead sexy.

Of course, there are a couple of hoops to climb before that combo could be engaged. Even the drug result will only happen one game in six, although 9 S7 I8 attacks or 8 S7 I8 reroll wounds attacks are perfectly fine.

The real kicker, though, comes in the Djinn Blade's rebellious genie, who will stab your Archon instead of his foes if given the chance! Then, all of that extra strength will tragically backfire, instantly ending the reign of terror in a manner that most of his foes would find highly amusing.

So is it worth the risk?

Firstly, what are the chances of skewering oneself upon the soul imbued monstrosity? Any double in the two extra attacks will trigger a loss of control, causing both of them to hit the wielder. Straight out, this is a 1/6 chance to wear two of your own attacks, although there is presumably nothing to stop the remaining six of your strikes from slaying your foe. Particularly galling, this is more than likely (2/3) to convert two of your own hits into two hits of yourself. Of course, if you are still swinging at base S3, this may not be a huge problem, especially if your shadowfield is intact. On the other hand, if you are still at S3, will you be able to scratch the enemy?

To calculate the utility of the Djinn blade, we can subtract the chances to autoflagellate from the overall chances to hit, remembering that 2/3 of these would otherwise have hit the foe. This leaves a 1/3 chance that both extra attacks will hit, a 4/9 chance that exactly one of the two attacks will hit, and a 2/9 chance that you will either fail to hit, or hit the wrong target. The expected return per combat phase is therefore 10/9ths of an extra hit on your enemy, and a 1/3rd of a hit on yours truly.

Now that this is known, comparisons may be drawn with some of the other weapons. The agoniser is the same cost, so beginning there, we can immediately see that the Djinn Blade will outperform it against T3 opponents, by 10/9ths of a hit, or 5/9ths of a wound (I am assuming a 3+ to hit, so adjust the maths if facing phoenix lords, bloodthirsters, avatars, daemon princes or hive tyrants). Equally, the agoniser will outperform the Djinn Blade against anything tougher.

Combat drugs may come partially to the rescue, as a +1S or reroll to wound is very handy. Likewise, after a couple of pain tokens, furious charge is a nifty boost.

Unlike the Husk Blade however, a Djinn Blade needs to do more than one wound to get anything from its sister wargear, the soul trap. The condition of having low S is not such a problem when only one wound is required; although the expected number of wounds drops sharply at T5 and six, one could still reasonably expect to deliver at least one wound per two combat phases. Presuming the target is a monstrous creature that will activate the Soul Trap, the Archon will then be free to resume reaving, this time at the much healthier S6.

The table below lists the expected output of wounds, assuming that the Archon gets the charge, and opposes enemies of WS < 7.


Next installment, I'll look at how many hits from each weapon are needed, on average, to knock off a variety of characters.

Haywire!

Nothing warms my black and twisted Dark Eldar heart more than basking in the flames from a destroyed enemy tank. However, since the Darklight Manufacturers Incorporated unionised, dark lances are neither as cheap nor as widely available as they used to be. And while I am hopeful that that means I will no longer be shipped dodgy dark lance knock-offs that bounce off armoured vehicles like summer rain, I feel I must investigate other options.

Among these, is the option of cheap haywire grenades that have suddenly become available to wyches. At half their previous cost they are suddenly a lot more attractive. But how effective are they?

The following table reflects the probability of achieving one or more of four levels of damage: Shaken or better, damaged (weapon destroyed or better), M-kill (immobilised or better), and the self-explanatory kill. Each figure is calculated by subtracting the probability of scoring 0 of these outcomes from 1, and therefore gives the chance of scoring at least the indicated result. The chance of failing entirely is therefore 1 - Shaken or better.

Results are illustrated against immobile vehicles, vehicles that travelled at combat speed (or immobilised dreadnoughts), and vehicles that travelled at cruising speed or more (and mobile walkers).


It can immediately be seen that chances of scoring a kill are relatively poor. The haywire grenade only penetrates on a 6, so this is not really its role. Instead, I prefer to look at the chance for damaging, as it will take only a couple of these results to cripple most enemy vehicles.

Immobile vehicles are easily pumped. Even 5 wyches with haywire have a 66% chance of immobilising permanently a vehicle that had the temerity to stand still in its last movement phase, and a squad of 10 even has a 45% chance of killing it outright. The astute will also recognise that the same fate may befall any other vehicle successfully immobilised by haywire attack. If the wyches cannot be cleared before the enemy assault phase, a free haywire attack will very likely be the end of the metal box in question.

Against vehicles moving at combat speed the numbers are a bit harder, but 5 wyches still have a 40% chance of scoring an immobilised result. This rises above 50% at 7 wyches, but does not reach 66% until 11 wyches are used. As this is too many to transport, I haven't bothered to reproduce that portion of the table.

Vehicles moving more than 6" are generally fairly safe from assault, and sadly haywire grenades are no exception. The best that could generally be hoped for here would be a shaken result - 5 wyches will still shake a fast moving vehicle 52% of the time, though they will cause lasting damage half as often.

By contrast, a ravager with 3 dark lances has a 45.9% chance of scoring an immobilisation or better, against any vehicle with AV12+. Albeit at any speed, from 36" away.

So, although I am not advocating taking haywire wych wenches in place of ravagers, I feel they have a place in the overall scheme of tank hunting.

But if they are taken, how many?

By dividing the probability of any given result with the points cost required to achieve it, we can gain an estimate of efficiency. The following table shows this efficiency for units of haywire wyches with between 5 - 10 members:


5 is the most efficient in all cases, with steep declines in shaking efficiency, but quite slow declines in M-kill and kill efficiency. As these latter two are generally more important than shaking vehicles, there is considerable leeway. Is 5 wyches enough? Efficiency is useless if your unit cannot be expected to perform its role. A unit of 5 performs well against stationary targets, middling against slow moving targets. A unit of 7 performs well against moving targets, but is overkill against stationary targets.

For me, I think it will be 5 wyches in a Venom. The lower profile (when it comes out) may help them fly under the radar, though they will cough up an easy kill point, and provide little assault support. I tend to play smaller point value games, and don't think I can spare a full unit of 10.

But let your own dark heart be the quide here. Whatever you choose, may your enemy's prize tanks roast like chestnuts!

Wych weapons 2: Razorflail and Hydragauntlets

If defensive weapons are your thing, then you've already decided on taking shardnets (or playing Nurgle). If you're tossed up between the two offensive choices of wych weapon, come with me as I do some analysis.

Firstly, what do they do?

The razorflail, inspired by Ivy's blade from the Soul Calibur series, is a combination of sword and whip. In game terms it provides a re-roll to both hit and wound - pretty tasty, and a benefit that sharply increases with the number of attacks made.

The hydraguantlets are spiky cestus-like guantlets (duh) that bear a passing resemblance to Voldo, of the same Soul Calibur series as Ivy. Perhaps when the Grey Knights are redone they will feature a Hadoken attack? Anyway, hydragauntlets provide a bonus +d6 attacks, rather than a bonus +1attack for a pair of weapons.

Naturally, if the weapons could be combined on one model, they would epitomise awesome. Naturally, they can not be combined. So which is better?

I have looked at the expected return for each weapon, against a variety of infantry, and the results are below. By expected return, I mean the fraction that is the product of the probability of killing an enemy, and the number of attacks made. For the hydragauntlets, I have assumed an average of +3.5 bonus attacks in all cases, as that is the mean result of many d6 rolls - hence it is the expected value of the d6, in so much as it contributes to the models overall number of attacks.

I have also looked at the effect of combat drugs. There are four possibilities that directly affect the chances of scoring a kill, and two that do not (these two are collectively termed no drug). Finally, I have considered the effects of charging (let's be frank, wyches should always be doing the charging), and also the furious charge, should you score a couple of pain tokens.

Forgive my shonky coding, I'm new to this whole web thing.

Firstly, against Guard or equivalent (most effective option in bold)

vs GEQ	Wych	Razorflail	Hydraknife
No charge
No drug	0.444	0.889	1.000
+1WS	0.444	0.889	1.000
+1 S	0.593	1.053	1.333
reroll W	0.667	0.889	1.500
+1A	0.667	1.333	1.222

Charge
No drug	0.667	1.333	1.222
+1WS	0.667	1.333	1.222
+1 S	0.889	1.580	1.630
reroll W	1.000	1.333	1.833
+1A	0.889	1.778	1.444

Furious Charge
No drug	0.889	1.580	1.630
+1WS	0.889	1.580	1.630
+1 S	1.111	1.728	2.037
reroll W	1.185	1.580	2.173
+1A	1.185	2.107	1.926

In most cases, the hydragauntlets are superior to the razorflail. Because it is not too challenging to either hit or wound a GEQ, the rerolls conferred by the razorflail are of generally poorer utility than the extra attacks provided by the hydra. Incidentally, it can be seen that rerolls to wound and +1A are clearly the drug of choice for GEQ slaying; if a squad of 8 normal wyches and 2 hydras charged, rerolling wounds, the expected death toll is 11.66. Look out, blob squad; look out wyches! If you blow through too easily, that Hellhound will cook you!

Anyway, onto something somewhat more challenging - Orks.

vs Ork	Wych	Razorflail	Hydraknife
No charge
No drug	0.278	0.694	0.625
+1WS	0.370	0.823	0.833
+1 S	0.417	0.938	0.938
reroll W	0.463	0.694	1.042
+1A	0.417	1.042	0.764

Charge
No drug	0.417	1.042	0.764
+1WS	0.556	1.235	1.019
+1 S	0.625	1.406	1.146
reroll W	0.694	1.042	1.273
+1A	0.556	1.389	0.903

Furious C
No drug	0.625	1.406	1.146
+1WS	0.833	1.667	1.528
+1 S	0.833	1.667	1.528
reroll W	0.938	1.406	1.719
+1A	0.833	1.875	1.354

Against the T4, the razorflail's reroll to wound comes into its own. Naturally, it gains nothing when drugs provide the same alternative - in these cases the hydras tend to trump the rerolls to hit, as you would expect. For Marines and equivalent, we merely adjust these figures for the better save, which doesn't alter the relative utility. However, if you want to plan for expected casualties vs MEQs, here goes:

vs MEQ	Wych	Razorflail	Hydraknife
No charge
No drug	0.111	0.278	0.250
+1WS	0.148	0.329	0.333
+1 S	0.167	0.375	0.375
reroll W	0.185	0.278	0.417
+1A	0.167	0.417	0.306

Charge
No drug	0.167	0.417	0.306
+1WS	0.222	0.494	0.407
+1 S	0.250	0.563	0.458
reroll W	0.278	0.417	0.509
+1A	0.222	0.556	0.361

Furious Charge
No drug	0.250	0.563	0.458
+1WS	0.333	0.667	0.611
+1 S	0.333	0.667	0.611
reroll W	0.375	0.563	0.688
+1A	0.333	0.750	0.542

As an aside, it is worth noting that in most instances, the expected return of a razorflail or hydra is around double that of a normal wych; as the cost is double, the weapon option is worth taking in instances where the return is more than double. I find this to be the case often enough.

In summary, razorflails are generally better than hydragauntlets against T4+, and hydragauntlets generally do more damage against T3-. Keeping in mind that more damage against very shooty armies like Guard may actually work to your detriment, if the wyches wipe their enemies in one turn only to get nuked, then I must recommend the razorflail > hydragauntlet.

Especially if you want your own little Ivy :)

Wych weapons 1: Shardnet and Impaler

Which weapon? (sorry)

Wych weapons have moved from being bulk to specialist equipment; so instead of upgrading all of the models in the squad to have some sharp hooks secreted about their person, a couple of select wyches may take their pick from three options.

On the bright side, all three are kinda neat. The downside for me is that I rather liked the look of all of the old wych weapons, and so my wyches had almost nothing but! Still, since we must adapt or die, the question now becomes: which should I take?

Shardnet and impaler

Two of the wych weapon options are purely offensive; the shardnet is the purely defensive upgrade. Every model adjacent to the wych fights with one less attack than normal. Incidentally, the wording suggests that models only ever lose one attack no matter how many shardnets they are in contact with; a pity, as it would otherwise be quite handy to gimp an enemy independent character completely.

There are two obvious drawbacks with the shardnet. Firstly, it is useless against models with only one attack. Secondly, since the wyches will almost always be striking first, casualties can be removed preferentially from those affected by the nets.

The second drawback can be ameliorated relatively simply, by engaging models that the enemy will not want to remove. This will often further our purposes directly, for example by impairing a veteran sergeant, we might reduce a power fist to one attack only, potentially losing fewer wyches without recourse to our new feel no pain capability.

A bonus of the shardnet, as compared to the old wych weapons, is that there is now no ceiling on whom it can affect. Carnifex? Tangle those claws. Wraithlord? Sorry, that sword won't be swinging through this. Every enemy model loses an attack, regardless of strength, whether they have additional close combat weapons, are a god given form, or just a grot with two pointy sticks.

So should I take one? Taking a shardnet will depend on the intended role of the wych squad in question. For tarpit wyches, it is the only choice. For slaughtering masses of basic troops, perhaps you want one of the others. If you want to limit the effectiveness of enemy upgrade characters, take one shardnet, and one of something else.

But which one? Stay tuned!

Dark Eldar are here, are dark

After thirteen-odd years, the Dark Eldar are back in town!

As my primary 40K army for the last ten years, I've got a lot of love for the little libertines.

Over the next few posts, I'll share some of my thoughts and mathematical analyses of the new codex.

But let me say before I move on - the flavour of this book, and the overall feel, were freaking awesome. I love it, and I hope that you do too!