Warhammer 40K Tabletop XXX: Imperium After Dark

[Eldan]

I'll make this short. General impression: I like it a lot. Certainly best 40k game I've had in years. Rules mostly work, a lot of stuff will take time getting used to (i.e. we both forgot command points existed.)

End result: 7:4 to eldar on points, but eldar almost wiped out.

Match-up wasn't ideal. I feel a lot of my monsters were good at killing big stuff, while he brought almost entirely aspect warriors, a hemlock and an avatar.

In general, higher move on monsters is a godsend. No more carnifexes slogging around for three turns, while being whittled down.

Hive tyrants are hive tyrants, they do what they always did. Except faster and with a ward save. However, tyrant guard are godly. Relatively cheap ablative wounds for your tyrant, even in melee.

Exocrine wasn't as impressive as it might have been, I feel. Shooting it at aspect warriors just feels unsatisfying, but there wasn't much else.

Carnifexes are awesome, but a lot of that was probably rolling. As in, it charged the avatar and trampled it into the mud first attack on the charge, no wounds in return.

Trygon prime feels a bit schizophrenic: shooting attack is good against light infantry (12 shots, medium strength, no AP), while it has what is probably the best melee profile in the army. Though since everything has split fire, I suppose it might work. Having it pop out on command with some hormagaunts was really nice to take objectives early.

Lictors: probably still need to be in groups. This one charged first turn and took two wounds off a farseer, then was slaughtered. If there had been two or three, it would have taken out the enemy's biggest force multiplier.

The toxicrene, sadly, never got to do much. I suspect rerolling to wound rolls might be awesome, except it only ever fought stuff with T3

Termagants: still fun to block about half hte width of the table with bodies and then let off a salvo of 60 S4 shots. About the same as they were.

[Haruspex_Pariah]

I just discovered that there's a shop in my town, less than half an hour's drive, that sells GW products. Gotta get down there personally to see the place, but if I can actually buy stuff without taking an airplane to the next state I might just get back into 40k.

Instead of, you know, theory crafting and lurking. Coincidentally, I've always been partial to Nurgle...

[Renegade Paladin]

We sent out 7th edition in style today, 2000 points Iron Warriors v. Guard.

Spoiler

Show

We rolled Emperor's Will with the predictable result. Chaos wound up winning 5-4 on secondaries (First Blood for me, Slay the Warlord and Linebreaker for him.)

[Blackhawk748]

I love how you have an Imp Guard helmet for terrain.

[Renegade Paladin]

I love how you have an Imp Guard helmet for terrain.

Oh, the dome? That's a miscast part from when I used to work in a plastics factory; I took home a bunch of junked miscasts to make terrain with years ago. That one was made into a Dark Angels-themed bunker; there's a door painted on the other side and a twin-linked assault cannon magnetized to the top.

Edit: I have to say, though, to fill 2k points my first impulse was to take a Baneblade, but the Chaos player really doesn't like superheavies, so I acquiesced and took a boatload of Russes instead.

Then he gave his Daemon Prince 2+/4++ and It Will Not Die. Couldn't scratch him. Should have taken the damn Baneblade.

[Haruspex_Pariah]

Edit: I have to say, though, to fill 2k points my first impulse was to take a Baneblade, but the Chaos player really doesn't like superheavies, so I acquiesced and took a boatload of Russes instead.

Then he gave his Daemon Prince 2+/4++ and It Will Not Die. Couldn't scratch him. Should have taken the damn Baneblade.

That's unfortunate.

At least in 8th Edition, Daemon Princes default to 3+/5++ and only a Chaos Daemons DP of Tzeentch can get the 4++. IWND doesn't seem to even exist anymore. And objections to superheavies won't have as much weight, with how units have been standardized.

[Cheesegear]

HUGE NERD ALERT.

Space Marines (n=74)
Mean Power Rating: 7.57 ± 4.18 Power¹
Mean Points:Power: 14.61 ± 7.31 Points per Power<sup>2

1</sup> Outliers are all Land Raider variants.
² Honour Guard are an outlier with 71 Points per Power, and Servitors, at 2.67 Points per Power*

*Complications; No matter what wargear you give a model/unit, it remains the same Power Rating. So basically everything skews higher.
Assumptions; Power Ratings are reliable. Points costs are not. I assume that GW/play-testers actually have figured out which units are better than others. However, I don't believe that they will ever get the 'by how much' part correct.

e.g.
Servitors (x4); x2 Servo-Arms, x2 Plasma Cannons - 74 Points
Shoots to 24.67 Points per Power, ~10 times what it was previously.

While
Intercessors (x5) - 100 Points
Remains on the curve because Bolt Rifles are free.

Conclusion
If you're a HUGE NERD, playing in a 'Power Rating only' meta, that disregards points, look for units like Servitors that break the curve. Because we have two variables on what 'Points' is - and I'm assuming that one of the variables actually is reliable - we can now mathhammer the **** out of the list-building process, without play-testing.

Yes. I'm aware that I'm totally breaking the 'casual' meta, and that I'm an A-hole. Alternatively, I'm doing a service, where I can tell casuals what the most efficient units are when you're Power, rather than Points.

EDIT:
As far as the models (not including wargear) are concerned, it looks like almost everything is costed at 13-15 Points per Power Rating. Obviously with some outliers.

[Gauntlet]

Conclusion
If you're a HUGE NERD, playing in a 'Power Rating only' meta, that disregards points, look for units like Servitors that break the curve. Because we have two variables on what 'Points' is - and I'm assuming that one of the variables actually is reliable - we can now mathhammer the **** out of the list-building process, without play-testing.

Yes. I'm aware that I'm totally breaking the 'casual' meta, and that I'm an A-hole. Alternatively, I'm doing a service, where I can tell casuals what the most efficient units are when you're Power, rather than Points.

EDIT:
As far as the models (not including wargear) are concerned, it looks like almost everything is costed at 13-15 Points per Power Rating. Obviously with some outliers.

When I was doing my maths, it looked to me like 1 power rating was equal to ~20 points, if you give the units the most expensive upgrades they can have. So the power rating for Servitors assumes you're going to be kitting them out with Plasma Cannons or similar, and the power rating for a Tacitcal Squad assumes they're going to be spending 'free' points on heavy and special weapons, etc.

[Cheesegear]

So the power rating for Servitors assumes you're going to be kitting them out with Plasma Cannons or similar

Then they're bonkers. Servitors are PR3? Servitors with Servo-Arms are PR3 no problem. With Plasma Cannons? You're dreaming.

[Gauntlet]

Then they're bonkers. Servitors are PR3? Servitors with Servo-Arms are PR3 no problem. With Plasma Cannons? You're dreaming.

Tactical Squads are PR3.

Devastator Squads are PR7.

Both units are basically exactly the same, until you give them more equipment.

Clearly Power Rating includes expected equipment, because if it didn't Tacticals and Devastators would have the same PR value (which would clearly be ridiculous). I don't think a Signum and an Armorium Cherub is worth doubling the squad's PR.

I don't think PR3 for two BS4+ Plasma Cannons is that crazy, really. You can get a BS3+ PCannon and a Plasmagun for the same power rating on Tactical Marines, and twice as many Plasma Cannons (on a Devastator Squad) costs 1PR more than two squads of servitors, and you get better BS, Signum and Cherub in exchange for fewer ablative wounds. Seems fine to me.

I don't think Plasma Servitors for 3 Power is above the curve much at all, to be honest.

[Cheesegear]

Tactical Squads are PR3.

What Tactical Squad are you looking at? I see PR5.

[LeSwordfish]

Even so, that still makes sense as compared to the devastators.

[Haruspex_Pariah]

Honour Guard are an outlier with 71 Points per Power, and Servitors, at 2.67 Points per Power*

e.g.
Servitors (x4); x2 Servo-Arms, x2 Plasma Cannons - 74 Points
Shoots to 24.67 Points per Power, ~10 times what it was previously.

How are you getting 2.67 for the Servitors? They come with servo-arms by default, so their minimum spec is 18.67 points per power.

[LeSwordfish]

A quick comparison I wanted to do, because I might want to swap weapons on my chaos lord.

Lightning Claw: Strength User (4), AP -2, Dam 1, re-roll to wound. 9 points.
Powerfist: Strength User x2 (8), AP -3, Dam D3, 20 points. Is it more than twice as good as the claw? What if you take two claws and get +1 attack? Is it worth having one of each?

Hit on the same amount. Aside from vs 5+ saves, 1/6 more wounds are saved by the claw than the fist always.

T	Claw (after saves)	Fist
2	0.97 (0.81)	0.83
3	0.89 (0.74)	0.83
4	0.75 (0.63)	0.83
5	0.55 (0.46)	0.66
6	0.55 (0.26)	0.66
7	0.55 (0.26)	0.66
8	0.31 (0.26)	0.50
9	0.31 (0.26)	0.33
10	0.31 (0.26)	0.33

The fist costs twice as much, so against W1 enemies needs to do twice the damage. And does it? Nope - it's noticably better, but not worth the points. If you're expecting to fight exclusively W1 enemies, you come out ahead with the claw. Unless you're only making one attack, the extra claw won't earn it's keep.

However, against W2 or more enemies the fist earns it's points back, because the average on a d3 is 2. So, the only situation you should buy a claw instead of a fist is when you expect to exclusively fight W1 enemies. There's also no point in having both, ever - the claw is never better than the fist, just better value. For the same points, a plasma pistol is probably better.

[Cheesegear]

How are you getting 2.67 for the Servitors?

Because I'm not being paid.
First, the idea that I'm going to go through all the units with their asinine fully itemised points costs is...Not what I'm doing if I'm not being paid. The issue being that the model's wargear, and the wargear's points cost, are on two seperate pages, may pages apart from each other. **** that. As Requizen (?) said, it almost feels like that the points system is deliberately botched, so no-one will want to use it, so GW can say that nobody uses points.

Second, the explicit use for Power Rating is that the points cost doesn't matter anyway. Since all wargear is free - all of it - there's no use in counting it, because no matter what, your unit remains the same Power Rating whether you give them four Heavy Bolters or four Plasma Cannons. When it comes to Power Rating, the only points that matters, is the cost you pay for the model, itself. Because Servo-Arms or Plasma Cannons, still PR3. Because all wargear is free.
It's also offset by the Devastator example. You could give them Boltguns (free), but they're still going to be PR7. I should give every unit their base weapons, but the assumption is that the PR system - allegedly - takes into account the unit options. So, while I'm not valuing Servo-Arms on a PR3 model, I'm also not taking into account the points cost on a PR7 model that should be armed with a Plasma Cannon.

Mostly though, I'm doing this the extremely quick and easy way. Because I really don't want to make complex spreadsheets in my own time. Complex spreadsheets are for work. Especially if I'm going to do all Indecies by Saturday. Quick and easy is how it's going to get done. As long as I apply the same 'flaw' to everything, it should end up...Okay. The mistakes made at the bottom are also - hopefully - made at the top. So it should even out, even if it's not perfect. With the Charts I make, I'll see the outliers for myself, and I'll make sure to note it. Not even close to perfect, I know. But I figure I've got at least three months before the first 8th Ed. tournament that I actually want to go to, makes itself known.

e.g; Imperial Guard seem amazing if you look solely at the numbers. But IG also have 8 unit entries that all have PRs over 25. While the next highest PR on a Guard model that isn't a Super-Heavy Tank, is only 13. Guard jump from 13 to 25, with nothing in between. That's going to give you some bizarre numbers. Unless you take it into account.

It's all just a guideline anyway - for casuals. Because that's what the Power Rating, is.
Scouts are 55 Points each (Boltguns are free), with a PR of 6.
9.6 ppp is under the 'average' 14.6 ppp. But there's no way that you'll be able to win games solely by spamming Scouts. At some point you're going to have to put in a Primaris Lieutenant or two (17.5), or Boltguns simply wont do enough.

TL;DR
I shaved my neckbeard years ago. Now I only have 5 o'clock shadow... Unless I'm being paid to not shave.

[lord_khaine]

Wow.. that was a rather long explanation. Cant help but think its a lot faster to just swiftly confess "whoops i forgot about the servo arms"

Hit on the same amount. Aside from vs 5+ saves, 1/6 more wounds are saved by the claw than the fist always.

Im kinda surprised. Does chaos lords really have a powerfist that dont have the regular -1 to hit?

[Cheesegear]

Wow.. that was a rather long explanation. Cant help but think its a lot faster to just swiftly confess "whoops i forgot about the servo arms"

But, I didn't. Not really. Just like I didn't forget that Honour Guard come stock with Power Weapons that increase their base cost. Here's a thing I put in my Huge Nerd Alert

*Complications; No matter what wargear you give a model/unit, it remains the same Power Rating. So basically everything skews higher.

Oh look. Seems like I didn't forget at all, and actually acknowledged it to begin with. Like I said. I didn't forget. I'm actually just lazy.

[Gauntlet]

As long as I apply the same 'flaw' to everything, it should end up...Okay. The mistakes made at the bottom are also - hopefully - made at the top. So it should even out, even if it's not perfect.

The problem is that the model/wargear points split is pretty inconsistent. Intercessors and Scouts cost the same number of PP, but Scouts only line up with Intercessors in points if you give them Sniper Rifles and a Missile Launcher.

If you disregard wargear costs when working out your points-to-PP ratios, then the most efficient options are going to be the ones which have 'baseline' wargear that doesn't change, because then their PP will reflect that.

The 'best' options in a Power Point game are unlikely to be those units - they're much more likely to be ones (like Devastators the Servitors you mentioned before) which have major optional power upgrades which don't affect the PP cost. If anything, you should be looking at the least efficient units in points-to-PP, since they're the ones most likely to have all those upgrades available to make them stronger than expected by PP.

[Cheesegear]

The 'best' options in a Power Point game are unlikely to be those units - they're much more likely to be ones (like Devastators the Servitors you mentioned before) which have major optional power upgrades which don't affect the PP cost.

But then you're just saying that the best options for any unit are simply the most expensive options. It's just not true.

If anything, you should be looking at the least efficient units in points-to-PP, since they're the ones most likely to have all those upgrades available to make them stronger than expected by PP.

No. I should construct a bell curve and spot any anomolies.
What you're talking about is effectively qualitative analysis (i.e; A Guide) A Tactical Squad is 65 Points and PR5. A Devastator Squad is also 65 Points, and PR7. Maybe I should look into that? A Tactical Squad has 3+ Armour, and is PR5. A Scout Squad is only 55 Points and PR6. Why? Qualitative analysis comes later. Probably when I actually have a tangible book in my hands.

[Haruspex_Pariah]

@Cheesegear: So you left the costs of the servo-arms out on purpose? Fair enough. I was asking because I thought there were 0-point servo-arms somewhere, or an exception which says the servitor point costs include their wargear.

I'm not questioning your methodology, though I appreciate that you shared your findings and thoughts, or the state of your neckbeard.

[Science Officer]

I just discovered that there's a shop in my town, less than half an hour's drive, that sells GW products. Gotta get down there personally to see the place, but if I can actually buy stuff without taking an airplane to the next state I might just get back into 40k.

Instead of, you know, theory crafting and lurking. Coincidentally, I've always been partial to Nurgle...

Well, GW recently rescinded their policy about selling their products online, so you can order stuff from places like Miniature Market that have good free-shipping deals. Of course, having a shop in town is very useful for finding opponents...

[Cheesegear]

@Cheesegear: So you left the costs of the servo-arms out on purpose?

I left the costs of a lot of things out. I feel like 'on purpose' might be a bit of a stretch. I was including base wargear costs, right up until Captain in Gravis Armour, and then I gave up trying. Pointing out that the Gravis Captain is the fourth entry in the entire thing. So, once again, I need to point out that pretty much everything is missing wargear. A Captain in Gravis Armour should be 137 Points. I have him written down as 102. I know I did this. Even if you're only just finding out now. Does it skew my results? It sure does! That's why I said it up front!

But no matter how many points he costs, he's still going to be PR7.

Even doing this 'mistake', I've done it consistently, and my results are consistent. 13-15 Points per Power (not including wargear) seems to be how it has been worked out. Does it matter that it might, probably should be 18-20, maybe? Not really. The margin remains the same. As long as I have consistent results, and I know why, then I don't care. I have mostly consistent results, and any inconsistencies I do have, I've been able to explain (Space Marine HQs top out at PR9, then go straight to Land Raider Excelsior at PR24. Or Imperial Guard with a gap between 13 and 25).

The inconsistencies I can't explain...I will effectively call a 'good' or 'bad' unit, depending on the way the inconsistency goes.
Such as like right now, where I'm learning that Honour Guard are 21 Points each, not 71, and that's what I get for not having a hard copy book in my hands, and reading the Dreadnought points cost while transcribing. Are people really that fussed over Servitors, that they missed the fact that I dramatically overcosted Honour Guard by 50 Points per model? Geez.

Fair enough. I was asking because I thought there were 0-point servo-arms somewhere

They're 0 Points when they're a fixed part of the model, and unique to the model (e.g; Mechatendrils on a Warpsmith are 0 Points, and unique to Warpsmiths). Since Servo-Arms on Servitors are optional and not unique to Servitors, they are not included as part of the model's base cost.

I'm not questioning your methodology

You should, though. Because eventually I'm going to have results, and I have to be able to explain how I got them. Unfortunately, it appears that one or more people did not read Complications and Assumptions, where "Everything should skew higher."

[LCP]

Speaking as someone who does quantitative analysis for a living: your "minor flaw" of everything skewing higher makes your analysis completely meaningless. You are comparing variable X = A + B (where B is occasionally 0 but can also be > A) to variable Y and approximating that X = A. Garbage in, garbage out.

Sure, you're not getting paid (and sure, it's not your fault that GW made the points system a pain in the arse). But don't waste your time making something that's just flat-out wrong - particularly across every index as you suggest. Better to do it right across 1 faction and compare the variation (perhaps between minimum and maximum wargear spend) than to do it wrong across every faction and make a wall of text that's no use to anyone. "It's skewed high" is only a useful statement if you have some idea how to correct for the skew - which you don't appear to. "Consistent" isn't the same as "useful" or "correct".

[Haruspex_Pariah]

Well, GW recently rescinded their policy about selling their products online, so you can order stuff from places like Miniature Market that have good free-shipping deals. Of course, having a shop in town is very useful for finding opponents...

I've always been a bit cautious of online sales and delivery, though maybe that's a bit unnecessary seeing how commonplace it is these days.

There's also the atmosphere of a physical shop and physical products that's not quite the same when you're browsing images online. People to talk to, painted samples on display, etc it all adds up.

They're trying to grow the local scene (for my town, not the whole country), which I didn't even know existed. I mean we play the usual video games here, DOTA, Counterstrike back in the day, etc, but I didn't know we had a market for miniatures. It'll be interesting to check out, for many reasons. The only meta that matters is the one you play in after all

[LeSwordfish]

If the Power Level system is designed with a reasonable amount of wargear (probably what GW's writers deemed "normal") in mind, then simply saying "correct as necessary" is meaningless. "All models have a 1:1 points:power rating, if you correct as needed."

(In addition, note that many things are more expensive for the first five than the next five? 2PL is not what you pay for +1 attack and +1Ld on the champion. Things without wargear options like Warp Talons don't have this, and things with fewer first-five-only options (Terminators, say) have it for less.

I happened to store a bunch of points values for Chaos stuff earlier this week (I wrote a quick list-building spreadsheet for my own collection, now it's on a Per Model basis), so let me see what analysis I can do with that.

	Points	PL	PpPl
Things You Can Outfit Differently
5 Raptors:	85	7	12.1
5 Raptors (Power Weapon, 2x Melta):	123	7	17.5
10 Raptors:	170	12	14.2
10 Raptors (Power Weapon, 2x Melta):	208	12	17.3
5 CSM:	65	5	13
5 CSM (Power Weapon, Plasma Gun):	82	5	16.4
10 CSM:	130	9	14.4
10 CSM (Plasma Gun, Power Weapon, ML)	172	9	19
Daemon Prince	146	10	14.6
Daemon Prince - Wings, Claws, Bolter	189	10	18.9
Chaos Lord	74	5	14.8
Chaos Lord - Powerfist, Plasma Pistol, Jump Pack	122	5	24.4
5x Terminators - Combi-Bolters, Power Axes	185	14	13
5x Terminators - RAc, 3x CB, 4x PW, 2x LC	261	14	18.6
5x Havoks	65	7	9.2
5x Havoks - 4x Missile Launchers	165	7	23.5
10x Havoks	130	10	13
10x Havoks - 4x Missile Launchers	230	10	23
Things That Are Pretty Much Always The Same
5x Warp Talons - Double Claws	140	9	15.6
Heldrake - Baleflamer, Claws	215	10	21.5
Typhus	164	9	18.3

This is, I should note, my own best guess at a "Normal" "fluffy" "balanced" unit - because I suspect that's what GW's rules writers would have done, rather than taking the two unrealistic situations of a maxed-out unit or a minimum nothing unit.

What i'm preliminarily seeing is that actually it doesn't seem to matter that much - of this small sample, there seems to be about as much variance between the non-bolded "bare" options, and the bolded "fully equipped" options. I still suspect that the system is designed to include wargear, though - even though they're anomalous when compared to the others, look how Havoks are almost precisely the same number of Points per Power for a squad of five as of ten! The same for Raptors. Without taking that into account, they'd be way off.

[Cheesegear]

Speaking as someone who does quantitative analysis for a living: your "minor flaw" of everything skewing higher makes your analysis completely meaningless.

I know. That's why I'm not trying very hard. Try Hard mode comes when I have the Index in my hand. But, by that point, I'll have all my formulas, tables and charts that just need data input.
Essentially what I'm doing is preliminaries. Full of flaws. But it's the best I'm willing to do, with what I have - that is, not a hard copy.

Besides, it's not like I'm writing 7th Ed. army lists I really want to play test right now. I actually don't have anything better to do.
I mean...I could paint...But until I know what I'm working towards, I don't know what to choose to paint first.

Garbage in, garbage out.

Correct. But I'll have a place to start from. To be refined later. And I am actually kind of bothered that no-one picked up that I'd written down Honour Guard at 71 Points per model, because that changes the mean and SD back down to sane levels, consistent with what I have for Dark Angels and Blood Angels.

"It's skewed high" is only a useful statement if you have some idea how to correct for the skew - which you don't appear to.

I know exactly how to correct it. By putting in the wargear costs. Which I don't really want or need to do. I'm still getting the results *I* want. The only way that wargear becomes relevant, is if it costs more or less depending on what Faction you're pulling it out of. At which point play-testing becomes necessary to determine whether or not a particular piece of wargear is worth more or less depending on what army it comes out of.

Space Marines pay 10 Points for a Heavy Bolter.
Imperial Guard pay 8.

But Imperial Guard also have a worse Power Rating. So, is the lower cost of the Heavy Bolter, a result of the lower Power Rating? Or does it compound things, and Heavy Weapon Squads have a double advantage being cheaper in both forms of the game, and spamming Heavy Weapon Squads will be a way to victory? (The answer is 'yes'. Heavy Weapon Squads are the ****).

[LCP]

Essentially what I'm doing is preliminaries. Full of flaws. But it's the best I'm willing to do,

If the best you're willing to do produces meaningless results, don't do it. It's just noise.

I know exactly how to correct it. By putting in the wargear costs.

That's not what I meant. I mean that if you put up these heavily biased approximate numbers while acknowledging their bias, then in order for them to be of any use to anyone you need a rule to let people compensate for the shift. Which seems very difficult to construct. Obviously if you do the maths right from the start then this isn't an issue.

And I am actually kind of bothered that no-one picked up...

If you make a big song and dance about how you're making a big nerdy effort and aren't willing to make any more, why are you surprised that people aren't willing to error-check your work for free?

I'm still getting the results *I* want.

Why do you want meaningless results?

[Cheesegear]

Basically what I've done; Is announced research before I've finished. Which always goes down well.

[LCP]

If by "early" you mean "before you finished actually doing the sums" then yes.

[Gauntlet]

But Imperial Guard also have a worse Power Rating. So, is the lower cost of the Heavy Bolter, a result of the lower Power Rating? Or does it compound things, and Heavy Weapon Squads have a double advantage being cheaper in both forms of the game, and spamming Heavy Weapon Squads will be a way to victory? (The answer is 'yes'. Heavy Weapon Squads are the ****).

Militarum Tempestus Command - 64 Points / 3 Power Points (including wargear)
Plasma Gun x4

^ This unit is one I expect to be one of the best choices around. It's four BS3+ plasma guns for an extremely cheap cost, and they appear immediately in rapid fire range as soon as you need them. You can take a Tempestor Prime for every two units, and give each of them rerolling 1s to hit, to mitigate getting your models killed. Alternatively, Roboute.

They're also, interestingly, better at killing big multiwound models than Meltaguns on the same models. You aren't within half range, so the Meltagun only averages 3.5 damage, while the Plasmaguns do 2 damage with twice as many shots, while being significantly better against anything with <5 wounds or more than 12" away.