This isn't completely correct. "Ranged" accuracy affects only our basic attack, i.e. Flurry of Bolts. "Tech" accuracy affects everything else, including both tech attacks and special ranged attacks. You can see this for yourself by hovering over your ranged accuracy number on your character sheet. The tooltip breaks it into "basic accuracy" and "special accuracy", where the special accuracy number is equal to your tech accuracy.

Just to be clear, accuracy has nothing to do with mitigation. It only affects hit/miss/dodge. Beyond that, as a Leth/Eng hybrid, these are the attacks that do not need accuracy to always hit an ops boss: Corrosive Grenade Corrosive Dart Interrogation Probe Explosive Probe Orbital Strike Are you really finding that 80% of your damage comes from those five abilities?

Couldn't have said it better myself.

I assume they are all snipers, yes?

Neither crit nor surge have a soft cap. If you're referring to when they shift in priority relative to other stats, then that depends on your other stats. The links in my signature can help to get a more precise answer than that. If you're referring to basic ranged accuracy, yes, that's true. Special ("tech") accuracy at that point reaches 110%, which is just enough to overcome operations bosses' innate 10% defense. "Tech" accuracy applies to all our abilities except Flurry of Bolts, which uses basic ("ranged") accuracy. For PvP, many gunslingers also prefer to reach 110% special accuracy, since that is the amount of innate defense inquisitors and consulars have. Other classes have 5%, and of course tanks could have much higher than 10%. Nope. Cunning over power all the way. Hopefully that helps!

Just a couple suggestions based on the log in that link: Corrosive Dart -- This ability was not used once. For pure MM, CD is generally a stronger filler than Explosive Probe (assuming it will tick all the way out and isn't being overwritten). Instead of extra Snipes to dump excess energy, try using CD. Of course, there will be times when CD is already applied and you still need to dump energy. Those are the best times for EP. Snipes other than to trigger FT should be a last resort. However, my bigger suggestion is: Activity -- By my count, only 230.75 out of the 341.67 total seconds were spent performing attacks, or only approximately 67.5% of the time. (This is with no alacrity and assumes that every Ambush had a Reactive Shot proc. With the high crit rates seen on Snipe and SoS -- 44% and 54%, respectively -- this assumption is likely not far from reality.) Some delay from latency is expected, but many of our abilities can be queued, which helps to mitigate that delay. With good focus on making sure abilities are used as soon as the GCD is over, I think you'll have no problem moving that 8.76s average FT period closer to the ideal 6.0s. Hope that helps!

No problem, but I hope it's not persistence that makes my point. I hope everyone will keep questioning things to see if they're true. When a conclusion is drawn, there are always assumptions that went into that conclusion, so ideally others will test to see if the conclusion holds true when using a different set of assumptions that could also be accurate. (It's science! ) There's really not much more to theorycrafting than that.

Yes, these two will do the same expected damage. However, case 2 delays Aimed Shot by an additional 3 seconds, meaning that over an average fight, we'll lose part of an Aimed Shot that we wouldn't lose with case 1 (1/4th to 1/6th gets turned into Charged Burst, depending on how much AS weaving we're doing in the rotation). By saving the 1 second of activation time, we'll also gain 1/6th of a TS and trade 2/5ths of a Flurry of Bolts for 2/5ths of a different filler (e.g. VS) to balance energy and time. I still found that to be a loss. Perhaps you get a different result due to different damage numbers, or maybe you have a completely different take on it. Either way, I'd like to see your math.

I agree, and the math I posted at the link above also agrees that this is still the better option in this situation. The analysis I performed can be easily adapted to show that Aimed Shot is still a better option than Charged Burst, even if Smuggler's Luck is active: Instead of 1850, the expected damage from a guaranteed-crit Charged Burst would be 2590, assuming 33% crit and 75% surge. This gain of 740 damage closes approximately half the deficit between CB and AS in this situation. AS is still the better choice in this case. My analysis also assumed that Speed Shot is still on cooldown as it comes time to use Aimed Shot. If SS is available, then using it in this situation means Trickshot will be delayed by 3 seconds. The energy difference in this case is a bit trickier to model, but what this scenario effectively does is trade away 1/2 a Trickshot and 1 Aimed Shot for 1/4 of a Speed Shot, part of a Flurry of Bolts, and part of a filler ability (for energy and time balancing). It's not likely to be a good deal. I'd be happy to be shown otherwise.

The only way to eventually accept statements as facts is by seeing objective evidence; re-stating a position does not increase understanding of it. This is how science works. (In my view, theorycrafting is a science. The theorycrafting that I've done has been approached scientifically.) There is no objective evidence that says MM is the worst spec in every situation. There is objective evidence that says it's not the worst in every situation -- specifically that it is quite competitive in an ideal stand-still environment, according to continually refined simulations. There has not been a large effort to gather empirical evidence, and to create scientific data indeed takes a large effort. Please, though, if you want true facts, help with the effort to create supporting data rather than immediately dismissing everything because it's not perfect.

Lethal Purpose is the proc I'm referring to. Get lucky and we end up with lots of energy to dump, get unlucky and we have to react to make the correct decisions that will keep us in the upper energy bracket as much as possible. Also, the simulation tools disagree that Marksmanship is unequivocally the weakest spec.

I apologize -- I probably should have used less provocative wording. The point I am trying to make is that if MM is so easy, then surely it leaves you with plenty of brainwidth to focus on improving your play. I agree that idealistic simulations are idealistic. However, part of sniper skill is performing as close to ideally as possible -- meaning maximizing the energy we use and using it as efficiently as possible. Unique to MM, this has an added dimension of keeping in cover as much as possible (for Sniper's Nest). So in MM spec, while we're not busy trying to calculate the next best ability to use because it's so easy to follow a consistent rotation, instead we spend extra effort choosing when and where to drop into cover to maximize our damage. It's a different proficiency to develop from Lethality's proc-based reactions and Engineering's cooldown-based ability prioritization and boss path prediction.

I can try.

The spreadsheet assumes extra energy coming from Lethal Purpose goes toward replacing Rifle Shot with Snipe whenever possible. Here's a thread in the gunslinger forums in which I calculated assuming using Explosive Probe as the replacement ability instead of Snipe. I decided to use Snipe in the spreadsheet because it's available for both the pure Lethality spec and for the Leth/Eng hybrid spec. (I'd love to hear feedback if we think a different ability would be better used for this purpose.) What this means is that if you find that Lethal Purpose increases your usage of a different ability besides Snipe, then the Spreadsheet is probably underestimating the benefit of Lethal Purpose, proportionally to: (that ability's DPE − Rifle Shot's DPE) / (Snipe's DPE − Rifle Shot's DPE) In short, it's probably pretty close. Use the spreadsheet for a pessimistic estimate, or a calculation like in the link above for an optimistic estimate. This. (Seriously, though -- I hope the post you made wasn't actually too long for people to read... I mean, they did come to the forums. No audiobooks here.)

I'm confused by this. If Quick Aim doesn't proc for Aimed Shot, you prefer to use an energy dump like XSFF or Sab Charge, rather than AS or CB for TS?

Do you have any evidence for this? I argue that we should use Aimed Shot whether Quick Aim is active or not. Here's my justification. If you have a link to yours, I'd like to compare math so we can see why our conclusions differ.

It's kind of understandable why this is frequently trumpeted around as "just the way it is" -- and nearly always is "easier" paired with "weaker". First there is a mountain, then there is no mountain, then there is a mountain. The MM spec seems easy and obvious. Nothing could be more simple as you look across the valley and plainly see the mountain. As you reach the base, you look around and your view is more obscured than you thought it'd be, confused by the trees and details you could not see from far away; you see no mountain. You climb and climb, passing the treeline and cresting the summit; now your view is clear and you can once again make sense of the mountain. The marksmanship spec can certainly be straightforward. If, however, you find it exceedingly easy to play in comparison with the other sniper specs, I suggest that you may be having a very easy time playing poorly.

This is a really cool idea. In addition to this great list of positive arguments, I also offer the iconic archetype from the movies -- one Mr. Vader who very clearly went from Jedi to Sith when they became FOTM with Patch 66.

What is the design philosophy behind melee/ranged defense versus force/tech resistance, both in PvP and PvE?

Of course you know that cover only gives ranged defense, so I assume you're talking about the defensive cooldown in Dodge and the accuracy debuff in Diversion.

No. Diminishing returns for crit rating (and any stat that has DR) start immediately at 0 rating. There is no point at which DR suddenly becomes an issue. It's happening the whole time, no matter how much crit rating you have. So how much crit rating do we want? Well, we want as much as we can get, along with as much power as we can get (and as much surge as we can get, etc.). The problem is that in order to get more of one stat, we either have to keep getting upgrades (which has its limits depending on how much is currently available) or we have to trade out one stat for another -- at an increasingly expensive exchange rate due to DR. At some point, the value of both options is equal, and that's where we want to be. So where is that point? Well, there's not a simple answer to that. You'll see lots of suggestions along the lines of, "Here's what my stats are, and I do well with them." One problem with those kinds of answers is that those are their stats and not yours; using their skills and not yours; and with their spec, ability selection, group composition, and combat environment, not yours. The other problem is that those quoted stats are very often not optimized anyway. They may be close enough to see success, but you probably didn't come here to ask, "How can I get 80% of the way to ideal stats?" Here is my typical approach for answering these kinds of questions: (1.) make an extremely generalized stat priority statement; so general that it's probably too vague to provide a full answer, then (2.) suggest that you try using the Sniper/Gunslinger Spreadsheet, which is linked in my signature; this will give you a precise answer. So here we go: Accuracy is good up to the soft cap, although it might not always be the top priority all the way to 10%. Surge is very strong at first, but has a very sharp DR. Crit starts out better than power, but passes it relatively quickly. If you have end-game gear, you probably want power over crit. Most snipers seem to over-value crit -- you'll want to get most of your crit chance from cunning. You can figure out the details for your specific situation by using the spreadsheet.

There are no soft caps for any stat except accuracy. Diminishing returns start applying as soon as you have any amount of crit rating, cunning, surge, or any other stat aside from power or tech power. It's a smooth curve from 0 rating all the way to infinity. The best way to find your stat weights is to use the Sniper/Gunslinger Spreadsheet, which is linked in my signature. There is further discussion about stats (and other topics) related to MM snipers in the compendium linked in my signature, as well.

Neither crit nor surge have any notable soft caps. To address stat weights: For DF, crit's value has been overstated, in my opinion. If we assume the following numbers: we have 75% bonus critical damage (i.e. surge) we're capable of 1500 DPS gaining more energy will allow us to occasionally use Sabotage Charge in place of a Rifle Shot we deal 2500 average damage from Sabotage Charge we deal 900 average damage from Flurry of Bolts ...then increasing our crit chance by 1% will add an estimated 13.52 DPS. Of that, 11.25 DPS is from the added damage of more frequent crits, and 2.27 is from the added energy regeneration. In other words, Fighting Spirit adds ~20% to crit's value (actually a little less due to the DR conversion from crit rating to crit chance). For example, if as SS you value crit over power until you have 200 crit rating, then as DF that cross-over point would be something less than 240 crit rating. Here's the math to back it up: (link) Feel free to do a save-as and substitute your own assumptions. If you want to find your own personalized stat weights, try out the stat weights calculator in the Sniper/Gunslinger Spreadsheet (linked in my signature). It will give you precise numbers based on your inputs.

Unfortunately, the math disagrees with this. For snipers, cunning has been shown to be 10-15% more potent than power, virtually no matter how much cunning or power or crit you already have. It's true that with obscene levels of cunning, its crit chance contribution will DR so heavily such that it no longer makes up for power's very slight advantage in bonus damage contribution, but that level is so far beyond what is attainable as to be irrelevant to gearing discussions.

The accuracy soft cap (against operation bosses) is 100% ranged/110% tech. Soft caps against other targets vary by their defense chance. However, that's not to say it's necessarily optimal for Lethality to be at the soft cap -- primarily because Corrosive Dart and Corrosive Grenade are tech attacks. If we have relatively low surge, then that's probably worth more than accuracy at a certain point. (The best way to figure out that point is to use the Sniper Spreadsheet.)