I know the final few minutes of this were posted yesterday (and I appreciate the enthusiasm there!) but I figured I'd also go ahead and post the full run and dive into some explanations I thought were interesting. First though I'd like to thank some major strategic contributors from Twitch – this section will be a little lengthy but it does have some insights as well, so skip at your own risk if you're here for details.
Big Thanks to:
Kog was the first to notice the way to determine the right Atraks from her replications, which is by looking for small changes in the health bar. He was also the first to suggested a ranged stun for the third stun, which was a strategic breakthrough. These critical contributions made the run possible, but I also have Kog to thank for a huge number of minor optimizations. His investment in the challenge really improved my outlook on it, too, so huge thanks to Kog for his above and beyond help.
Fryco pointed me in the direction of mods that relieved a ton of pressure from adds and allowed me to focus near single-mindedly on the rest of the mechanics. Specifically, arc resist mods are just crazy good in this encounter and I didn't know that, but the enhanced operator augment mod is actually insane here too. With this setup I was nearly invincible, it took massive misplays for me to die to adds.
Wildfire corrected my understanding of the wipe mechanic's timing. I was under the impression that a strict timer started at the beginning of the encounter, but Wildfire pointed out that the timer actually starts after every stun. This helped correct a huge, needless amount of rushing on my end that made consistency possible in a way it would have taken forever to arrive at on my own.
GrimAgent suggested crouching to cleanse the replication from myself, which seemed to make the self-cleanse "hitbox" more forgiving. It's hard to overstate the importance of consistent cleanses, but runs depends entirely on it – a slow cleanse (we're talking ~an extra second in a lot of cases) means one less guess on the boss, which reduces chances of success by 25% on the spot. A "very slow" cleanse that's 3 seconds late is a wipe. It's a kind of wonky mechanic that needs to be done near perfectly, so a little extra forgiveness with it helped massively. Grim also was around for a lot of the pre-attempt theorycrafting to get runs off the ground. Also thanks to Grim for pretty much all graphical stuff that went into the video (thumbnail and RNG explanation image).
Butters also helped with quite a bit of testing during the early phases when I still needed a person to rely on for some help in the encounter while I was attempting various solo experiments, so thank you very much for that help Butters.
Faldo_ found the best cleanse spot, which was the doorway of the airlock itself. There aren't many places to check within the airlock and I had tried what I thought was everything with varying levels of success, but Faldo's approach turned out by-far best. Interestingly, the doorway is too short to cleanse on without crouching, so it seems like this development came in two small steps for a major improvement to consistency. Faldo is also the first player to achieve final stand in this solo which is a huge achievement.
One thing I found that I'm personally satisfied with is the "soccer" strategy of cleansing two replications at once after juggling them. For a long time I was cleansing replications in individual airlocks after juggling them (for speed), but after deciding that keeping more airlocks functional was a higher priority, the cleanse-one-kick-one approach was my idea. This whole mechanic came to my complete satisfaction because the replications sometimes bounce away immediately after stuns, which I complained about as "the wrong time to play soccer" every time it caused a wipe (which wasn't uncommon), so to have the kicking become part of the intended strategy was serendipitous to say the least.
Kyro and Lemons suggested I hold onto my super for movement reasons instead of using it to clear servitors. While I don't use that strategy in this run, it was an impactful change that allowed me to continue runs that would have otherwise been impossible.
Disco from my day 1 raid team was the first person I heard mention the idea of self-cleansing with ricochet rounds. I'm not sure how many people thought of this or if crediting anyone is really appropriate, but Disco considered using RR to self-cleanse in the middle of our world's first raid attempt, and I think that's very clever of him to have noticed so early on.
And finally, my gf Vanessa who had to babysit our dog and let me game until odd hours of the night so I could run this to my heart's content.
A lot of others suggested I use Lament skating for speed, but I prioritized consistency with what I am already familiar with over gaining a small (but significant) boost to speed that would disrupt my intuition. Overall this run is so heavily dependent on speed that anyone attempting it should at a minimum be comfortable with everything I do, but mastering sword movement on top of that would be a big help.
Interesting details about the run:
I'm gonna start this section with the two most frequently asked questions:
In order to determine the correct replication without scanner, you damage the replications as much as you can without actually forcing wrong stun / wipe. By doing this, the correct replication will take noticeable damage on its health bar, but incorrect ones won't wipe you. This doesn't entirely eliminate RNG, but it increases your odds enough to make this a realistic challenge.
In order to cleanse the replications from yourself as operator, you use ricochet rounds against a wall in the airlock.
This is probably the most interesting solo we've ever had in Destiny in my opinion. Without necessarily speaking to its difficulty, it has everything a solo challenge has ever had in Destiny bundled together. There is a major RNG barrier to success (which I'll get into in much more detail), there is a incredible stress test on movement speed and precision, inaccurate shots are often punished with a wipe, and anything short of the best build quickly begins to push the run into the realms of impossibility. On top of these factors, the narrow timeline for elevator cleanses combined with the 4-stun DPS cycle seems to almost suggest Bungie deliberately intended to keep this option possible, which is probably just a coincidence, but it's amazing to see exactly how narrowly this challenge avoids impossibility. These opinions on the run are absolutely subject to change if any more strategic breakthroughs are found post-completion, but barring that possibility this challenge is the ultimate gauntlet of Destiny mastery and dedication. I'll expand on all those points one at a time, with RNG last because it's the most interesting to me.
As far as movement is concerned, every wasted instant hurts your chances of success substantially during the DPS phase. An icarus dash without a boost, an unnecessary clamber, a bodyblock from a wretch, or any other miniscule thing is likely to reduce your chance of completion by 25% on the spot. This is because, while stuns in space are effectively a given when playing well, you need to immediately cleanse the replication in an airlock and get back to the orbital elevator pods afterwards. During this window, any tiny mistake robs you of a guess on the base side replications – hence, even very small movement hiccups can reduce chances of success by 25% (or 33% on the second stun). In my runs I was not able to test all replications at base side even a single time – even when my space side stun was as convenient as possible – so you're going as fast as possible to fight for the best odds, not to guarantee success.
Movement hiccups after a base side stun are a bit more forgiving, but small mistakes there can compromise your ability to pick a good position to stun from on space side. Ideally on a third stun you are able to stun from right next to an active airlock, but getting into position takes time – if you trip up base side, it's likely you'll have to stun from a suboptimal position for the third stun, which greatly reduces your chances of making it to the final stun in base afterwards. It's hard to overstate how impactful even tiny movement slip-ups can be in this run. Also, this run is subject to the typical nuances most Destiny challenges are, such as kicking around replication instead of picking it up for no apparent reason / sword attacks tracking to enemies instead of the direction you're facing / etc.
Again, noting that I'll go deeper into this later, RNG can make your forgiveness for movement better or worse, ranging from totally impossible, to absolutely unforgiving, to "please don't let me mess this up, this is as 'easy' as it gets." In so many words, your speed and precision with movement are major decisive factors in your odds of success.
While there isn't much to say about aim, there are two critical points where accuracy is do-or-die. During any "testing" of the bosses, a missed grenade launcher shot is a major time waste. This, like with movement, can easily mean a single missed shot reduces your chances of success by 25% or 33% depending on the phase.
For example, if I am going base side after a good 3rd space stun and I've already killed close right base replicant, I still need to test two of the three remaining among left, middle, and far right base side replicants. The optimal way to do this is to stand by middle and shoot left without moving (as there's a 2/3 chance I'll be moving farther from the real one if I do, and I still have time to get to left if it's the correct one since it's my first guess), then shoot right while moving toward it (guess is now a 50/50 and I'm running low on time so I want to be as close to both options as possible), then either attack right or middle after I've ruled out the others. In this scenario, if I miss the shot on the left or right side replicants, my odds of success drop from 100% (4th stun is guaranteed by accurate gameplay following a good RNG 3rd stun) down to 66% on the spot. You simply do not have time redos with the base side stuns, so precision in aiming here is absolutely critical. Also, I assume it sounds easy enough to hit a boss with a grenade launcher, but it really is more difficult than it sounds, particularly when the boss randomly meanders behind partial cover.
As with movement, good RNG can help relieve the punishment of poorly aimed shots, but this is a challenge in which you want to rely on RNG as little as possible since you're forced to rely on it to a significant extent by design.
The second place where aiming is critical is during the cleanse. As stated before, you cleanse by using ricochet rounds against the airlock door while crouched for best consistency. What's not immediately obvious about this is that a slightly angled shot (left or right of direct center) will be a miss that doesn't cleanse you. Same with too high or too low. The cleanse happens as follows: get to airlock door-> crouch -> push yourself up against the door -> aim directly at it, at the right height, on center, without any clear visual indication you're aiming at the right spot -> fire -> trust you got it right and keep moving. The hitbox isn't exactly "tiny" per se, but it is absolutely small enough that your aim needs to be careful and deliberate every time you cleanse.
I have never actually felt the need to perfect a modded loadout for my character before this challenge. During other difficult runs like GM nightfall solos I would use the necessities and wing it from there, but in this case, mods and gear were insanely important.
Kinetic weapon: Needs to be ricochet rounds. Strong preference for mid range (so not SMGs ideally) and high RoF (so not HCs ideally) for cleansing and optimal engagement range. Feeding frenzy is a valuable perk since you can't crit shanks, and rampage or swashbuckler are ideal damage perks since you don't want to waste time reloading (particularly wearing transversives, which improves speed and is therefore probably best). I used a False Promises that dropped in the middle of my early attempts in the most serendipitous imaginable fashion At the end of my runs, it had exactly 8000 kills on it. Prior to that I used a RR/Rapid hit/Kill Clip Sacred Provenance, which was also very good and would have done the job similarly well.
Energy weapon: Needs to do exactly enough damage to clearly hurt the right replication but not cause a wipe on the wrong replications. Also needs to be usable at long range. Strong preference for mid-air accuracy for movement reasons. I used truthteller, and I'm honestly not sure anything other than a grenade launcher can meet all the demands of the challenge. As an incredible bonus, blinding grenades disable servitor immunity which makes them far less of a hassle to deal with (particularly because they can make the boss immune under bad circumstances), and greatly reduces enemy add pressure (which is one important piece of the "I feel invincible" build).
Heavy weapon: Ideally, Two Tailed Fox, but since I didn't have that, I used Lament. This is pretty much exclusively for boss DPS.
My armor mods were built around two key concepts: 1. maximizing DPS, and 2. staying alive. Shocking.
As far as staying alive, I used two arc resistance mods (which are insanely strong, I had no idea). This alone makes an enormous difference in the boss room because almost every bit of damage thrown at you is arc, so those mods are unbelievably high value for their cost. On top of that, I used the raid specific "enhanced operator augment" mod, which is absolutely busted. The way this mod works is that whenever you're down to red health, you get periodic bursts of health regeneration. I did spec into recovery for the most part as a general stat, but so, so many times I was hit hard enough to be low red only for this mod to kick in and bail me out. With the combined effect of high recovery, double arc resist, and op mod (an appropriate shortening), the enemies in the encounter almost never killed me. The sheer damage resistance and health resilience was actually hard to believe.
For the damage aspect, I exclusively focused on maintaining lucent blade. Since lucent blade doesn't stack with things like rifts, wells, or much else, this was an easy way to give my damage a strong buff without having to change gameplay to make it happen. I ran supercharged (+2 charge cap), charged up (+1 charge cap), and stacks on stacks (x2 charge gathered) to minimize the amount of intentional charge gathering I'd have to do, taking charge to get the charges (orb pickup = +1 charge), and lucent blade as the singular way to spend those charges. I would've loved to use protective light as well just to see how unkillable I would be, but anything that might take my charges before a DPS phase wasn't an option.
The rest of my mod slots were spent on things like grenade scavenger (artifact mod) and finder, which all helped, but ultimately weren't critical in the run. My split of recovery and mobility was tier 8 and tier 6, and those numbers also helped but were probably not critical.
On the uncanny "did Bungie intend this?" design of this challenge:
Replications are the most significant limiting factor to success in this challenge. You can only cleanse them in space, but you have to pick them up in base, too. This means that you'll definitely have times where you can't cleanse instantly after picking up regardless of your RNG – on the first stun you can, on the second you can't, etc. Here's why that's interesting – as a result of this design, the optimal pathing is effectively as follows:
1st stun / cleanse 1st / run base 2nd stun / run space 3rd stun / cleanse 2nd / run base 4th stun / run space
… so 5th would theoretically be
5th stun / cleanse 3rd / run base
at which point you'd be left with replications 4 and 5 uncleansed while headed the direction exactly opposite of where you need to be to cleanse them. If the DPS phase ran any shorter cleansing would be trivial, and if it ran any longer, it would be impossible. To me, this is quite a coincidence.
But that's not all there is to it – replication balls wipe you if left unattended for ~45 seconds, and this timer is reset when knocked off a player (notably, NOT when picked up). As a result of this, combined with the ~13 second elevator ride and pickup mechanics, this is the timeline of a ball juggle to get two balls to space at once after fourth stun:
- Start by dropping 3rd at base elevator, resetting it
- Grab 4th and drop it at base elevator to reset it, too
- Pickup 3rd / go to space / drop 3rd / go to base
- Pickup 4th / RESET 4th / Pickup 4th / go to space
- Drop 4th / pickup 3rd / RESET 3rd / begin soccer
During this section, you have to leave stuns unattended for nearly their entire timer as you go up and down the elevator. As it takes some time to pickup the ball, drop the ball, and move between flat walls and the elevator pods – on top of the elevator movement itself – it's not unusual to run those timers dangerously low. This actually wiped me at least a handful of times due to going slightly too slow – it just so happens that my successful run was very clean in this regard so I had a very generous 6 seconds every time at least.
All this said to reinforce one point: A longer DPS cycle would be impossible because you'd have to juggle during a DPS phase, which there isn't time for, because the time allotted by the replication mechanic and elevators makes it so. In my opinion, quite the interesting coincidence if this is unintended.
And finally, on RNG:
This challenge is NOT a Riven solo. Riven's eyes was a 1/45 pure odds that you couldn't help at all, combined with the 1/2 chance of her even going to the right side for a start the fight. Riven solo was a beast of its own, but the true "difficulty" of it was almost entirely found in fighting against terrible odds. Atraks-1 is very different in the sense that, while it does have a large RNG component, there are a myriad of ways you can mitigate it and keep your chances of success far better than the 1/90 on a fresh Riven start.
By my calculations – which are summarized in
this image link – you have a 29.3% chance of any run being completable at the outset, considering there is a 54.2% chance of success on any single DPS cycle, and you need two of those in a row.
The information in that image is subject to two major considerations:
Firstly, boss movements can totally let the goose loose. If a boss stands in a particularly terrible spot, it becomes much harder to check it and can reduce your chances of success by forcing you to guess with less testing either by committing to the weird spot or by ignoring it and hoping it's not the correct one. In any case, the overall good/bad value of the stun cycles I've covered are not absolute by any means.
Secondly, the testing order also matters a lot. The clearest example is that any run in which the 2nd stun is on base left I have considered an irredeemable wipe. This is because testing all four in base is nearly impossible, so it makes sense to prioritize checking base far and base close (both of which are on the right side), so there's a high probability of finding the right one. Because the limiting factor is travel time, you are forced to check "sides" rather than specific replications.
It's very possible that improvements to routing and testing speed could come later and improve the wipe / difficult scenarios to better statuses, but as of right now, the probability of it being a completable run is somewhere around 29.3% with the current knowledge – so in that way, it is a very RNG based challenge (as the variability in difficulty from the top end of that 29.3% to the bottom of it is very high), but as far as completability goes, it isn't just hoping for the best all the time. Rather, this challenge is entirely about overcoming bad scenarios.
Overall, I love this raid encounter, and while this challenge is very RNG heavy, and 75% is not 75% somehow, I enjoyed doing it a lot.
Also, HUGE congratulations to Tier 1's Vendetta who also just completed this challenge. Incredible achievement on his part and he was impressively quick to get a completed run in. Can't say I'm too surprised though, he's an insane gamer. Major kudos to him!
Source: Original link
© Post "Solo Atraks-1, Fallen Exo + Comprehensive Writeup" for game Destiny 2.
Top 10 Most Anticipated Video Games of 2020
2020 will have something to satisfy classic and modern gamers alike. To be eligible for the list, the game must be confirmed for 2020, or there should be good reason to expect its release in that year. Therefore, upcoming games with a mere announcement and no discernible release date will not be included.
Top 15 NEW Games of 2020 [FIRST HALF]
2020 has a ton to look forward to...in the video gaming world. Here are fifteen games we're looking forward to in the first half of 2020.