I've tried to run the Pentagon mission twice now for the weekly project, firstly on Heroic with a matchmade group of randoms (someone else was lead), then again solo on Hard (to speedrun and get it over with). Both times as soon as I/we got the marauder's health down by maybe 20-50%, the game threw a delta-03. Reconnecting put the whole group back at the start of the mission.
Looking on the discord, I see others complaining about the same thing in different circumstances - eg being unable to complete Liberty Island.

I have yet to test this more fully (eg, is it every marauder in every circumstance? I completed legendary Roosevelt last Tuesday without a hitch) but figured I'd post this now as it seems like a pretty major problem if it's blocking players from completing WONY, and I'll add as I do more research.

Frustratingly, research will be slow as I have to run the whole mission again to reach the drone (and having completed the weekly invasion, I won't have any 'aerial recon' marauders in DC to test on).

It sounds like you're a 'hoarder', trying to maintain a collection of every potentially-useful item in the game - obviously this game doesn't suit that approach (hence your post), but you're more likely to successfully change your approach than change the game. I'd advocate the following:

Firstly, instead of the 'one of everything' wish-list, limit yourself to items in builds you actually play or particularly want to play: you want a short enough list here that you can maintain a mental list of what particular items, secondary rolls & talents you need (instead of having to actually compare against your existing items). If you haven't already, I'd recommend taking a look at some recommended builds other people have put together to get an idea of what's most useful and what isn't.

Secondly, develop a habit of glancing at loot drops when you pick them up, quickly judging them against your mental wish-list, and marking as junk anything that doesn't fit - which is likely to be the vast majority of it. Once you get used to it, it takes barely a second per loot item to evaluate and grab, far quicker than manually going through everything later. This way, whenever your inventory gets full, you can simply hit the 'scrap everything marked as junk' key and it'll all go away.
You may potentially scrap some good item that would have been handy in a future build you haven't considered yet, but that's a risk worth taking for the time & effort it saves you, and there will always be more loot drops in the future.

On a side note, I suspect a 'bottomless stash' would turn out to be a terrible idea, without good discipline it would be far too easy to let it build up to an unmanageably huge task to sort & tidy!

Just tested this for myself and it does appear to be true. This is pretty major, with a lot of players unknowingly reducing their potential maximum expertise strength by donating all their junk. If this is intended behavior, it really ought to be made more visible so people realise before they 'lose' too much potential.

There are currently 249 weapons (including named and exotics as separate items), 23 brand sets, 13 gear sets, 44 named/exotic gear pieces, 41 skill variants, and 6 specialisation weapons. So that's 376 individual proficiencies to gain and 10 levels of each. That's obviously not counting any new items added in the new season.

First off, everything I talk about here is from the Intelligence Annex here: http://ubi.li/PFoYL so please refer back to that for anything you wish to confirm/clarify/contradict.

I'm starting this thread because I'm seeing quite a lot of misunderstandings about the new Expertise/Proficiency systems, and would like to try and clarify some of the less obvious points about how it is planned to function, and also some of the rather surprising ways in which it could affect our gear choices.

Explanations
First off, we need to look at the two interconnected systems of Proficiency and Expertise.
Proficiency is earned on a per-item-type basis, and affects which items can be improved. Proficiencies have exactly 10 levels each, which are earned by playing with the item(s) equipped, or by donating items and/or materials.
Expertise is earned on an account-wide basis, and affects how far items can be improved. The number of possible expertise levels is currently unknown.

For Proficiency, items are grouped by type as follows:

Weapons (Black Market AKM, P416, SOCOM Mk20 SSR, etc).
Brands (Gila Guard, Alps Summit Armament, Walker, Harris & Co., etc).
Gearsets (Striker’s Battlegear, Aces and Eights, Hunter's Fury, etc).
Named Items (Fox’s Prayer, The White Death, The Hollow Man, etc).
Exotics (Liberty, Diamondback, "Tardigrade" Armor System, etc).
Skill Variants (Explosive Sticky Bomb, Incinerator Turret, Defender Drone, etc).
Specialization Weapons (Survivalist Crossbow, K8-JetStream Flamethrower, P-017 Missile Launcher, etc).

By my count, this means there are currently 249 weapons (including named and exotics as separate items), 23 brand sets + improvised (probably treated as a single brand set), 13 gear sets, 44 named/exotic gear pieces, 41 skill variants, and 6 specialisation weapons. So that's 377 individual proficiencies to gain and 10 levels of each, giving a total of 3770 possible proficiency levels to gain. That's obviously not counting any new items added in the new season.

Now, proficiency levels go up to 10 for each item, and no item can be upgraded until its proficiency level has actually reached 10. As far as "which items can be improved" then, the individual proficiency levels are simply progress markers and offer no benefit until you reach 10. They do offer a benefit to your Expertise level, but we'll get to that later. It's also important to note that once a profiency has reached 10, it cannot be improved any further. This means that as soon as you can improve an item (even by only 1%), continuing to use that item will not increase the amount by which you can improve it! (Except by going down the donation route, explained later.)

Expertise controls how far you can improve eligible items, and is increased by gaining proficiency ranks for any proficiency. At this stage, we do not know how many proficiency ranks are required to gain a single expertise rank (very likely the devs want to do further testing before deciding on the best number), which means we do not know what the current maximum number of improvements you could apply to an item actually is.

Examples
For this reason, if one is focused primarily on improving items by use rather than by donation, we get the following results (for the sake of the example, I'm going to assume each expertise rank costs 100 proficiency ranks):

A standard high-end loadout likely has 10 different proficiency classes equipped (1 specialisation weapon, 3 regular weapons, 4 different brand set/named/exotic items (assuming 3pc of one brand or 2pc of two), and two skill variants.
By playing enough with this build equipped, one would therefore achieve max proficiency in 10 proficiencies, giving 100 proficiency points total towards your expertise. If that equals one expertise rank, you could then upgrade all 10 items/skills by 1%. However, continuing to play with these items will not increase your proficiencies any further, as they're all now max level. You could farm items & materials for donation and improve other proficiencies that way, but the proficiency xp you earn in that time by playing will all be wasted.

If you then equipped a completely different set (different items, skills & specialisation) you would lose that 1% bonus, but you could earn another 100 proficiency ranks - and once you're there, you can upgrade both your new loadout and your previous one by another 1%.

Perversely then, the fastest way to gain the most improvement in your gear is by always using items you can't improve yet, until you finally gain all proficiencies and can go back to your favourite loadout, suddenly gaining a huge power increase in the process. In practice of course, you'd be using donations to improve whatever items you least want to play with, and perhaps mixing your improved gear into your loadouts to help 'carry' the weaker, unimproved items. Rather than the usual 'Jack of all trades, master of none' setup, here we have to be jack of all trades before we can be a master of any of them!

Observations
One little piece of phrasing about leveling proficiencies through gameplay caught my eye:

Using the items in the field and getting kill XP with them

Note that it specifies kill XP here, presumably to exclude the 'activity complete' xp that makes up the vast majority of a players xp gains. I would guess that this is to prevent players exploiting the system by playing most of a mission/activity with their strong, upgraded gear then switching to an unleveled loadout to get the completion xp. However, this does raise the question of whether DPS builds in co-op will level their proficiencies far faster than support builds such as healers, tanks and CC?

The donations system will be great in finally giving us a use for all the junk we find (for longer-term players who have already maxed the builds they want, 99% of drops get ignored or trashed), however I'm thinking it will make exotics much more difficult to level up in this way, as their drops are far rarer. It'll also likely see an end to a lot of the exotic sharing that currently goes on, which will hit newer raiders especially hard (I suspect quite a large majority of people got their first Eagle Bearer/ Ravenous as a share from a more experienced raider!)

Suggestions
I would personally like to see a progression system for proficiencies beyond level 10, so we're not effectively penalised for using these items. Perhaps a traditional xp curve on subsequent levels (so each level takes longer to earn than the previous) to gain item-specific points instead of contributing to the global expertise level (points that could be spent on specific handling attributes or similar, like Keeners Watch). For exotics, it would be good if improvements could somehow be made to their talents (which is what makes them so unique and useful) instead of just base damage/armor.

Looking over all the back-and-forth about HB and 'fun', I think a large part of the problem is how we view the difficulty options. It seems to be now that everyone expects Heroic to be the default endgame difficulty, and anything lower is some sort of sign of weakness. Almost all matchmaking is for Heroic as a result, which made sense before (towards the end of the 'Great Content Drought of 2021', when the majority of remaining players were long-timers with meta builds and lots of experience) but now, with the influx of new and returning players, I think we need to make an effort to push 'Challenging' as the standard difficulty, where players should go if they want to have fun. Heroic can then be recast as 'where people go when they really want to push themselves'. It's MEANT to be punishing (just look at the lack of checkpoints in missions), we just need to keep reminding ourselves of that, and check why we choose to play it.

To nerf Heroic (as HB effectively did before the fix) doesn't add anything to the game, it just subtracts something: the sense of challenge and achievement that keeps people coming back and playing long after the 'fun' of being OP has worn off.

One argument in favour of keeping them raid locked which I haven't seen anyone put forward yet is this:

Many players are instinctively reluctant to try the raids due to (their idea of) the pressure & stress of demanding team gameplay (eg fear of judgement of others, letting others down, etc). These players would obviously happily avoid raids and get the exotics elsewhere if they could. However, there's a very real possibility that when they DO give in and get involved in the raids (spurred on by the exclusives) they might actually find significant enjoyment in doing so, and maybe even become keen raiders themselves. After all, group play can as easily be a hugely positive, affirmative experience as it can be a negative one. Such players would never discover this without the attraction of those exclusives.

That said, my personal taste would be to limit raid exlusives to apparel & appearance mods - albeit preferably very desirable ones to maintain as much attraction as possible. Just not anything that has actual gameplay value.

Changing your world difficulty will not reset your manhunt progress, so feel free to lower it for this mission.

However, enemies will keep on spawning during certain parts of the mission regardless of difficulty: you should ignore them as much as possible and focus on the objectives (eg disabling the emp things), only kill enemies if they are getting right in your way, or once your objective changes to 'clear remaining enemies'.

It helps a lot to have a fairly tanky build (good armor and regen) to help you ignore all the incoming damage while you get those objectives done.

Update: Upon re-running the mission in order to record the issue, I found that this time the game saved my progress after each disconnect, so I was able to complete the mission. I've attached the video for reference, the four disconnects occur at approximate timestamps of:
00:39, 02:46, 03:11 and 05:11.

The timing of the disconnects seemed to be pretty random, not linked to any specific event.

Have to disagree with some of the posters in this thread about the usefulness of snipers - I play sniper more than anything else, in open world and in missions, solo and grouped, and it works just fine even with all directives on (I don't generally lag behind on kills or damage in group missions, unless everyone else is full DPS and mow everything down before I can aim...) With a high-end build with Mantis and headhunter as the key elements, your decoy gets reset every time you get a headshot kill so it doesn't matter that it doesn't last very long, you can just keep throwing them out. I do usually run an AR as a secondary for situations where I need to put out some consistent damage and don't have any suitable enemies to build headhunter stacks from (eg to get the helmet off a heavy) or for the occasional rusher when I can't seem to line up a headshot fast enough.

Needless to say, you DO need consistently good aim, or you'll end up doing little to no damage. It's very much an all-or-nothing playstyle, and if I'm having an 'off day' I'll just switch to an AR or skill build instead. It's also worth noting that enemies generally make easier targets solo than when grouped - they tend to move a lot more erratically when being shot by others, when it's just you they don't appreciate the danger until its too late, and are much more likely to stand still or run in nice straight lines for you

I do agree that HF/sniper is not a particularly useful combo though, you're essentially picking two conflicting playstyles and will only ever get the use of half your gear - pretty much the opposite of synergy.

I assume the OP and As1r0nimo were referring to this snippet:

At the end of Season 7 in December, we will be organizing an in-game event, reintroducing some player-favourite global events, associated with new rewards and an apparel event. We will reveal more information about this event closer to its availability date.

This is presumably supposed to cover the December and January 'dead time' in-between seasons. They haven't specified whether the current season will be bought to a close before these events, or whether it will be continuing in the background (until Feb) to allow newer players to get their season rewards & do the manhunt, so it's anyone's guess as to what happens next. I had expected them to say something by now, but they haven't - and impatience in circumstances like this serves no valid purpose whatsoever. Best to just stop anticipating, focus on other games, but keep an eye out for when they finally decide to say something. If they're delaying it, it's presumably because the alternative is a buggy mess, and none of us want that.

@NorthernSarcasm Check the difficulty requirements: a number of the field research objectives require them to be done at a specific minimum difficulty (often Hard). Note that this requirement is only displayed when you view the objectives at the specialisation desk, and doesn't show on the UI task-tracker thing.

@Battle-Phil yup, you still have to farm the specialization points on any new character. Activate the specialization you want, then earn points by doing daily & weekly projects, bounties, invasions etc or just by gaining levels.

@thenewmalcolm this exotic might require you to get the blueprint before it enters your loot pool: the blueprint can be obtained by completing the Captain Lewis manhunt mission (requires unlocking by reaching season level 53 iirc).

To activate the manhunt mission once unlocked, head to Jefferson Trade Centre, open the mission options and look for the 'toggle mission mode' prompt.

@Ubi-Massilia Unfortunately the work around I described above no longer works, the last patch (which fixed the Pentagon Breach problem) also made Darpa inaccessible to anyone who hasn't unlocked it previously.

The mission can still be run by joining another player, but can no longer be permanently unlocked.

I recently tested if running Pentagon on Story difficulty at lvl 30 WT5 would successfully unlock Darpa (as it did prior to the last patch), and can confirm that it does not. This means that, to my knowledge, there is now no workaround to unlocking Darpa on a bugged character at all.

It can still be run in another player's session, of course, but that will not unlock it permanently for the bugged character.

@Cyber_Skull_PT The damage stacks used to be multiplicative with each other instead of additive - meaning at 100 stacks you actually got ~170% amplified damage (instead of 100% as it is now), this made it substantially stronger in pve than any other build, which is why it got nerfed.

@CeddyRuxpin If you read the old threads then you already know all the arguments for and against the nerf, there's really nothing more to say and neither side is going to convince the other that their arguments are better. The devs made their choice, and resurrecting that whole debate will achieve nothing but raising tempers.

@MrAATX50 ok thanks - my suggestion was based on how (I think) it used to work before the bug first appeared, so figured that would be the first thing to try. The next would be to check if you can still unlock Darpa by running Pentagon on story difficulty at level 30, but obviously that requires finding someone with a level 30 map.

@MrAATX50 Did you run Pentagon on Story difficulty? Once you complete it and clear the final area you need to follow the prompts through a side door and down an elevator which should take you to the beginning of Darpa. I'm not sure if it works on higher difficulties.

@Dedsixx Do you have the Warlords dlc? Have you completed it and reached level 40?

(Newer gearsets and brandsets like Ceska require the dlc. Theoretically so does Countdown, but for some reason it's possible to matchmake into Countdown at level 30.)

@JoeTanker67 When the monorail is already there and no enemies get off it, that's a separate bug, and one that has been around for years. It seems to be caused by restarting the mission after getting on the monorail (whether you complete the whole mission or not). The workaround in that case is to travel to DC and back between each run of the mission.