doomagent13
-
Posts
74 -
Joined
-
Last visited
Posts posted by doomagent13
-
-
Just as a forewarning, I do not know the details of how cache optimization works. This is all based on fixing a problem I had, as well as observing the file system while optimization ran.
22.7.0 added an automatic trigger for cache optimization if enough space was wasted. It had been a while since I manually ran optimization, so it triggered. However, it got stuck at 37%. Repeatedly, for extended periods of time, and despite running Verify. Between Resource Monitor and going to the folder with the cache files, I figured out the issue. The way it performs the optimization involves more or less duplicating the cache file being optimized. It wouldn't be THAT big of an issue, except the largest cache file is OVER 14 GB. (To be specific, it is "F.TextureDx9.cache") At the time, I only had about 10 GB free hard drive space, so the optimization couldn't be finished. (It's a 250 GB partition on a 4-year old MacBook Pro.) I ended up uninstalling Source FilmMaker to free up enough space.
There are two problems that jump out at me. First, the launcher doesn't give any indication if it fails due to a lack of disk space. It just sits there, part way through the process. Not even the logfiles gave any indication of what the problem was. Second, it has to duplicate such a large file. I don't know if optimization could be changed to no longer rewrite the cache next to the original, but I would think the biggest cache file could be broken into 2 or 3 separate files. Just for comparison, the 5 largest cache files right after optimization weigh in at ~14.8 GB, ~2.7 GB, ~1.7 GB, ~1.1 GB, and ~350 MB. Given the current number of cache files, I am inclined to think 1 or 2 more would not have any significant impact on load times, while it could potentially make updates and optimization faster and/or more efficient.
-
If I recall correctly, it got up to a total of 25 different keys: 6 for T1-3, and 7 for T4. Getting a specific T1 or T2 key was fairly easy: just run an appropriate level endless mission long enough. A specific T3 key was a bit harder, due to having only a 50% chance between T2 and T3, but still doable. T4 keys were split in 2 groups, with a key from one of those 2 groups GUARANTEED after round 4 of interception; getting a specific T4 key was arguably the easiest kind of key to do so for. (Which group was determined by enemy level.)
It only took a substantial amount of work to get a specific T3 key, which was at least partially mitigated by building up a collection of other T3 keys for when you needed them. Sabotage keys may have seemed rarer due to higher demand due to lower supply because they did not exist as long as the other key mission types.
-
1 hour ago, Phatose said:
You should probably add "Every player requires a relic for a rewards, thus preventing leeching behavior from being rewarding" to your list of pros.
That's also a con: You need a matching relic for the fissure missions to be worth doing at all. (You need several for the endless missions.) With the obvious exception of the relic contents, and to some extent void traces, everything can be gotten elsewhere without the annoying fissure-spawning. It was usually pretty easy to quickly assemble a squad for key runs; casual relic runs are completely reliant on the matchmaker to get a squad at all.
-
To put it bluntly, the problem with the relic system isn't the relics. Having the correct relic is more useful than having the correct key was. The problem is getting the correct relic. If I recall correctly, the old system ended up with 25 keys, evenly distributed. (1 per Tower per mission type, except interception was only T4. There were 27 keys if you include ODS and ODD, but those are manufactured.) Furthermore, every now and then which key was the correct one could change. The relic system has no limit on the number of kinds of relics, or any reason for even distribution. Relics also do not change, meaning that every group of new primes comes in a set of completely new relics. Few, if any, primes have had more than one part in a single key/relic, with the possible exception of the different rotations in the endless keys. This means that for every new prime you need to get at least 3 or 4 new relics to even have a chance of getting that prime. Even if they were all common, you would almost certainly not get the correct part on the first run of each kind of relic. If I have the math and numbers right, a 4-man squad, all with the same unrefined relic, has less than a 70% chance of receiving a specific COMMON item. It would take 4 runs with such a squad to reach 99% chance for a specific COMMON drop. That means 4 identical relics per person. I can't check right now, and the wiki hasn't been updated, but someone said there was something like 11 currently available Axi relics. I don't even want to TRY doing the math for how many drops you would need to get in order to reach 50% chance of getting 4 of a specific Axi relic, let alone for 90% or 99%.
If someone wants to try doing that math, and then try to explain how this system is better than the old one, PLEASE do so...
-
I don't get the hate for T3S; it usually got fun within 15-20 minutes. (T1S rot C is a completely different story... Just reaching 20 minutes was hard enough; I don't think I ever went for 40...)
Relics do provide a better chance than non-endless keys did for a specific part, but are by no means better overall. (Endless keys, even for rot C, depend on how long you stay on average.) The problem is that, to have a guaranteed CHANCE at something, you must first acquire the correct relic. That mostly was not a problem with the keys. With the limited number of possible keys, the rewards shifted around enough that at least most keys remained moderately relevant. I can't say the same for most of the original relics, a decent number of which I still have 50+ of. Aside from the ones that have a Forma as a common drop, they are mostly useless. (Baro hasn't had enough new stuff for a while for any substantial ducat farming to be necessary, and Formas usually come up often enough without specifically picking relics for them.)
I can understand wanting people to not have everything they need to get a new prime immediately upon release, but always requiring the farming of a completely new set of relics goes too far. A better option might have been to initially leave one or two slots per relic undetermined, with a large and/or regularly shifting pool to pick from if it is rolled. This undetermined slot could later be filled with a specific part before eventually being retired, which could be used to reduce the number of new relics needed every time a new group of primes are released and/or vaulted.
-
I'm pretty sure a jump-kick can break containers, although hitting lower than head height can be tricky...
-
I think Nyx and Banshee are mostly fine.
For Zephyr, I'd say move the jump from the ground from Tail-Wind to Dive Bomb, and shorten the duration of Tornado in exchange for more of something else, but that's about it.
-
Try re-rolling the Riven until you get +Fire Rate and some form of +Damage... I forget exactly what my Riven mod's stats are, but my Vulklok has a Fire Rate of .459, with around 1k damage and both crit mods.
But yes, it being usable without a Riven would be nice...
-
4 hours ago, JHarlequin said:
As well as it takes a lot of weapon slots to have one of each type of weapon available at any time. Slots that aren't free. Weapon only modifiers are crap for this reason also.
Aren't there only 6 possible weapon type modifiers? (Melee, Secondary, Bow, Sniper, Shotgun, and Assault Rifle/everything else?) It's been a long time, but I vaguely recall either starting with 8 weapon slots or at least reaching 8 with the starting plat... By the time you reach sorties, you should be able to trade for plat anyway...
-
Whether or not old Bladestorm meets the definition of AoE, it was nothing like any of the other true AoE abilities. It auto-targeted every enemy in a certain area, and then SEQUENTIALLY killed them one by one. More or less EVERY other AoE ability SIMULTANEOUSLY affects every enemy within a certain area. Furthermore, they have some effect beyond pure damage -- CC and/or debuff to be specific. (Miasma has a short stun.) Bladestorm did not, and still does not, have any real CC aspect. (Stunning an enemy for the few seconds it takes to kill them is not CC.) To put it bluntly, Bladestorm was, and mostly still is, a no-risk auto-kill against almost anything, and nothing more. Aside from possibly getting ragdolled enemies stuck in awkward locations, no other abilities are anywhere near as reliable at killing high level enemies. Every other ability that can be reasonably certain to kill at high levels involves using the enemies' damage or health against them, and only Maim comes close to not having some CC aspect.
Requiring targeting for truly AoE abilities, beyond specifying the center/origin, makes absolutely no sense. Unless there are obstructions, an explosion cannot arbitrarily hit only some targets within its blast radius. (Shaped charges just make an explosion be distinctly non-spherical.)
As far as the power of the various CC abilities, they generally balance potency and duration. Stomp and Avalanche completely stop enemies, but do not last that long. Bastille completely stops enemies and lasts longer, but can only affect a limited number of enemies. Molecular Prime makes enemies take double damage and can last fairly long, but only slows down enemies. Chaos can last fairly long, but only makes enemies fight each other. Bladestorm (eventually) outright killed enemies, without any real risk or effort. (It made W+M1 seem quite skillful.) DE mildly increases the effort required for Bladestorm, and you want abilities that already involve some risk, and require moderate effort to take advantage of, to take even more effort? You MIGHT be able to present a vaguely reasonable argument if Bladestorm did not make Ash and his targets invulnerable, although its risk and effort requirements would still be much lower than other abilities. (Compare the survivability of a Kavat to that of a Kubrow.)
Bladestorm has LONG been an anomaly in the relative balance of powerful abilities. It is slightly less so now, but is still very much low-risk, high-reward for minimal effort.
Do keep in mind that some people primarily only use the powerful CC abilities when really necessary, such as to revive a teammate. Or would you rather they let you die because there isn't enough time to target all the nearby enemies?
-
Is it possible that the different faction bonus damages apply to just the Corrupted that had been that faction? Meaning, maybe bonus damage against Grineer also applies bonus damage to Corrupted Grineer, but not the other Corrupted.
-
Endo should be completely removed as a sortie reward, or at least scaled up to actually be vaguely comparable to the other rewards. 10k MIGHT be enough to not feel like a complete waste of a reward. The only crafting material that should even be remotely considered as a possible sortie reward is a Forma, and even that is questionable. Personally, I would cut the sortie reward possibilities down to:
Riven Mod
Greater Lens
Orokin Catalyst BP
Legendary Core
and MAYBE a completed Exilus Adapter.
As for Bladestorm, its victims need to not ignore external damage while being attacked. Not infrequently, I will be about to shoot and kill an enemy only for Bladestorm to come around to them, keeping them alive until Bladestorm kills them a few seconds later. And it is not just the Ash/Specter blocking the shot, explosives also fail.
-
That looks nice and all, but...
Shouldn't Valkyr Prime resemble Gersemi Valkyr? Valkyr Prime resembling normal Valkyr just does not work (chrono)logically. The Corpus would have had to have experimented on Valkyr before Valkyr Prime existed. The Orokin then would have had to intentionally include some of the Corpus changes. However, the Corpus experiments were overseen by Alad V, who is still alive. Furthermore, the non-Prime Warframes are at least implied to be mass-produced, slightly inferior versions of the Prime Warframes, which requires the Prime version to have come first. On top of all that, where would Gersemi Valkyr have come from?
From what we know so far from lore, the order of creation should be Valkyr Prime -> Gersemi Valkyr -> post-Corpus Valkyr. What can be seen of the appearance of Valkyr Prime suggests it was Gersemi Valkyr -> post-Corpus Valkyr -> Valkyr Prime...
-
Don't forget about the Vulkar's augment. An extra 60% damage after all other mods have been applied is nothing to laugh at.
-
A few times I've had this happen, only to find that the mods had switched to a previously empty config, leaving the one I had filled empty. I think it may have started around the time of the Halloween event...
-
Maybe it could just auto-auto-resolve if you have an exact duplicate of the mod in conflict that is not currently in use... Maybe it could even go so far as to not complain at all in such a situation...
-
Did the reactor alert, only to get back and find progress was not saved, before getting kicked back to the login screen. Oh well, it's not like I really needed yet another reactor lying around...
-
Congratulations DE, you killed the Vulklock. With 6.66 seconds between shots, I may as well put Sweeper Prime on my Diriga. It would probably be more effective, despite being a SHOTGUN used at up to sniping range... Diriga was the one companion that had really felt useful other than Carrier (Prime), and that was in part due to the Vulklock.
I guess I wasted the 6 Formas I used between Diriga and Vulklock. Oh well. Back to Carrier Prime...
-
Who says they don't already do that? We just don't have any vampires to test on...
-
I think the problem relates to how they made Antimatter Drops not charge each other. Antimatter Drop's explosion is radiation damage. Some limited testing strongly suggests that Antimatter Drop NO LONGER ABSORBS ANY RADIATION DAMAGE WHATSOEVER!!!
While I can understand not wanting Antimatter Drop to charge itself, doing it in such a crude manner WITHOUT ANY MENTION OR EXPLAINATION is downright ridiculous. In all seriousness, not absorbing any radiation damage is much more game-breaking than being able to charge off of other Antimatter Drop explosions.
It is quite possible that this is how it has been since 16.4.4 which stated:
Explosions from Antimatter Drop will no longer fuel the damage of other nearby AMD projectiles
Note the lack of any mention of ignoring all radiation damage...
-
Seeing as they already released one set of dual-stat mods for Archwing, I'd say it's a safe bet the rest will be coming, eventually.
-
Carrier due to Vacuum. Helios is the only other one I ever use, so it can scan stuff for me. Depending on how useful I find Detect Vulnerability to be once I get it, Helios might start seeing more use.
-
Normal Tornado is primarily useful for CC and procs... Unless the tornados do a much better job of chasing enemies, you need the pull-in to get the number of procs you want. It seems to me this augment would turn Tornado from what it is into a larger number of small, (very) slowly moving elemental damage fields, that enemies can probably avoid/pass through without too much trouble...
-
If it is reasonably easy to make them stack, this could be a great way to block a doorway or other appropriately sized passage for an extended period of time without needing to pay attention to it. That's about all I can think of where it would justify giving up a slot for it.
[22.20.5] Spores Revisited Feedback MEGATHREAD
in Warframes
Posted
Being able to recast Spore on a new group of enemies without removing it from a previous group is much better. To be honest, detonating all spores never really made sense when leaving them would definitely kill the victims. Needing to detonate the current batch of spores before you could use Spore on another group of enemies made even less sense.
Having it not spread at all when a spore kills seems likely to cause complaints of it "stealing" kills, preventing it from spreading to other enemies. If some non-Spore damage has been done to an enemy that dies to a spore, it should spread.
Unless I misunderstood something, any enemy hit by Spore is guaranteed to eventually die without any further interaction, unless the spores get removed by something. (With Nullifiers being the main, if not only, way for spores to be removed without spreading to other enemies.) Although your sentinel/companion would kill faster at anything but the lowest levels, this seems like a rather questionable situation, particularly for a first ability. The only ability I can think of that might come close to such a guarantee of a kill is Link, and even that comparison has issues. Every other ability requires some interaction and/or has a limit to the damage it can do.
To be honest, I think you've gone the wrong direction with Spore. It was a contagious debuff that was unlikely to be lethal unless you got it carrying toxic damage. Now, it does lots of damage, with a minor debuff that is mostly useless against 2 of the 3 main factions, and only somewhat useful against the 4th. My suggestion would be to get rid of the scaling damage, switch the damage type to toxic or viral, put back a duration, and have any popped spore spread all status effects the victim was suffering from. It may not have gotten very many kills, but seeing the constant viral statuses was rather satisfying.
Personally, I mostly liked the interaction between Saryn's abilities, although I can see how some players would have trouble understanding them. The only interaction that I found hard to work with was getting targets afflicted with both toxic and viral in order to maximize the damage from Miasma. I ended up primarily using Miasma for the brief stun, usually to revive someone. The interactions between Spore, Molt, and Toxic Lash were fairly straightforward and easy to use.