jcm.8921

They should have just added like a 1s ICD to the carapace generation portion of it, or a limit to how many carapace stacks it can generate within a period of time. The main problem with Harbinger's synergy with Death Magic IMO is the ease and speed at which it can generate carapace stacks, primarily from this trait and Harbinger's ability to rapidly apply conditions.

Reading this thread, I'm honestly convinced that them fixing the UI should be the first thing that they do to Harbinger's defensive capabilities, before any nerfs.

IMO, a number of things really need to happen: 1. The LFG needs to show an indicator of roughly how populated a map is, so that both players and commanders are better able to gauge when they should start filling up an overflow map. It doesn't need to be a straight number, it can be something as simple as just a little icon that's blue when the map's fresh and underpopulated, green when the map's got a decent population that's under the soft cap, yellow/orange when the map's hit the soft cap and the game will attempt to spin up a new map, and red when the map's hit the hard cap and will no longer accept additional players. 2. There needs to be a way to view the actual map ID rather than trying to represent it via the last octet/number of the IP address of the server you're currently connected to, and there should be a way for players to select which map they're currently in, to give players more control over which maps they end up in, and to minimise the chance that a player ends up in a legitimately dead map when they're trying to participate in a meta. I don't buy the whole "but doing so would make team work redundant because everybody would join the same map" aspect, because what we have now is just a significantly less convenient version of this with individual maps listed in the LFG, so this should have similar behaviour, except it would make it significantly easier for people to spread out across instances. 3. There needs to be a way to queue for a full map, on the off chance that you really want to get into that map, ie the rest of your guild or party is in there and you want to join. This is Guild Wars 2, we shouldn't be outright preventing friends and guild mates from running events, metas, bounties, etc together, they should 100% be able to do open world content together, and this is arguably the best way of allowing that. Even with this change, I still don't buy the whole "but doing so would make team work redundant because everybody would join the same map" aspect, especially if the above two changes are implemented alongside this. If this were to function similarly to WvW's map queuing system, players should still be able to enter and play in less populated maps while queuing for the map they really want to play in, and when a spot does open up in that other map, they should be given a choice to either move over to that other map or stay in their current map and forfeit their spot to someone else in the queue.

IMO, the follow-up attack on GS4 should be a bit faster. Just feels super clunky to block and absorb a ton of attacks, then hit it again only to have it be easily dodged or side-stepped due to it being super slow.

Honestly wouldn't even mind this, would probably even allow ANet to buff the skills a tad, since they wouldn't have to worry about them overshadowing another legend or throwing balance out of whack when paired with another legend.

Have to agree with everyone else who's expressed confusion over the identity of Vindicator, ie why it's a hybrid DPS/support spec and not just a straight power DPS spec. Hell, I'd even settle for still having a more hybrid straight power DPS that annihilates everything in sight on Archemorus, with a more aggressively defensive brawler that deals less damage but can support themselves and allies while still remaining aggressive in the thick of things on Saint Viktor. Would seriously hate to see GS be wasted on such a hybrid spec, if Vindicator releases like this.

Playing with it a bit, I have to agree with others that the Alliance legend is in a really bad spot, and simply going with the tried and true Shiro/Jalis combo feels better. I like the duality aspect of the Alliance stance as a concept, but in its current iteration I find it to be extremely clunky. The flip mechanic on the Alliance skills really break up your toolkit in a way that does not feel good at all, and the only way to remedy that is to eat the energy and cooldown cost to flip things back either individually (very painful) or through Alliance Tactics (RIP energy and cooldown), which in itself has issues that, again, others have stated, in that the F2 feels extremely mediocre on other legends. Removing the cooldown when it flips would immediately made it better, but it still leaves you having to use the opposing flip to get back to the flip you really want, on top of the fact that the F2 is still pretty meh on other legends. Because of this, I'd actually like to see the duality nature of the Alliance legend become an overarching mechanic on a new F3, that not only switches which skills you have available in the Alliance legend, but also interacts with a new F2, that ideally would have different effects based on which side of the Alliance is currently "channeled" via the F3. Archemorus could give the F2 an offensive effect, while Saint Viktor could give it a defensive effect, with the specific effect changing based on the particular legend you're channeling via the F1. Aside from that, Vindicator is in a pretty decent spot and honestly just needs some tweaking. The greatsword kit overall feels pretty decent, with my only complaints being that I feel the GS 2 isn't as impactful as it should be, the GS 3 definitely has a range bug or quirk with an untargeted dash, and the GS 5 also feels pretty unsatisfying. I like the idea of the dodge, too, but I do feel that the recharge is way too slow, even with all the endurance and vigor Vindicator pumps out, and I definitely dislike that aftercast animation, it needs to not lock you in place when you're coming back down. There's also too many traits that focus on the dodge, and overall I feel that Vindicator needs some improvements in the power damage department, as I was led to believe that this was meant to be the power DPS spec of Revenant. Artwork and sound design needs some work, too, but I'll chalk that up to this just being a beta preview, and I'll trust that the artwork and sound design are improved closer to release.

Just because I've seen ESO thrown around a few times now (someone outright asked how expensive it is), I'll give a breakdown of ESO's cost as a former player of around 3.5-4 years who's sunk around 4k hours into it. ESO is a buy-to-play game with a quarterly release cycle (Q1 - dungeon DLC, Q2 - expansion, Q3 - dungeon DLC, Q4 - story/zone DLC), with paid expansions, and DLCs that can be either accessed by paying for them upfront using the in-game premium currency, Crowns, or by subscribing through the optional ESO+ subscription to gain access to the DLC during the subscription, on top of receiving a bit over the same amount of Crowns as the subscription costs. The base game costs US$20 from the official ESO storefront, and offers the first expansion, the Morrowind chapter, bundled in. However, there's also an all-expansions-in-one bundle, the ESO Collection: Blackwood edition, that contains all 5 expansions (Blackwood, Greymoor, Elsweyr, Summerset, Morrowind) for US$60. To make the comparison fair, we'll assume a new player wants to get everything the game currently has to offer, so they'll choose the ESO Collection: Blackwood edition, for US$60. ESO also has 17 individual DLCs as of this moment, which, if you were to purchase the 3 currently available DLC bundles, would cost exactly 24000 Crowns. Because there's an optional subscription involved, our new player essentially has three options here: They could spend US$175 to purchase 24000 Crowns through the 21000 Crown pack (worth US$150) and the 3000 Crown pack (worth US$25) to ultimately purchase all currently released DLC upfront to permanently unlock them, bringing the total to US$235, or US$60 for the base game + all expansions, and US$175 for all DLCs. They could subscribe for around 15 months, as the ESO+ subscription gives 1650 Crowns for each month you've subscribed, meaning that by spending US$15/month for 15 months, or US$225 in total, they could permanently purchase all currently released 17 individual DLCs and finally let their subscription lapse, bringing the total to US$285. They could just subscribe in perpetuity, which essentially means that ESO becomes a subscription-based game for them, costing them US$60 as an entry fee, and US$15/month to maintain access to all currently released and future DLCs. Note the italicised parts (currently released and future DLCs), because the 24000 Crowns figure it based on all currently released DLC, but by Q1 next year, another DLC will have released, which will add another probably 1500 Crowns onto the total. This isn't the case in GW2, as you can unlock GW2's DLCs (Living World episodes) for free, provided you log in during the claim period after the episode drops. Sure, ANet should absolutely make the need to purchase Living World content separately more obvious, and an all-in-one bundle probably wouldn't hurt, but GW2 is insanely cheap for what it offers, essentially having a US$50 all-expansions-in-one bundle, plus US$55 for LWS2, LWS3, LWS4 and IBS, bringing the total to just US$105, assuming you can't claim any episodes during the current Return To Living World promotion, you didn't purchase the EoD Collection Ultimate Edition to get the 4000 gems (this'd probably be more expensive, regardless, if you can't convert gold to the remaining 160 gems needed for all the LW bundles), or you can't purchase the expansions from an official reseller that is able to offer them at a discount. No monthly sub at all, base game is completely free, they give out the Living World episodes completely for free provided you're actively playing during their release, all they ask is US$105 for all story and map content currently in the game, and that's it, anything else you spend money on is convenience.

That's why it's considered to be a beta, because these specs are unfinished and ANet is wanting feedback on them, as well as wanting to give players a preview of them ahead of time, to keep the hypetrain chugging along by building anticipation for the specs that players enjoyed trying out. Both Heart of Thorns and Path of Fire had this same beta preview of their new elite specs, so this isn't anything new, ANet's just continuing tradition, here.

Rev takes a bit of time to get used to the energy mechanic, because of the exact thing you've ran into: you don't have enough energy to go full throttle all the time. You need to conserve your energy and, assuming you have Charged Mists selected from the Invocation trait line, swap legends at or below 10% to gain more energy in your swapped-to legend (specifically, you'll have 75% energy when you swap into your other legend), so you have more energy to go full throttle in your swapped-to legend for longer. Rev used to be a class I sucked on, because I wasn't used to the energy mechanic and wasn't used to knowing when to conserve vs when to go all out, but as I played it more (specifically, as I leveled it naturally, in general I feel leveling naturally eases you into each class's mechanics very well) I became more comfortable with it, and got the hang of the ebb and flow of the class. In general, condi Rev has a harder time managing its energy well due to how reliant you are on legend swapping and your weapon skills, but power Rev is pretty easy provided you're running sword/x, as the sword auto chain deals pretty good damage and consumes no energy, so if you're energy starved and your legend swap is on cooldown, then you can just spam autos and cast your other skills whenever you have the energy, then shortly before your legend swap comes off cooldown, dump all your excess energy, swap, and repeat.

They probably wouldn't need to. It'd only be the reference implementation that requires DX11/DX12/Vulkan, ie the "I don't want to write this myself, I just want to grab what AMD has, throw it in and get it working" implementation. The actual shaders and core concept probably work on older versions of these APIs (definitely OpenGL, there's only a few things that Vulkan has over modern OpenGL, and none of them would keep you from implementing FSR in an OpenGL game), so, if a developer wanted to put the time and effort in, they could write it themselves for whatever API they're using. It's just a matter of if they want to write it themselves.

Believe it or not, FSR might actually make performance worse, not better. FSR helps performance by allowing the game to run at a lower resolution internally, without losing as much image quality as if it were to use a "dumb" upscaler like bilinear filtering (which is often the default upscaler), similar to setting the game's render sampling to 'Subsample', just with a better upscaling method. Doing so moves some load off the GPU, allowing the GPU to render images faster, while also putting more load on the CPU, as the CPU has to give commands to the GPU faster, for the GPU to keep busy. This typically isn't a problem with a fast CPU and a game that is able to efficiently utilise the CPU, but if you don't have either of those things (or worse, if you don't have both), then the CPU can hold the GPU back, preventing it from pushing out as many frames as it could otherwise push out, leading to a CPU bottleneck. Unfortunately, GW2 is one of those games that makes horrible use of the CPU, to the point where if you have even a somewhat remotely balanced system, with a CPU and GPU that pair well together (ie the CPU isn't so much faster than the GPU, and vice versa), you'll pretty much always be hitting a CPU bottleneck. I'm on a system with a Ryzen 7 5800X and a 3090, and even at 1440p with the DX12 mod (which reduces CPU overhead) and render sampling set to 'Supersample' (meaning that, assuming 'Supersample' runs the game at 2x resolution, I'm running the game at 5120x2880, above 4K), I still really only see 60-70% GPU usage at absolute maximum, with a single CPU core hovering right around 80-90%: a classic CPU bottleneck. To even see the GPU start to become the bottleneck, you'd either need to run that game at a stupid resolution, or you'd need to be using such an old graphics card that any modern CPU would run rings around it, even a Pentium. That's how bad it is. Because of this, there's a 99.9% chance that FSR will result in worse performance, because it'll make the CPU bottleneck the vast majority of people have with GW2 that much worse. Even with the DX12 mod, it'd still probably make the CPU bottleneck worse, hurting performance.