[Poll] Should there be a Graphics update?

[Stormcrow.7513]

Start.GW3 with a modern engine. Problem solved.

 

However, since that won't happen i would love a graphic update. I use p912pxy and reshade now anyways since i can't handle the games innate blur.

[firedragon.8953]

Higher resolution textures (older content) would be nice.

 

But, I think the game actually still looks great. I wish the performance was better. If FPS could be improved by 30% that’d be pretty nice.

[mercury ranique.2170]

> @"MatyrGustav.6210" said:

> > @"kharmin.7683" said:

> > Pretty sure GW2 won't come to consoles because of the method by which their updates would have to be approved and such. There are several threads on that topic that can be found with the search function.

> >

>

> This article states it will be. Hopefully its real.

> https://www.justpushstart.com/2012/02/guild-wars-2-announced-for-consoles/#:~:text=Guild%20Wars%202%2C%20a%20sequel,to%20consoles%20in%20the%20future.

>

> > Any "future proofing" would probably entail a massive undertaking. We've already been told how convoluted and cumbersome the existing code is. To future proof the game, they might need to start all over rather than try to fix the antiquated code. I wouldn't expect anything to really change in this aspect short of releasing GW3, which I don't see happening. /shrug

>

> Thats a bummer.

 

The linked article is from 2012, around the release date. I am not sure how serious it is, but later statements on the old forum has been very firm saying no.

[zealex.9410]

A better post processing or unbundling of what post processing does into their own options, an upgrade to vulkan and dx12 and finally some touch ups to models and the lighting system should also get nuked because its bad.

 

 

All that and a robust system that hides effects, mount skins and abilities of other players will breathe new life into the game.

[MatyrGustav.6210]

> @"zealex.9410" said:

> A better post processing or unbundling of what post processing does into their own options, an upgrade to vulkan and dx12 and finally some touch ups to models and the lighting system should also get nuked because its bad.

>

>

> All that and a robust system that hides effects, mount skins and abilities of other players will breathe new life into the game.

 

Yes i would like better lighting. At times i feel like the game has a Soft & Dreamy glow effect that i want gone. Lol.

 

I wish the addition of DX12 (lets hope) there was also a stylized low Texture optional mode to improve performance. Id hope it would be optimized for the landscape to look like Rime in a way.

 

[Tyncale.1629]

Using Reshade with Tekkits settings completely overhauled the look of the game for me. It is crisp, sharp, beautiful colors and far away vista's and none of that glowy, blurry bloomy shit anymore.

[SweetPotato.7456]

voted other,

 

I cannot afford new computer.

[maddoctor.2738]

A graphics update isn't needed, as in upgrade of assets, models and textures, at least not for current maps, old maps could certainly use an overhaul though. A graphics engine update is another story. I voted yes because an engine overhaul is badly needed, the game runs really badly and an overhaul of the engine can make it run much faster, as we already see with dx12py. The problem with dx12py is it can only help in places where the main performance penalty is outside the main thread of the game, where most of the calculations happen. When the main thread becomes the bottleneck, then dx12py doesn't do anything.

 

As for those with older systems, an upgrade in performance will either help them too (based on exactly how old it is) or at the very least not negatively affect their performance. In other words, it cannot get any worse, any performance improvements will only make the game better.

[TheQuickFox.3826]

A native DX11/DX12/Vulkan renderer would be nice. [DX12 for Windows 7 is also a possibility](https://devblogs.microsoft.com/directx/porting-directx-12-games-to-windows-7/), just like World of Warcraft has. For me and the rest of the Windows 7 gamers it is important that we at least can continue to play GW2 on this OS.

[Astralporing.1957]

> @"Danikat.8537" said:

> Can someone explain for me what difference upgrading to DX12 would make if they don't change the graphics? Would it just change some of the lighting effects and shaders?

People that ask for dx12 usually think it will result in visibly better performance. It's doubtful that the actual effects would be even close to what they hope for, but hey, people can dream.

 

But as for answer to the OP's question: There are different kinds of "should", and different kinds of graphics updates.

 

As far as quality of graphics goes, i think GW2 is in a good place and doesn't need any upgrades. It might use some more display options so, for example, people can choose the amount of sfx spam they want to see better, but that's not a change to graphics itself.

 

Now, what might be really nice is some kinds of _performance_ upgrades - although, again, those might not necessarily be graphics-based. Most bottlenecks supposedly are on a much deeper level, in the core of the game engine (and possibly in the client-server communication routines).

 

The real question however always is - what would we need to sacrifice, and for what level of improvement. Engine upgrades are generally a complicated matter, requiring a lot of time and resources. Resources that would then not be used for other things.

 

For example, i'd really not want to hear that sure, the engine update happened, but we had to sacrifice the whole next expac for it. And, considering how LS5's quality went down since they started working on expac, which suggests they don;t have enough resources now to work on two major projects at the same time, it's quite likely that the engine update might actually cost us far more than just the next expansion.

 

So, would we really want _that_?

 

[knite.1542]

I'm more concerned with the game performance. I won't pretend to know/understand why the game performs the way it does, but I will say that the performance I get compared to the hardware I have is terrible.

[Cynder.2509]

It would be nice if they implemented DX12 support in their engine to make the game run better on newer hardware. I mean WoW got an entire engine update too so why not GW2? I know there's a DX12 support mod out there but I couldn't get it to work so far as i use GW2Hook and they're not compatible. I think they could bake this technology within their own engine. And Yes, I know that the game relies on a heavily modified version of the GW1 engine and that it would take a lot of money, recources and time for that but they could at least use a bit of the d912pxy mod. I mean even ESO runs better at some parts for me than GW2 and even ESO has some problems for me at the moment. ESO used to run on ultra at 100 to 140 fps for me when GW2 used to run at 150 to 200 fps at highest settings but both ESO and GW2 seemed to have lost performance with every new update. Especially PoF , LSS4 and Drizzlewood Coast run terrible for me now considering to how they ran at day 1 and considering that other games like Monster Hunter World for example runs better at ultra for me. I can even get Cyberpunk 2077 get to work as it seems with my Ryzen 5 3600 and 1660 ti (Yes, I'm on a budget of 1000€ as I'm not a rich spoiled Twitch streamer getting money stuffed in their butt for not doing any real hard work) better than GW2. I used to play GW2 on a very shitty non gaming PC from 2012 until 2018 and barely got like 10fps on itback then but it really pains me to see how well other games run better than GW2 used to run at firt before every update. I had a FX 6300 and a GTX 1050 before I upgraded to my current system in January and even on my previous system (which was an upgrade from my shitty non gamer PC) GW2 ran better before PoF was released. Also, yes, I'm aware that MMOs are much more dependent on other things than singleplayer games but that doesn't mean that this game couldn't need an overall overhaul. Don't get me wrong, I have no problem with graphics since I grew up playing PS1 and PS2 games but still things could be better in terms of performance.

TLDR: This game needs better optimisation for newer hardware.

[Zephire.8049]

Visually the game is fine, imo. Performance, however, is a different matter. From what devs have said and implied in the past, doing a visual overhaul would be not much different than an engine overhaul when it comes to the about of work needed, so it would be better to to an engine overhaul rather than focus just on graphics. With a new engine, a graphics update is likely to happen since they're already in the guts of the game meaning everyone would benefit from it. This includes devs who are limited in what they can and can't do due to the age and jerry rigging of the current engine.

 

A graphics update that's only a graphics update would be a bandaid that wouldn't actually fix server issues, broken events, lag, etc. It may or may not be a prettier mess, but it would still be a mess.

[Infusion.7149]

I'd rather better multithreading , though DX12 would probably help a bit for Windows systems it won't help overall for anything that is not the render thread. It's heavy on one thread due to the sequential nature of some operations and [Amdahl's law](https://en.wikipedia.org/wiki/Amdahl's_law) inevitably applies. Something akin to the introduction of the 64 bit client would be a huge step in the right direction (prior to this people would crash when hitting a combined 4GB RAM+VRAM limit).

Some stats: Most people are on quad cores by now even if they play on laptops, which isn't the case when the game launched ; Steam's hardware survey has 91.07% DX12 capable GPUs and the most common cards are Pascal era low midtier cards (GTX 1060 / GTX 1050 Ti) and on the AMD side it's Polaris type RX 580 / RX 570 cards. Meanwhile 45.08% have 4 cores and 25.70% have 6 cores. 89.28% are on Windows 10 64 bit.

 

> @"TheQuickFox.3826" said:

> A native DX11/DX12/Vulkan renderer would be nice. [DX12 for Windows 7 is also a possibility](https://devblogs.microsoft.com/directx/porting-directx-12-games-to-windows-7/), just like World of Warcraft has. For me and the rest of the Windows 7 gamers it is important that we at least can continue to play GW2 on this OS.

Linux users tend to use DXVK so d912pxy isn't a necessity there.

 

Ultimately what needs to be optimized is the multi-threading that isn't addressed by DXVK / d912pxy. Would be helpful if shadows and reflections are able to offload to GPU as well.

[Dawdler.8521]

> @"Cynder.2509" said:

> I mean WoW got an entire engine update too so why not GW2?

And for everyone asking "will it make the game look better", did that make WoW look better?

 

No it still look like bloody WoW.

[zealex.9410]

> @"Dawdler.8521" said:

> > @"Cynder.2509" said:

> > I mean WoW got an entire engine update too so why not GW2?

> And for everyone asking "will it make the game look better", did that make WoW look better?

>

> No it still look like bloody WoW.

 

It certainly made wow run better tho.

 

As for visuals concerned, a big part of why the game can look bad is the visual clutter for abilities and such.

 

Visually the game looks fine but most ppl can even up the graohics options because if you do that and then go play wvw or a meta your pc will melt.

[maddoctor.2738]

> @"Astralporing.1957" said:

> The real question however always is - what would we need to sacrifice, and for what level of improvement. Engine upgrades are generally a complicated matter, requiring a lot of time and resources. Resources that would then not be used for other things.

 

It's not like the writers, composers, asset creators and programmers are gonna be affected by an engine performance optimization. If anything, it's gonna make the work of the artists faster, it's puzzling to this day why pre Draconis Mons maps use a different rendering solution than maps released after it. This causes the asset creators to make sure their creations look well on both systems, and we've seen cases of a dye or a skin looking good on one, but horrible on the other. Streamlining the game to use one solution for all maps, either a completely new/overhauled one, or at the very least the one used after Draconis Mons, would certainly help improve the game and cut down on resources and time needed to create new assets.

[Astralporing.1957]

> @"maddoctor.2738" said:

> > @"Astralporing.1957" said:

> > The real question however always is - what would we need to sacrifice, and for what level of improvement. Engine upgrades are generally a complicated matter, requiring a lot of time and resources. Resources that would then not be used for other things.

>

> It's not like the writers, composers, asset creators and programmers are gonna be affected by an engine performance optimization. If anything, it's gonna make the work of the artists faster, it's puzzling to this day why pre Draconis Mons maps use a different rendering solution than maps released after it. This causes the asset creators to make sure their creations look well on both systems, and we've seen cases of a dye or a skin looking good on one, but horrible on the other. Streamlining the game to use one solution for all maps, either a completely new/overhauled one, or at the very least the one used after Draconis Mons, would certainly help improve the game and cut down on resources and time needed to create new assets.

You're right on writers and composers - i doubt they are the ones that decide the release schedule, though. You're not necessarily right about asset creators and other devs however. Remember that the reason why it took them a while to bring back SAB and Queen's Pavillon was that the original instances were no longer compatible with the engine and they had to adjust them first. Now, think about making a new LS episode while also working on that rework for _all_ currently existing maps. Not that you'd be able to make new stuff either, until the changes were finalized to at least some degree.

 

Also, again, we're likely not even talking about the rendering engine - we're probably talking about underlying core threads that call upon that engine, and about the client-network communication.

 

It's not the car chassis that needs a rework, it's what's under it that does. And here we are talking about adding some spoilers and paint job.

[maddoctor.2738]

> @"Astralporing.1957" said:

> > @"maddoctor.2738" said:

> > > @"Astralporing.1957" said:

> > > The real question however always is - what would we need to sacrifice, and for what level of improvement. Engine upgrades are generally a complicated matter, requiring a lot of time and resources. Resources that would then not be used for other things.

> >

> > It's not like the writers, composers, asset creators and programmers are gonna be affected by an engine performance optimization. If anything, it's gonna make the work of the artists faster, it's puzzling to this day why pre Draconis Mons maps use a different rendering solution than maps released after it. This causes the asset creators to make sure their creations look well on both systems, and we've seen cases of a dye or a skin looking good on one, but horrible on the other. Streamlining the game to use one solution for all maps, either a completely new/overhauled one, or at the very least the one used after Draconis Mons, would certainly help improve the game and cut down on resources and time needed to create new assets.

> You're right on writers and composers - i doubt they are the ones that decide the release schedule, though. You're not necessarily right about asset creators and other devs however. Remember that the reason why it took them a while to bring back SAB and Queen's Pavillon was that the original instances were no longer compatible with the engine and they had to adjust them first. Now, think about making a new LS episode while also working on that rework for _all_ currently existing maps. Not that you'd be able to make new stuff either, until the changes were finalized to at least some degree.

>

> Also, again, we're likely not even talking about the rendering engine - we're probably talking about underlying core threads that call upon that engine, and about the client-network communication.

>

> It's not the car chassis that needs a rework, it's what's under it that does. And here we are talking about adding some spoilers and paint job.

 

Queen's Pavilion and SAB were affected by changes to how jumping works (physics?) it wasn't about graphic performance. When they changed the graphics engine with Draconis Mons it didn't cause them to rewrite all assets from scratch either. In the end it depends on which part of the engine they are gonna tweak to increase performance. We are talking about graphics here, if they finally add a proper way to handle reflections and shadows, the current system is heavily taxing the CPU on these two effects where they could be done on the GPU exclusively, it won't affect jumping or skill usage, the rest of the game should be fine.

[Astralporing.1957]

> @"maddoctor.2738" said:

> Queen's Pavilion and SAB were affected by changes to how jumping works (physics?) it wasn't about graphic performance. When they changed the graphics engine with Draconis Mons it didn't cause them to rewrite all assets from scratch either. In the end it depends on which part of the engine they are gonna tweak to increase performance. We are talking about graphics here, if they finally add a proper way to handle reflections and shadows, the current system is heavily taxing the CPU on these two effects where they could be done on the GPU exclusively, it won't affect jumping or skill usage, the rest of the game should be fine.

It's not the reflections or shadow that are the main problem here. Sure, they can result in a performance hit, but that's on the top of the real problems. The stuff you speak of is clientwise, while the real issue lies in what's happening on the _servers_.

The first clear sign of problems to come was the Karka southsun event where most of the mobs weren't taxing your computer at all, because they weren't even visible. Then came the WvW Stonemist fights, where, again, the game was often unable to even show most of the participants, and thus you ended up being killed by people you never even saw. All those things are _not_ graphic engine problems. They are game engine problems that seem to be buried much more deeply into the code. The stuff that needs to be optimized (and reworked to make use of multiple threads, so it will be able to make use of multiple cores way better) seems to lie at the very basis of the game's programming, long before graphic calls are even taken into consideration.

 

BTW, they seem to be actually working on all that, but it is a very, very slow process. Probably not only due to the sheer scale of it, but also to the fact that they have to do it on a working game, without disrupting it too much.

 

Yeah, some renderer upgrades might be nice, but i personally don't think they would help that much - and they would mostly be seen by the people with the high-end computers, so those that are in the _least_ need of performance fixes.

 

[Dawdler.8521]

> @"Astralporing.1957" said:

> The first clear sign of problems to come was the Karka southsun event where most of the mobs weren't taxing your computer at all, because they weren't even visible. Then came the WvW Stonemist fights, where, again, the game was often unable to even show most of the participants, and thus you ended up being killed by people you never even saw. All those things are _not_ graphic engine problems. They are game engine problems that seem to be buried much more deeply into the code. The stuff that needs to be optimized (and reworked to make use of multiple threads, so it will be able to make use of multiple cores way better) seems to lie at the very basis of the game's programming, long before graphic calls are even taken into consideration.

>

> BTW, they seem to be actually working on all that, but it is a very, very slow process. Probably not only due to the sheer scale of it, but also to the fact that they have to do it on a working game, without disrupting it too much.

... uh unless Southsun came out *before the game even released*, this was noticed in WvW fights first. Not just SM either, entire WvW during large fights.

 

And incidently, Anet also fixed that *many* years ago because it was a graphics engine issue - the engine was culling players way, way too much. This was during the time when Anet actually paid attention to WvW, listened to community feedback and worked to fix WvW issues, unlike today **WHEN WE STILL HAVE A FREAKIN INVISIBLE WALL ON DBL GARRI THATS BEEN BROKEN FOR LIKE 5 YEARS**.

 

But otherwise, I really have no idea what you are on about.

[SpellboundTutor.4257]

Do I think the game needs a graphics update? Maybe a little bit. Even on max settings, I can still see a bit of pixelation on the models...

 

...but I think the top priority is upgrading GW2 from DX9 to DX12, or even DX11. The game is getting harder and harder to run on DX9, I feel, and eventually something is going to break.

[maddoctor.2738]

> @"Astralporing.1957" said:

> It's not the reflections or shadow that are the main problem here.

 

When I'm outside Queensdale, looking towards the mountains, reflections and shadows are what is causing extremely high performance problems because both of those are CPU intensive in this game. And I can easily tell when I get sub 60 fps while my GPU is chilling (makes no noise). Putting both on the lowest settings causes my GPU fans to ramp up quickly, also increasing my overall FPS.

 

I also see a big improvement in general gameplay when activating dx12py. Sure in cases where the ONLY problem is the main thread, rendering performance plays little role, but in every other situation in the game, it does have a big impact. Neglecting performance optimizations, those reflections and shadows are using an ancient method of calculation that barely touches the GPU, just because "it wouldn't solve ALL performance problems" is just unacceptable to me. Any kind of performance gain IS a performance gain

 

Edit: in a very simple example, yes in mass open world encounters and in WVW, they might need optimizations on the server side to fix all the issues, but that's not all the game has to offer. Just chilling in my home instance can see massive performance benefits when tweaking settings, using dx12py, or both. Why wouldn't the company be interested in resolving those issues (that are exclusively graphic dependent) and instead put it on the back burner claiming "server/main thread issues"

[Tic Tac Toe.1482]

Yes, most people would want a full redo. And a new PC to run it all!

 

While that's not realistic, a few optimizations for performance, the main rendering thread, and multi core CPUs wouldn't hurt, it's a lot easier (and less expensive) said than done.

[Fenom.9457]

I don’t think the game really is in serious need of it but if anet has a chance to upgrade without hurting other things like the pace of content, then by all means! It certainly won’t hurt anything. I also appreciate whenever anet takes the time to upgrade old models or locations if they can justify it in the current story