Guild Wars 2

1.) Yep, GW2 engine has heavy problems in dense populated areas like cities or WvW and AMD users have even more problems, no excuse here. They simply should have put more effort into optimizing this from the beta, since it makes WvW unplayable.

2.) The “OverPowered” class observation is a subjective and psychological phenomenon that happens in most competitive MMO games. The balance is probably not 100% perfect, but overall is working. If any class would be actually so much overpowered, given the easy nature of sPvP to reroll, every match would be consist of 80% of those “overpowered” classes, which is simply not the case.
- Jumping Puzzles, agree: Why they put PvE content in a “abstract” and isolated PvP gamemode/world is beyond me.

3.) All classes have good and valid range options, simply use them. The main problem i have is that i want to save and redo my traits inside a dungeon, so we can quickly adapt to changing situations.

4.) Yep, agree: GW2 economy/crafting system is badly designed and is basically a re-hash of all the older MMO’s, nothing new here on the base problems, except for massive crafting XP. Wish they would have analyzed EvE-Online or SWG better, for some fresh ideas on a stable and interesting economy.

5.) Yep also agree, except for the endgame exotic items nothing new here, same old crappy system than any other WoWish MMO. The main reasons u craft is for xp and the ability to craft the endgear items, which is pretty bad design indeed.

Maybe this helps: Think of GW2 in terms of investment and “fun” extracted. So for 40$ u get a pretty good game that will entertain u for at least 20-80+ hours, which is quite a good value. The main problem is that many buyers simply expected something “extraordinary”, which is simply not the case. So look at GW2 from a money/fun standpoint and u will realize it has a fantastic ratio. If u expected more than thats your problem for being a little naive and now disappointed.

Antiriad.7160:

everyone reporting “cpu bottleneck” seems to be using an nvida card.
Are there any reports of people getting poor fps on an AMD card actually?
Because there was one dood claiming he uses a Q6600 and AMD card and had great performance..

First its a “claim”, combined with vague formulated fps statements and personal bias towards a manufacturer. Secondly the “CPU bottleneck” has nothing to-do at all with the GPU manufacturer used and lastly NO not everyone reporting this had used a nVidia card.
Just google for GW2 benchmarks.

If u plan to play and enjoy sPvP or WvW, yes :p
If u just play solo or with some friends in PvE, simply try it with your current system.

PS: The i5-k series seem to be the favorites for Gw2.

Try setting your sound quality and buffer sizes to the middle, this is a know problem since release and yet no “real” fix.

The i7-3740QM coupled with a half decent GPU or lower settings, should be able to deliver around 15-20 FPS in those heavy door zerg situations, which i think is “playable”.

Yes even a 550Ti can run on maxed settings. However in what situation u can run max. depends on your playstyle and CPU. Your AMD will only be able to run WvW at around 15-20 FPS and heavy zergs fights at 5-7 FPS. On the other hand in normal solo/small group outdoor PvE u will probably reach 60 FPS (vsync=on).
Larger events in PvE will suffer from the same problem like WvW, basically every activity u do with more than 5 players will start to degrade your FPS.
Your system will range from 3 FPS in zerg WvW to 100 FPS in solo PvE story instances.

I also think sPvP will be hardly enjoyable with this CPU, given its competitive nature.

The problem with this poorly designed stat, is not the “farmer”, since he simply uses his MF set anyways, like the WvW player uses his PvP set. Its the players that still have problems or level up and know there is a MF stat. Those players have 2 conflicting goals, DPS/survivability/Utility vs MF/Loot.
Now the question is why do players have a problem to decide/balance those 2 directions? I think the answer is, that all stats/gear is making u as a player more powerful in some sort, more dmg or more HP, more survivability. On the other hand getting better loot does the opposite, it makes u less powerful.
Thats like telling a car fan/modder to exchange his 400 HP engine for a more fuel efficiency one.
It simply don’t resonate well with most players that are used to “pimp” there avatars for more power, while it also generates envy if they don’t use MF.

The problem is GW2 has no rewarding loot system, since there is no rewarding or “cool” loot. Even if u find a exotic item, its not rewarding, since most of the time u either cant use it, the stat combo don’t matches your play-style or u already have bought a exotic. This is true with all loot/items in GW2, so most simply sell those at the TP. This is not rewarding at all. As example if u would get a legendary-weapon matching your build/class this would be considered “rewarding”, simply because legendary’s are so hard to get.

So yes i agree with u, improving a already boring and unrewarding loot-system is meaningless, better spend time on the base system or other areas (WvW hint hint).

Uh i seem to follow u around :p

Nah your observations are wrong, don’t use the forum, but actual benchmarks from “professionals”, anything else is suspect to errors and all kinds of bad data/conclusions.

http://www.tomshardware.com/reviews/guild-wars-2-performance-benchmark,3268-4.html
http://www.notebookcheck.net/Guild-Wars-2-Benchmarked.81604.0.html

1) avoid AMD CPU’s completely for GW2, especially for WvW
2) any midrange GPU is fine for PvE/WvW
3) PvE is playable with a “normal” system and a older dual core CPU, but GW2 still favors older intel CPU’s over newer AMD
4) large WvW battels are “bad”, without a overclocked i5/i7
5) GW2 has problems utilizing more than “2.5” cores.

ahhh be carefully with such “simple” advices! U ignore the whole memory timing component, which may result in a un-bootbale PC and the need to reset it via the cmos jumper. Many older boards also have old FSB switches to run the ram multiplier differently from the FSB. I advice to google for a proper Q6600 overclocking guide if u are interested.

Like this one: http://archive.atomicmpc.com.au/forums.asp?s=2&c=27&t=496
(U need to copy&paste the link, don’t click + redirect)

zerph.1482:

The CPU is significantly faster than the GPU, the GPU is ALWAYS the bottlekneck. Unless of course you do have a inferior cpu.

What? So we simply ignore all GW2 benchmarks and general game engine design of the last 15 years? A game “can” be CPU or GPU limited or even both/none at the same time. This depends on several factors and while most games are GPU limited, GW2 in a WvW scenario is CPU limited in a extreme way. Most GPU’s even go into lower power save states in those GW2 situations, since the CPU can’t “deliver” the GPU data fast enough.

Also comparing a CPU with a GPU is like comparing a car to a ship.

PS: Try use http://www.hwinfo.com/ , in combination with the rivatuner OSD display and make sure hyperthreading is disabled. This will show better whats going on in WvW vs PvE.

@roqoco Sorry got of topic, without adding much to this topic.

I think MF (assuming it works as advertised) in GW2 simply don’t adds anything important, since its just a plain passive/indirect gold increase stat. So removing/replacing it would be not a big problem.
I think a MF or similar system can improve a item/loot system, similar to what achievements do with other mechanics. I still would make the MF “effect” rare and desirable, without competing with other rune/armor/buff slots. Like a potion or special “lucky charm” slot, that enabled/enhance the effect. The main idea would be to make the item and “rare” loot hunting more rewarding.
I don’t see it adding much as passive gold increase stat/mechanic, it must add more than this. So maybe like a “discover” or “event” system, designed around the loot/gather/chest mechanic.
A good example would be the dog/pet u could send to find rare items. (forgot what game it was) So rather than a passive stat it should be coupled to a event, similar to what runes do on crits. So as example having 5% more passive dmg is oki, but summoning a cool pet is way cooler, rewarding from a gameplay perspective.

“None of the arguments against MF depend on how effective it is, sure, like poison, the more MF you have the worse the results, but that doesn’t mean that a little poison is somehow a good thing, does it?”

Really? So a 50% MF cap, with very little effect would make u draw the same conclusions than no cap with extreme effectiveness?

If u just want to have a discussion as intellectual training, than this is fine, but to use your devil example until we agree that the devil actually exists and is responsible for XY, any discussion on that subject is meaningless.

Your poisen example is also fishy, compare MF to alcohol, cigarettes or cars. We know those are “bad” or even “poisen”, but strangely they are still present and not banned. If MF would be ultimately “bad” per definition, putting it in the game would also be considered “bad”, so is Anet simply a “bad” company?

roqoco.4053:

All the arguments made here have been answered previously in the thread and miss the important points, which are to do with the effectiveness of the reward system and the health of the economy – not one person’s anecdotes about obtaining legendary weapons.

Since some of us don’t even agree on the base statement that MF actually impacts drops “significantly”, he has a point. If we don’t first agree that MF works by formula XY, any conclusion or discussion based on it, is meaningless and hypothetically.

As example, there could be a cap for MF effectiveness? Maybe 50%?
MF store booster could work differently than ingame %.
As shown by some tests MF could also not work at all or it could be extremely effective.

U have no conclusive basis for this discussion anyway. The best we can do is to discuss the concept of MF in a general way and observe/study its effects in other games.

roqoco.4053:

@Andypandy – It’s probably best to avoid posting links to suspect external sites, you may wish to edit your post to make the point without that link.

Lol, did not noticed this. The same post was in this forum, but i could not find it again, so i googled it and put the first link down, sorry

Here is the org. post, found it again!
https://forum-en.gw2archive.eu/forum/game/gw2/Magic-Find-No-Significant-Benefits/first#post437694

The point was, that i have reviewed 4 larger “actual” done test’s, rather than anecdotal evidence, hearsay or placebo and neither of those could convince me that MF has a “significant” meaning for me. All those test had problems in its setup and data analysis and the conclusions they came too.
Some of them suffer from heavy confirmation bias, since they want MF to “work”, even if the actual data is “fishy” at best. I wish they would double blind there tests+analysis, to remove this bias.

Just wanted to throw in a “negative” test, to show that its still not 100% clear how and if MF works, as it is often overstated as “100% fact”.

GW2 was and is still CPU bound in heavy populated areas. We never had big problems in solo story or sparsely populated areas, since lunch. The main issue most players have are WvW or large world-boss encounters. I for once did not notice any significant change since lunch, which matches what several benchmark sites tested over the last 2 months.

In WvW in a huge zerg, even with a 4GHz i7 u don’t get stable 60FPS, most players “try” to reach around 30FPS so they can play. I have like 5-10FPS if i “try” to play near a zerg that smashes on doors and like 15-25FPS otherwise, while i get 100 FPS in solo PvE.

The thing is “normal” PvE was always playable for most players, WvW and big events are the problem and in those areas the performance is the same since lunch, while adding the “invisible army” bug.

So far the only “known” way to play WvW with acceptable framerates, is to buy those i5/7-k models and overclock those, which is kinda sad since it adds a steep $ investment to play WvW, if u not already own such a CPU.

PS: My FPS are with a E8500 at 4.3GHz

Squall Leonhart.2075:

i would be surprised if there was much hand written SSE optimisation at all.

Normally the math lib is, since u can get them for free as lesser GPL. They use intrinsic and proper inlining hints, which results in “perfect” math lib code. So at least your basic vector, matrix and quaternion operations use less instructions, but u just need SSE1 for it. SSE2 comes in if u need more stable, precise matrix math to lower your operation errors adding up, thats more of a stability fix than a performance increase.
AVX is nice for image filters or special subroutines, which ofc get a huge boost, but optimizing a function by 500% that only contributes to 1% overall performance is not “worth” it. They should focus on better threaded code, since its a really hard job to write good threaded code, without extensive and lazy locking.

Several problems for real world diversify.

1) Runes + traits are fixed for time x. This means u can’t switch/change them quickly, while u also have to pay if u actually do.

2) No ability to save builds, makes the actual build “building” painfully.

3) Most traits + skills result in some straight forward designed synergy, they have very low “random” synergy. What i mean is the “hey what happens if i combine x + y + z” moment in GW1.

4) Combo fields….. no comment

5) Runes are destroyed on change, requiring u to handle multiple weapon and armor sets, without a system to actually “handle” them.

6) Very limited “working” and “different” builds, compared to GW1. GW2 “builds” are often just minor variations and play very similar to each other.

From my experience every class has like 4-6 “working” builds and a dozen variations. Thats still more than with a old skilltree system, but is a fraction of what GW1 had.

Its actually surprising that the WoW MOP skill-system change, looks less restricting than what GW2 has?

Jazhara Knightmage.4389:

would love to see a compile of this game with sse2-3-4 and avx support…..bet that would boost perf nicely…..add in FMA support and i bet amd chips would actually start to compete…..(would have to use open64 or gcc’s newer builds for that, since icc blocks avx on non-intel chips)

Nope u wont, first games use floats for almost everything, so SSE1 is enough for your normal vector3 and matrix math. Secondly the most slowdowns and performance problems are not in any SSE/AVX routines, but branching code, cache misses, memory access, thread synchronizations, STL containers, error handling code, drawcalls, traces and ofc the infamous locking aka mutex. Even if we add the best AVX compile using the best compiler out there, wont help much since that are loophole optimizations that hardly contribute to the actual FPS.

The best hope is DX11 and optimize branching or cache misses for AMD and they need to get there stuff together and invest in better threaded code.

@Jazhara Knightmage.4389 like noted GW2 uses the default VS2010 compiler+linker, also the performance boost by ICC is none existent for most games, while the ICC does some crappy loop unrolling job and adds crashes.
Most game company’s don’t use intels compiler, since there are no real speed gains over the much more stable VS one. The other problem is that while u develop the game u always use VS compiler, since its 10 times faster to compile code, therefor the whole development time u test with the VS compiler, therefor u should be really carefully to than ship a way less tested ICC release.

ICC is good for plugins or ffdshow, but more complex programs should stay away.

“toms” just released a updated analysis, maybe this helps?

http://www.tomshardware.com/reviews/guild-wars-2-performance-benchmark,3268-7.html

Looks really bad for AMD, even the lowest Pentium dual core outperforms any AMD.

I don’t think they “optimized” anything, so Intel CPU’s just simply brute-force there way to the better FPS. I guess for some odd reason the specific “slowdown” code runs better on Intel CPU’s, maybe lucky branch prediction or some other nifty technical detail.

Also i don’t see a “best played on Intel” logo or advertisement, u normally use if u actually worked together with Intel.

Problem is most players have only minor problems in PvE/Events, while they have huge to “impossible to play” in worldevents or WvW.

This means u may have “playable” 20-30 FPS in PvE, with drops to 10 FPS, while u may have 5 FPS in WvW fights.

Ah oki, 3500$ should work for WvW, obviously not everyone is committed to such a investment to play GW2 WvW, if they have no problems with other PC games. My friend can play WoW/LoL/WoT/BF3 just fine (med/low settings), so ofc he won’t upgrade for GW2 if he isn’t already sold.
The other problem is that even if he wants to upgrade, most other PC games are GPU bound, so while he may boost BF3 performance with a GPU upgrade, it wont do any good for GW2.

Its actually a little sad, we really tried to find a cheap way for him to upgrade for GW2 and the lowest “good” we could come with was around the 400$ mark. The problem is he compares GW2 to WoW/BF3 so ofc its hard to convince him to upgrade, if those games run “fine”.

cheapest kit for any “salvage” item
mystic forge kit (check wiki) for any green 70+
best/mystic kit for any “rare/exotic” lvl73+ item
blue kit for everything else

vendor any heavy/wood armor/weapon if u are after money

Since there are no monthly fees, people wont “really” leave GW2, they may come back after each new content or PvP update.
The main problem with “mostly consistant 60 fps” is that u obviously don’t play WvW, which was/is the “big” selling point for many GW2 players, but u need a abysmal beefed up PC to get even consistent 30 FPS in a large fight. Most of my friends did actually spend like 300-700$ and build/upgrade there PC’s only for GW2. So we talk about a pretty steep entry barrier, if u enjoy competitive WvW.
I tried to get 2 friends to play WvW with me and they got like 5 FPS in WvW, so they are back in PandaLand

Squall Leonhart.2075:

oh wow, way to make yourself look foolish and validate my post in the same instance :thumbs:

I have no problem “looking foolish” in your eye’s, since u do the same in mine :p

Just had to commend on simply overblown and out of context statements. U might want to check the difference between CPU features, architecture and how those impact game FPS. Only because a i7 has better virtualisation or AES encryption support, don’t mean “they are completely different chips.”.
I also already gave a testable example by linking the game benchmarks, which indicate that those chips are -3-5% “different” regarding gaming performance. If a 5% difference makes a chip “complete different” than we simply have a different understanding of “complete different”.

“2. MAKES THEM COMPLETELY DIFFERENT CHIPS”

Yeah CAPS will make my statement even more valid!

Seems u like to “ride” on semantics, instead of the actual discussion. So just for the sake of argument, yes a i5 and a i7 is “different”. Does those differences matter for most games? No they only effect FPS in the range of 0-5% and the GPU is what still matters for most PC games.

U are happy?

Unfortunately there wont be a “quick” patch, there will be gradual optimizations, since also Intel users have big problems in WvW. I have no clue what they screwed up so much, that GW2 is this CPU bound and favors Intel CPU’s. If it was a small problem the AMD guys would have figured it out by now. Normally big game company’s work together with CPU/GPU manufactures and get feedback on performance problems. Maybe its a problem in some of the middleware they use?

The only choice u have is either grab a i5 or stop playing. I don’t think selling is a option.

PS: Maybe u or someone else has a old intel dual core + mobo? A E8400 at 4Ghz should already double your FPS in GW2.

Squall Leonhart.2075:

the difference between i5 and i7 is physical, they are completely different chips.

What a bunch of Bollocks :p Where do u people get this stuff from? At least do some minimal research before u post on something. A i5 and i7 of the same line, have the exact same architecture (sandy-bridge, ivy…). The differences are by removing and shifting some features around, those features mainly are targeted towards servers or professional applications. If u check the i3->i5->i7 features u clearly see this.

The funny part is that a quad core i3 or pentium would be the new gaming CPU, since u get 90% of the performance of a i7. Intel choose to cut turbo-boost and only offers dual core versions for a reason, they don’t want to cannibalize the more expansive i5/i7 market.
Anyone remember the “old” celeron debacle?

Crawford.4135:

Stop telling people to purchase i5 processors for heaving gaming!

Seriously folks, know what you’re buying. I would NEVER touch an i5 for a gaming system, ever.

Cheaper? Yes, better? No.

Than pls also stop presenting people “anecdotal” evidence or hearsay!
I’m really glad that we have so many gaming benchmark sites, which simply blow your crazy assumption out of the water. Seems u have no real clue how gaming code and pipelines work, the larger caches mainly help programs which more rigid code paths and data access like databases, 3d modeling, photoshop.

So actually the i7 is for special professional niche programs, that will benefit from the larger cache and even than i would run a benchmark and see if they actually perform better, since larger cache and performance is a quite complex field and not a simple 1:1 relation.

Just one example of dozen test’s u can find:
http://www.anandtech.com/show/4083/the-sandy-bridge-review-intel-core-i7-2600k-i5-2500k-core-i3-2100-tested/20

Just to-be clear if u think the 1-5% performance increase is worth the higher price, go for it. I just want to make the point that any “reasonable” person would go for a i5 over a i7, since u take the saved money and simply buy a better GPU. Hence for most PC games going with a i3 or Pentium will yield more FPS, if u can get the next higher GPU for the saved money. Unlike GW2 most PC games are actually GPU bound.

Squall Leonhart.2075:

im fairly sure they are using Opengl 3.0 on the mac port

Ofc they do, since there is no D3D9 on MAC, but there is D3D9->Mac OpenGL wrapper middleware and company’s out there, that can “port” a game quite fast to MAC using wrappers. What this means is that u can “port” a lazy D3D9 engine to mac, but this wont help with DX11, since those wrappers wont work here.

Means if they use a wrapper, we are not one bit closer to a Dx11 version, while if they actually have a “clean” OpenGL abstraction layer for there render core, abstracting also a Dx11 version is much more easy.

@Vexus.5423;

No the issue can’t be “dye” or “data” related, since u have a max. of 65,507 bytes per UDP packet. U actually only need very few bytes to encode a full MMO character, since u use lookup values and can highly compress those data. I can encode all dye and armor informations in the same size than the position/velocity value.

As example:
512 color values = 9 bit
64 armor prefix = 6 bit
512 armor types = 9 bit
256 rune types = 8 bit

So we need 32bit for each armor slot, if we assume 12 slots we need 48 bytes to fully describe the “dynamic” part of the player. We may need a additional 48 bytes for the “static” part, aka hair + face + size + race.

The static part can actually be cached and saved, removing the need to send those more than once per map. Thats actually what EvE online does, they store/cache all static data on your local hard-drive and simply recheck from time to time.

So we have 48bytes for armor + dye and we also need at least 24bytes to send position and velocity.
Now how can it be that GW2 can update 24bytes (pos+velocity) in a very high frequency, but we cant update an extra 48bytes with a much much lower update frequency?

In comparison all this physics information at very high frequency for every effect and projectile, is much more problematic than the actual player + armor data, since those are low frequency or even only “once” updates.

Also the often noted “streaming” of those armor parts and textures, is a false assumption for this problem, since at 1-2GB PC GPU Ram u can probably hold all mesh and texture data in the GPU ram, without the need to stream anything. WvW is using a fixed mesh and texture set, there is nothing changing here. We also don’t talk about ultra high quality mesh characters or textures. So the actual GPU ram amount is more than enough to hold all this data.

There must be something else Anet has problems with, that we cant simply find by “reasoning” or speculating. My guess is its some low level technical problem in there client-server and database code.

I hope they can fix/optimize this somehow, since its really annoying.

I just realized some good news for a DX11 update :p
They have a MAC version running, this means they already have a “good” abstracted graphics engine layer. This means moving from DX9 to DX11 is a lot less time consuming, since all important layers/interfaces are already abstracted.

I only really hope they don’t use those D3D mac wrappers, that “translate” D3D9 calls, since than we are back to square one.

@deltaconnected.4058

Hey the ingame display looks like the EVGA precision thinggy, how did u enable the CPU stats? I only get GPU data displayed.

I think the op is spot on, GW1 != GW2. Even at lunch GW1 had numerically and “realistically” more combinations to build your character. Depending on your class GW2 seem to have even less “working” skill/trait combinations than the archaic WoW. I pretty much have exactly 2 “builds” i like, one for PvE one for WvW and except for minor tweaks, i don’t see any reason why i would change those. In GW1 i never had “a build” i was constantly changing class and skill combinations, this is completely absent in GW2.

Btw i just looked at the WoW-MOP, skill-system change and it looks like its even less restrictive than the weapon/trait system currently in GW2? It seems u can pick whatever talent/skill and “glyph” u like now, without any restrictions?

I especially don’t like the tiered trait and “shoehorned” minor trait system, let me pick what i actually want and not what a game designer “thought” is good for me. In my build i rarely need any of the minor traits at all, they seem like waste of points.

The Dude.6942:

the phantasms which bypass LOS are those which are spawned AT the target, ie the melee ones. you know, those which are often pretty much instakilled, sometimes even before they attack once.

sure, a berserker from a glass cannon mesmer hurts. but so does pretty much anything from a glass cannon, regardless of class. remember they take alot of damage as well.

What other phantasm ignores LOS?

AE skills work on a completely different level, but common we all know i can target someone on a wall and run to the wall, while he might be where-ever inside and i can cast berserker’s all day long, as long as i’m in range. The best part is, if he gets down somewhere inside, u simply keep spawning more berserker’s to finish him.

We all know its one of the main reasons why GS is so cool in WvW and such a “ability” would probably be stated in the tooltip, if its intentional.

So if we use common logic, we will notice no other “working” direct target-able skill ignores LOS this way. Hence this rule is even true for other MMO’s thats why we correctly assume its “bugged”.

So yes its a bug and they will fix it.

Chubbysumo.4736:

This told me that there was some bad code, and stuff kept getting processed and handled by the CPU and GPU, and then gets discarded(so called “bad GPU/CPU calls”). Since bad calls are not read by your GPU as “used” time, your GPU and CPU usage says extraordinarily low while its workload is actually high, and your game performance suffers.

Where did u read this or is this a conclusion by yourself? To make it short, NO thats not how hardware performance counters in your CPU/GPU work at all. There are no “bad” calls that stress your CPU/GPU that the hardware counters wont measure. So either your measuring software is bad or your workload to heat-production conclusion don’t hold.

I’m a programmer and after reading: “(so called “bad GPU/CPU calls)” i nearly spit my coffee :p This is complete “Bollocks” in this context.

If i had to guess i could speculate that u mean cache or pipeline “trashing”, which means filling data/buffers only to “throw” them away if the code branches unpredicted. The problem is all this “trashing” still uses CPU/GPU cycles, that can be measured by internal counters or simply timed using the various CPU timers.

MoosieOfDoom.4926:

So whats that says is that the game just has a bug in the code? And that it is just bad luck that i am one for the few that have this problem from day 1. With some days/weeks without problems…

I can just speculate, so ofc in your particular case i would also assume a “system” specific problem, since otherwise more players “should” encounter this problem. After yesterdays crash’s i would point to the patch in my case, which contradicts your pre-patch crashes.

I mean soon u reach a point where, except from changing your mobo/cpu, u have done everything that remotely relates to such problems.

Cbuzz.5083:

Core i5-3570K (3.4G,L3:6M,iGPU,4C,77W,rev.E1) with a decent aftermarket cooler, then overclock it to 4.5Ghz, then your good to go.

Keep in mind “good to go” means, u might get 30-45 frames instead of 20-30

“Model ‘0×006a28’: Can’t link child ‘0×00d120’ to parent bone ‘aura’, there’s no bone with that name. Linking to root instead
Model ‘0×006a28’: Can’t unlink child ‘0×00d120’ from parent bone ‘aura’, there’s no bone with that name.”

Yesterday for the first time ever, i also had 3 crashes in short succession, with similar errors. Since i did not change anything on my system, i guess the recent patch is a cause for this.

The actual “Exception: c0000005” is a illegal memory access in “no mens land” so somehow a pointer is corrupted, which may be caused by all kinds of stuffs, ranging from bad RAM/over-clocking to bugs in the game code, used .dlls, bad error handling or simply not correctly initializing/clearing a pointer variable.

The error trace code seem to fallback/crash for some animation/“bone” related calls, which point to a actual game code problem.

I also hit the same problem, updating your GPU is “nearly” useless for WvW, since WvW atm is heavy CPU limited. The solution atm is to get as much raw CPU power as possible, which favors the i5 “k” series cpu’s and over-clock it as high as u feel “comfortable” with.
Hence for GW2 it might even be usefully to get a “old” sandy-bridge over a “ivy”, since with decent cooling u can get higher over-clocking results.

The cpu/mobo “muslumgurseks.4951” listed are a good combo, alternatively this mobo might get better over-clocking results: http://www.newegg.com/Product/Product.aspx?Item=N82E16813157293

It works like this:

Moving from DX9 to DX11 for a existing game is mainly a software optimization, with additional rendering features, which might be used by the game or not. The FPS increase on PC comes mainly from the reduced “drawcall” costs, which “probably” is also a issues in GW2, especially in WvW.

Modern GPU’s don’t run in a DX9/11 “mode” the driver basically abstracts all down to the hardware, so your GPU shader lighting code for a dx9 game is doing the exact same things than for a dx11 game, assuming the same HLSL shader. The difference is that the DX11 game can use additional commands that may improve FPS, the closest similarity would be going from MMX to SSE instructions.

As for 32bit vs 64bit, thats overrated from the FPS standpoint. The main reason is the 2GB 32bit RAM limit per process, so by going 64bit games don’t necessary need “good” memory management. If u have 4GB ram for your game, u can simply get away with “bad” memory code and memory leaks, which otherwise would crash the game over time.

If we check the features added to a 64bit CPU + OS, we mainly get more registers, which don’t impact game FPS greatly. The branch prediction and pipeline trashing costs or cache misses, play a much larger role in game code and those are the same for a 32bit OS using the same CPU.

The improvements some 64bit game releases see, are more a result of using a updated compiler or some additional code optimizations, unrelated to the actual 64bit recompile.

As for updating, going 64bit is relatively easy for none network games, often a simple recompile. The problem is that network code might need to-be adapted or reviewed.
Going DX11 from a DX9 engine, is quite more time consuming and often shadows/lighting is updated for the new DX11 code, which means the need to update/test shaders and the whole pipeline. Considering this amount of time, most DX9 games only go DX11 if they can “advertise” with those cool new features/improvements, so the work pays off with potential new buyers/subscribers.

VirtualBS.3165:

I’ve been using Virtu MVP (HD3000 + GTX 560 Ti, in “i” mode, with the monitor connected to the iGPU via HDMI), with both Virtual V-sync and HyperFormance, and can honestly say that it works 100% in GW2.

Ah oki so at least no crashes or graphic glitches? Do u get actual better FPS or u just use the “vsync smooth” feature? What FPS u are getting in PvE with “HyperFormance” turned on and off, if u have time to test it?

Yeah sorry, but instead of useless whining and kittening from trolls like the above 2, i was hoping to get someone to simply turn MVP on, start GW2, write down FPS and repeat the same without MVP. I guess this was to naive of me in a GW2 forum. Instead i get useless troll answers.

So neither u not the guy above u answered any of my questions. U both simply repeat “hearsay” and i already edited and posted a benchmark i found with some newer drivers, yet u took the time to quote without actually reviewing the article, that would have show that the “opinion” from the first guy, is at least partially false, outdated or irrelevant.

I also did not ask about general “experience” i specifically asked if someone “tried” it for GW2, since i don’t care if it don’t do anything in all other PC games out there!
Also i obviously have no MVP board, since otherwise i would not have to ask here, so how can i test it?

PS: At least thx for the bump….

Thx for your personal opinion, but i rather go with hard facts than hearsay. Yes i understand that its not “great” or otherwise every game review site would recommend it, however i saw some valid and positive benchmarks, for some modern games, using the latest lucid drivers.

The biggest issue seems stability and compatibility for games, but thats why i ask here. I’m not interested in any “hearsay” or personal opinion or if it don’t work with game XY. I would like specific GW2 test results, especially under 60 FPS systems with vsync off and only MVP “Hyperformance” mode. This means i’m interested in pure FPS gains/loss and not the “special vsync” feature over 60FPS.

As example, here is a test with newer drivers, but i cant find any test that uses the latest drivers. As shown it also “seem” to work for <60FPS as well, which was not seen on reviews before the newer drivers.

http://www.hardwaresecrets.com/article/Virtu-Universal-MVP-Review/1548

Hi,

simple question does anyone out there has this feature working for GW2 and if so does it increase FPS?
Than since ivy-bridge has ~30% more GPU performance, this may effect MVP for GW2 too.

So if u have such a board/cpu can u post result with MVP on/off, while in GPU (none WvW) heavy areas?

Thx

PS: If possible also post a short system spec, board/cpu/gpu/MVP settings, GW2 test area and if the monitor is connected to the iGPU or external gpu.

@VirtualBS.3165 Sure thats all true, the HPET is the superior method to time things in applications period. However we are discussing if a more precise timer can have any FPS impact at all in a game engine and here the answer is no, since it don’t effect any performance relevant game engine subsystem.

The OP uses the common excuse, “Well it worked for me, so it might not work for everyone and u are all arrogant requesting proof.”

The problem i have with any of those “magic” FPS fixes is simply the magnitude, normally those differences are due to bad test setups in largely unstable environments or simple total overstated.

Its like saying: “Well if u use those engine additives u will get 300 more hourse-power!”
The OP did not even respond to the simple request if he could reliable reproduce this “phenomenon” of his, by enabling/disabling the tweak.

The problem is once u “spoke out” at whatever standpoint or claim, u will hardly ever review or take it back, if there is a chance u might be wrong. Thats basic human psychology, so we will probably not see any actual additional tests from the OP on this.

Just noticed the OP topic: “LAA bit (might increase FPS on high-end rigs)”

Why does everyone assume more available memory has anything to-do with FPS? At best it reduces loading times and prevents “hick-ups” from page swapping on HDD and in GW2 case prevents this out of memory crash.

VirtualBS.3165:

It all depends on the used function. For all we know, they can be used thousands of time per frame. At 60Hz, each frame is ~16.6ms (1.66×10^-2 s) — most of those timer functions have precisions around 10^-8 to 10^-7. In the worst case (10^-7), they can be called 166.000 times per frame if need be.

Um u seem to misunderstand what they are talking about, first of timing code is used a dozen of times per frame not " thousands of time". Its also common practice to actually only update your internal game timer only once per frame, so game calls get all the same cached timing result within a frame, thats needed since otherwise strange timing errors would happen.

Than we talk about a change from one function call that takes 3000 cycles and replacing it with a 100 cycle timer call. They simply refer to the problem of profiling correctly, if the timing call itself takes significant time. So if a timer call takes 3000 cycles and u want to time a 100 cycle call, u obviously need to eliminate the 3000 cycles of the timer call itself.

As example a expansive d3d “DrawIndexedPrimitive” and “SetStreamSource” call also takes around 1000-3000 cycles or even up to 10000 cycles. The difference is those calls are actually called 200000-1000000 times per frame on complex scenes.

So reducing a function call that makes up less then 1% of the per frame cycles, even by a factor of 1000 times wont yield any measurable results. As noted its used to correctly profile code or get correct timeouts for network or other timer critical tasks like video playback.

Odoakar.4759:

Wow, who crapped in your coffee? I’m just saying that I noticed an increase in performance for me, and that it might work for others. Jeesh…

What u did was claiming a 160% FPS increase by some software, without doing proper testing. Thats like promoting “snake oil” and since this happens all over the place, i simply noted whats wrong with this kind of testing and how u can improve it.

So do proper testing, post the results here and we can get some usefully informations out of it.

BTW here are the things “game booster” does:

1) Close “unneeded” programs! Wow what a feature now i don’t have to close my word or pdf reader anymore, before starting GW2! Wont affect FPS, but at least memory.

2) “optimize your system” Yay all those little registry hacks in yet a other tool, without actually affecting game FPS.

3) “game folder de-fragmentation” Finally a feature that might help users with old HDDs, that don’t run other de-fragment tools. Useless for SSD users.

4) “driver update checker” Again a usefully feature, that might actually increase game FPS if u run a outdated driver.

So no overclocking features for a “game optimizer” the only FPS relevant feature, is the driver updater. So yes if u need yet a extra tool to tell u your driver needs a update, than this is the tool for u, anyone else can just use the default driver update feature from nVidia/Ati.

So did the tool install a new GPU driver? If so what version was installed before and what version now? This might actually be some usefully information.