My performance in game reflects this from Tom's 100%
Seriously making me consider going out and buying a new board and SB i5.
My system runs the game "OK" but I'm honestly at medium on a fair amount of stuff to keep 30-40+ FPS at 1080p.
This is with 8gb Kingston RAM, HD 6870 OC'd and PhenomII x4 OC'd to 4.0ghz!
Very sad.
Yep, unfortunately that's really quite bad for your system, you should be cranking 60+ all the time.
I'd written this in another thread as well and IIRC there was some doubt about it... So I've been watching my FPS more closely since the headstart.
I have an i5 750 and an ATI 8850, both on factory clock (that'd be just 2.66 on the i750) and 6GB RAM. And I'm getting constant 40+ FPS on the "Best Appearance" setting. It fluctuates between 40 and 70 and it's usually 50+ rather than 40+; and it seldom dips below 40. Mind, this is only PVE - the open world and cities - as I haven't been able to step into WvWvW.
The game performs much better than I expected. I can even go for the hardest jump puzzles on "Best Appearance" setting, what more could I ask for?
Yeah I'm a bit disappointed by this as well. My computer is able to play every other game on full quality, but GW2 has a hard time in some places. It has to be the CPU because my graphics card is an absolute beast...
Oh well. I suppose it is about time to upgrade again anyway!
My performance in game reflects this from Tom's 100%
Seriously making me consider going out and buying a new board and SB i5.
My system runs the game "OK" but I'm honestly at medium on a fair amount of stuff to keep 30-40+ FPS at 1080p.
This is with 8gb Kingston RAM, HD 6870 OC'd and PhenomII x4 OC'd to 4.0ghz!
Very sad.
Don't see Phenom CPU's there, only the newer FX series. AFAIK the FX series is a 'lesser' than the Phenom. If they're equaly in performance and I look at the 6-core FX compared to my Phenom X6 1090T (non OC'd), I think my performance is better (might be due to GTX460 SLI instead of an ATi card).
That isn't a broad Intel versus AMD comparison. It's only a Sandy Bridge versus Llano and Bulldozer comparison. Bulldozer is a broken mess, and Llano is a laptop chip, so they aren't going to hold up well against Sandy Bridge in other games, either, but will probably tend to do at least somewhat better than in GW2.
I'd be interested in seeing how other architectures compare. Due to architectural similarities, you can largely guess Ivy Bridge from Sandy Bridge, just by adding 5% or so. Athlon II and Phenom II are both fairly similar to Llano, and you could typically subtract 5% or add 5%, respectively, though there can be considerable variation based on how much use it gets out of L2 or L3 cache.
But Sandy Bridge results don't necessarily tell you that much about older Lynnfield and Clarkdale processors. Will those crush a Phenom II, too? How about if we go back further to a Core 2 Duo/Quad?
It also doesn't tell you anything about relative performance on Bobcat or Atom cores, though given how processor-heavy the game is, it's a pretty safe bet that it's unplayable on both. That's a shame, as GW1 ran very well on an AMD E-350.
But it is starting to look like GW2 has a rather bad case of badly-coded syndrome. If an FX-8150 or a Phenom II X6 1100T can't really run the game that well even at minimum settings, the problem isn't the processors. The problem is the game.
Thats a terrible frame rate with AA off you should always be above 30fps for a smooth ride.
Anti-aliasing puts essentially no load on the processor whatsoever. In some cases, it's as simple as the processor telling the video card to turn on anti-aliasing, and then never touching it again. That's why you can often force anti-aliasing through video drivers. FXAA post-processing is a little more involved than that, but the added processor load amounts to a rounding error.
Thats a terrible frame rate with AA off you should always be above 30fps for a smooth ride.
Anti-aliasing puts essentially no load on the processor whatsoever. In some cases, it's as simple as the processor telling the video card to turn on anti-aliasing, and then never touching it again. That's why you can often force anti-aliasing through video drivers. FXAA post-processing is a little more involved than that, but the added processor load amounts to a rounding error.
Yep... I suspect that my bottleneck is about 65% GPU (ATI 5770), 35% CPU (Phenom 9850 BE 2.5x4).
I'll probably upgrade both in the near future(the 965: 3.4x4), but now I'm a little worried to, seeing as how the problem seems to be hit and miss with the AMD chips.
My computer runs it fine maxed out even though I have a Phenomenah 2 (x6). A good graphics card is a lot more useful than a good CPU while you game.
I have a Nvidia 480 GTX BTW, a bit old but still pretty good as long as you have enough airflow.
But it was a good idea to post it for people who plan to upgrade for GW2. Then again, this might just be some optimization issue that might get fixed any day, the game isnt even really releasing until tomorrow so I wouldnt worry too much if you have a AMD.
Doesn't look like they tested WvW which is where the biggest issues will arise. I've gotten as low as 10 fps in stonemist castle fights which is completely unplayable.
Originally posted by jusomdude Doesn't look like they tested WvW which is where the biggest issues will arise. I've gotten as low as 10 fps in stonemist castle fights which is completely unplayable.
The trouble with testing WvWvW is that it's not repeatable. If there are 20 players in the scene when you test one card, and then 30 when you test another, the numbers aren't comparable. Given the choice between measuring things that are easy to measure, and measuring what you actually want to know, they're likely to choose the former.
Runs fine for me in the triple monitor EyeFinity portrait resolution of 3240x1920. (how's that for a fun string of buzzwords)
Also, Tom's Hardware therefore Intel. They have a history of Intel coming out on top irrespective of the rest of the "review". I actually stopped using their website around two years ago or so, and I'm not surprised by the pot-shot that they take at the AMD hardware on the linked page.
So they test a 220€ intel CPI against AMD's 150€ CPU. Seems legit.
By the way a year and a half back when I bought this machine I chose AMD's 1055T over i5. It outperformed it in practicaly all the tests and it was cheaper. And I'm also not sorry for making that choice, since all the games ran great. I'm runing GW2 with 60fps on average... 1055T and 1090T were great CPUs. However, I don't know about these latest models. And for now, I don't really care much
Not sure what all the fuss is about. I'm running on a quad core 965 BE OC'd to 3.8 gHz with crossfired 6870's OCd to 975 (proc) & 1250 (Memory).
Three monitor Eyefinity setup 5965X1080 and getting ~65-70 FPS in open world and ~ 45 FPS in WvWvW... not the smoothest, but a far cry better performance than I was getting in SW:TOR.
Right now I have everything set to max (except reflections... kills framerates) and am perfectly happy with the performance.
That's wierd... it's been the smoothest running game I've played in recent history.
I'm running a 9850 (2.5x4) at stock clock
8 gigs at 800 mhz
ATI 5770 OC'd quite a bit, can't remember the numbers, though.
1080P all settings cranked, but no AA
I'm getting solid FPS; wouldn't be surprised if it never dipped below 15, and hovers +30.
What the what?
So under 30 fps (as low as 15 holy cow) is smooth to you?
Thank god you added some numbers to your statement.
You misunderstand. Majority of the time its at 30+ fps.
Those specs are similar to my laptops. Mine is at 30-50 fps and dips at around 15 fps when too much loading happens (no ssd). Its a 5870m overclocked to desktop 5770 specs and i7 720 (1.8-2.4 ghz) and its running pretty smooth.
''/\/\'' Posted using Iphone bunni ( o.o) (")(") **This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
Originally posted by Z3R01 AMD Phenom IIx6 1100t Black (4.0 OC'd), AMD HD6970 2gig, 16gb Ram, 120g SSD+ 2tb HDDx2, Win7 Pro.GW2 60fps Lock with Vsync on all the time except in massive WvWvW zerg where i can see a drop to 40-45.
I'm running a similar system but with a Nvidia GTX 560 TI and only 8gb of Ram and I am seeing the same performance.
The game will make use of as many cores as you can toss at it. Across all 6 corse its averaging between 50-75% cpu load with each core being actively used.
At that point I imagine its more of other factors that are causing cpu loss other then the cpu. Graphics Card Bandwidth, SSD / HD access, ect.
Crusing on 15 FPS in open space, it jump up to 35-40 when in a cave, or a place where you can't look far away into distance. That's on auto-detect settings (medium shader, low shadows, low LoD, low animations... that's all I remember), 1920x1080, game looks great.
The fun part is I can change settings to lowest, disable everything, lower resolution, and I won't gain a single FPS. Obvious bottleneck is obvious :P
I didn't try the WvW yet, I don't know if I dare...
Anyway, AMD has better price/performance ratio than Intel.
Originally posted by alilsneaky What the what?So under 30 fps (as low as 15 holy cow) is smooth to you? Thank god you added some numbers to your statement.
Most people don't complain that TV and Movies aren't smooth - they run around 30/24FPS. It's not so much about the actual frame rate as it is that most people just perceive when/how often the frame rate changes. It's especially bad with VSync On, as you get the 50% penalty (but for some people, like me, playing with VSync off isn't an option because tearing is more distracting)
Comments
AMD Phenom IIx6 1100t Black (4.0 OC'd), AMD HD6970 2gig, 16gb Ram, 120g SSD+ 2tb HDDx2, Win7 Pro.
GW2 60fps Lock with Vsync on all the time except in massive WvWvW zerg where i can see a drop to 40-45.
Playing: Nothing
Looking forward to: Nothing
Yep, unfortunately that's really quite bad for your system, you should be cranking 60+ all the time.
I'd written this in another thread as well and IIRC there was some doubt about it... So I've been watching my FPS more closely since the headstart.
I have an i5 750 and an ATI 8850, both on factory clock (that'd be just 2.66 on the i750) and 6GB RAM. And I'm getting constant 40+ FPS on the "Best Appearance" setting. It fluctuates between 40 and 70 and it's usually 50+ rather than 40+; and it seldom dips below 40. Mind, this is only PVE - the open world and cities - as I haven't been able to step into WvWvW.
The game performs much better than I expected. I can even go for the hardest jump puzzles on "Best Appearance" setting, what more could I ask for?
P.S.: I have V-Sync off, though.
AMD Phenum II x4 3.6Ghz 975 black edition
8 gig Ram
Nvidia GeForce GTX 760
Yeah I'm a bit disappointed by this as well. My computer is able to play every other game on full quality, but GW2 has a hard time in some places. It has to be the CPU because my graphics card is an absolute beast...
Oh well. I suppose it is about time to upgrade again anyway!
Don't see Phenom CPU's there, only the newer FX series. AFAIK the FX series is a 'lesser' than the Phenom. If they're equaly in performance and I look at the 6-core FX compared to my Phenom X6 1090T (non OC'd), I think my performance is better (might be due to GTX460 SLI instead of an ATi card).
That isn't a broad Intel versus AMD comparison. It's only a Sandy Bridge versus Llano and Bulldozer comparison. Bulldozer is a broken mess, and Llano is a laptop chip, so they aren't going to hold up well against Sandy Bridge in other games, either, but will probably tend to do at least somewhat better than in GW2.
I'd be interested in seeing how other architectures compare. Due to architectural similarities, you can largely guess Ivy Bridge from Sandy Bridge, just by adding 5% or so. Athlon II and Phenom II are both fairly similar to Llano, and you could typically subtract 5% or add 5%, respectively, though there can be considerable variation based on how much use it gets out of L2 or L3 cache.
But Sandy Bridge results don't necessarily tell you that much about older Lynnfield and Clarkdale processors. Will those crush a Phenom II, too? How about if we go back further to a Core 2 Duo/Quad?
It also doesn't tell you anything about relative performance on Bobcat or Atom cores, though given how processor-heavy the game is, it's a pretty safe bet that it's unplayable on both. That's a shame, as GW1 ran very well on an AMD E-350.
But it is starting to look like GW2 has a rather bad case of badly-coded syndrome. If an FX-8150 or a Phenom II X6 1100T can't really run the game that well even at minimum settings, the problem isn't the processors. The problem is the game.
Anti-aliasing puts essentially no load on the processor whatsoever. In some cases, it's as simple as the processor telling the video card to turn on anti-aliasing, and then never touching it again. That's why you can often force anti-aliasing through video drivers. FXAA post-processing is a little more involved than that, but the added processor load amounts to a rounding error.
Yep... I suspect that my bottleneck is about 65% GPU (ATI 5770), 35% CPU (Phenom 9850 BE 2.5x4).
I'll probably upgrade both in the near future(the 965: 3.4x4), but now I'm a little worried to, seeing as how the problem seems to be hit and miss with the AMD chips.
My computer runs it fine maxed out even though I have a Phenomenah 2 (x6). A good graphics card is a lot more useful than a good CPU while you game.
I have a Nvidia 480 GTX BTW, a bit old but still pretty good as long as you have enough airflow.
But it was a good idea to post it for people who plan to upgrade for GW2. Then again, this might just be some optimization issue that might get fixed any day, the game isnt even really releasing until tomorrow so I wouldnt worry too much if you have a AMD.
A FX-8000 behind a Pentium DC?
what madness is this?
This review was based on performance during a BWE event prior to most of the newer optimizations.
As such, you need to take the review with a grain (or two) of salt.
DarkSpace Developer - Play DarkSpace - Play For Free!
Medusa Engine SDK - Free MMO Game Engine
Hampton Roads/East Coast Video Gamers Association
The trouble with testing WvWvW is that it's not repeatable. If there are 20 players in the scene when you test one card, and then 30 when you test another, the numbers aren't comparable. Given the choice between measuring things that are easy to measure, and measuring what you actually want to know, they're likely to choose the former.
Runs fine for me in the triple monitor EyeFinity portrait resolution of 3240x1920. (how's that for a fun string of buzzwords)
Also, Tom's Hardware therefore Intel. They have a history of Intel coming out on top irrespective of the rest of the "review". I actually stopped using their website around two years ago or so, and I'm not surprised by the pot-shot that they take at the AMD hardware on the linked page.
edit: DxDiag: http://pastebin.com/HNkR63PX
So they test a 220€ intel CPI against AMD's 150€ CPU. Seems legit.
By the way a year and a half back when I bought this machine I chose AMD's 1055T over i5. It outperformed it in practicaly all the tests and it was cheaper. And I'm also not sorry for making that choice, since all the games ran great. I'm runing GW2 with 60fps on average... 1055T and 1090T were great CPUs. However, I don't know about these latest models. And for now, I don't really care much
Not sure what all the fuss is about. I'm running on a quad core 965 BE OC'd to 3.8 gHz with crossfired 6870's OCd to 975 (proc) & 1250 (Memory).
Three monitor Eyefinity setup 5965X1080 and getting ~65-70 FPS in open world and ~ 45 FPS in WvWvW... not the smoothest, but a far cry better performance than I was getting in SW:TOR.
Right now I have everything set to max (except reflections... kills framerates) and am perfectly happy with the performance.
You misunderstand. Majority of the time its at 30+ fps.
Those specs are similar to my laptops. Mine is at 30-50 fps and dips at around 15 fps when too much loading happens (no ssd). Its a 5870m overclocked to desktop 5770 specs and i7 720 (1.8-2.4 ghz) and its running pretty smooth.
''/\/\'' Posted using Iphone bunni
( o.o)
(")(")
**This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
I'm running a similar system but with a Nvidia GTX 560 TI and only 8gb of Ram and I am seeing the same performance.
The game will make use of as many cores as you can toss at it. Across all 6 corse its averaging between 50-75% cpu load with each core being actively used.
At that point I imagine its more of other factors that are causing cpu loss other then the cpu. Graphics Card Bandwidth, SSD / HD access, ect.
AMD Athlon X2 5600+, Radeon HD 5770, WinXP 32 bit, 4GB DDR2
Crusing on 15 FPS in open space, it jump up to 35-40 when in a cave, or a place where you can't look far away into distance. That's on auto-detect settings (medium shader, low shadows, low LoD, low animations... that's all I remember), 1920x1080, game looks great.
The fun part is I can change settings to lowest, disable everything, lower resolution, and I won't gain a single FPS. Obvious bottleneck is obvious :P
I didn't try the WvW yet, I don't know if I dare...
Anyway, AMD has better price/performance ratio than Intel.
Most people don't complain that TV and Movies aren't smooth - they run around 30/24FPS. It's not so much about the actual frame rate as it is that most people just perceive when/how often the frame rate changes. It's especially bad with VSync On, as you get the 50% penalty (but for some people, like me, playing with VSync off isn't an option because tearing is more distracting)
The world we know is going away http://www.graystatemovie.com/
Look up Agenda 21 as well.