51 réponses à ce sujet

[mjordan79]

Notick wrote...

Maybe when I have a little more energy, I'll compile a list for you of average FPS for each game in their current technology standing.

I've been programming games sense the early GodWars/Dragon Realm MUDS, along with an avid modder (Q2 Mod's, HL2 Mods, PlanetSide source-code mods, Counter Strike mods...) C, C++, C#, Java, VB, Lisp (Autocad language). I'm not just speaking from my ass either.

Having built over 75 computers for myself, and over 300 for clients... Hundreds if not thousands of hours of research over the years... You'd be surprised.

Use your time for better things. Such list is not needed nor requested. Just look at a bunch of benchmarks. You'd be even more surprised.
BTW, I don't need to show off my titles to demonstrate a game runs poorly. For three reasons: it doesn't fix the game. I may discover that someone has far better skills than mine. And it's a very poor argument to a flawed discussion that doesn't add anything to the discussion itself.

Modifié par mjordan79, 24 mars 2011 - 04:36 .

[Jazharah]

mjordan79 wrote...

Jazharah wrote...

mjordan79 wrote...

Notick wrote...

I'm not writing this to bash anyone, be clear. I just want to try to be
constructive. For me, Dragon Age 2 has no reason to be heavy like that.

I think someone was lying.

Again, you're being picky. The average game today doesn't run at the high FPS you're hoping to see. I'm sorry that fact upsets you.

I have more than 60 games on my shelf. Only 4 or 5 of them runs poorly like Dragon Age. So you are the one who is definitely lying. A game that produces 32 fps using a GTX 480 has performance problems. Period. There is not much to discuss. Also, I'm not the one who is trying to justify in every way WHY the game runs poorly. For me it just runs bad. I'm not interested WHY. That's for developer's interests, not centainly mine.

Funny, I bet it's some kind of setting for you.

I'm running DA2 on a laptop with an ATI Mobility chip with 512 MB gfx RAM.... @ 45 fps.

Maybe looking into 'why' would help more than crying as if your diaper is full and want BW to change it for you.

You can't compare apples to oranges, so you can't compare notebooks to desktops and you can't compare low resolutions such those used on a laptop with an "high" resolution such 1920x1200. In fact, I'm not even discussing about the performance of the game on low end systems.

Considering I'm running at 1280x1024 with DX11 enabled with all settings on the highest possible, except for the high res texture pack, it's not that much of an orange. But that's not my point.

My point was that *maybe* if you'd ask *why* you'd be able to make DA2 run much smoother. *Maybe* if you'd go to the bathroom you'd get the **** out of your pants you'd get a clean butt, cause BW sure as hell ain't gonna change your diaper.

[Notick]

.

Modifié par Notick, 24 mars 2011 - 04:50 .

[Notick]

mjordan79 wrote...

Notick wrote...

Maybe
when I have a little more energy, I'll compile a list for you of average
FPS for each game in their current technology standing.

I've
been programming games sense the early GodWars/Dragon Realm MUDS, along
with an avid modder (Q2 Mod's, HL2 Mods, PlanetSide source-code mods,
Counter Strike mods...) C, C++, C#, Java, VB, Lisp (Autocad language).
I'm not just speaking from my ass either.

Having built over 75
computers for myself, and over 300 for clients... Hundreds if not
thousands of hours of research over the years... You'd be
surprised.

Use your time for better things. Such list is
not needed nor requested. Just look at a bunch of benchmarks. You'd be
even more surprised.
BTW, I don't need to show off my titles to
demonstrate a game runs poorly. For three reasons: it doesn't fix the
game. I may discover that someone has far better skills than mine. And
it's a very poor argument to a flawed discussion that doesn't add
anything to the discussion itself.

Then don't bring up how many games are 'on your shelf'....

Sorry if I seem argumentative, I am not trying to be. The problem I have with the basis of your post - the data is based off of one game. Also, in defense of all of the arguments, you're using generalities and personal  experiences, not hard data. (Even no data from your own testing...)

I recommend going here

Cycle through all the games they have listed, and see the average FPS of all of the games they have posted. The best running game was on the 580 GTx @ 52.2 FPS for Battlefield: Bad Company 2. The worst was 33.5 FPS on the 580 GTx in Civilization 5 (which, I would consider, a comparable game to DA2 in application)

Modifié par Notick, 24 mars 2011 - 04:50 .

[mjordan79]

Jazharah wrote...

Considering I'm running at 1280x1024 with DX11 enabled with all settings on the highest possible, except for the high res texture pack, it's not that much of an orange. But that's not my point.

Ok. That's the evidence the game has some sort of scalability problems, then. The exact reason I have opened this thread.

My point was that *maybe* if you'd ask *why* you'd be able to make DA2 run much smoother. *Maybe* if you'd go to the bathroom you'd get the **** out of your pants you'd get a clean butt, cause BW sure as hell ain't gonna change your diaper.

There are many threads on how to optimize the performance and WHY the performance are bad. And that's not my point either. The point is I'm just dissatisfied with the performance level of the game. I'm not claiming BW have to fix the game either. I'm just saying that I won't buy DA 3 if this kind of trend is maintained. As a customer, I have the right to complain on the product I have bought, right? Just consider this thread a feedback for the next game. But I see there are people who are more than happy to buy quad SLI systems to run under the average graphics. That's fine.

[mjordan79]

Notick wrote...

I recommend going here

Cycle through all the games they have listed, and see the average FPS of all of the games they have posted. The best running game was on the 580 GTx @ 52.2 FPS for Battlefield: Bad Company 2. The worst was 33.5 FPS on the 580 GTx in Civilization 5 (which, I would consider, a comparable game to DA2 in application)

Did you see the resolution those games are tested? 5760x1200 .... Three times the resolution I was taking for the comparison. :DNot only that, but they're using SLI, twice the power I was considering. Let's normalize the obtained values to a single 1920x1200 monitor with a single card: let's assume performance is a reversed linear function of the resolution (it isn't), that means the worst performing game runs at 50fps on a single 1920x1200 monitor with a single GTX 580. In fact, (50 / 3) * 2 = 33.3 fps, so the calculation is correct. Far above the performance of Dragon Age 2. And we're speaking about the WORST case with the worst performing game. You know 6 programming languages and you can't understand this?

Come on. Have you finished the arguments? You're quite good at mirror climbing. Do you practice it as a sport?

Modifié par mjordan79, 24 mars 2011 - 05:35 .

[Cobaltleo]

ok im running on a newly build machine. amd 955 PII x4, xfx Radeon 6850 4 gigs of kingston ram and windows 7 64 bit and at 1920x1080 im getting butter smooth graphics, with everything on the highest settings with only slight chop on boss fights if i have the high resolution textures enabled.(which honestly im not seeing a difference except on fenris's weapon handle if he's in the back when we enter the docks or lowtown). Is there any chance your issue is steeming from this being a amd/ati optimized game? similar to how crysis was optimized for nvidia cards?

[mjordan79]

Cobaltleo wrote...

ok im running on a newly build machine. amd 955 PII x4, xfx Radeon 6850 4 gigs of kingston ram and windows 7 64 bit and at 1920x1080 im getting butter smooth graphics, with everything on the highest settings with only slight chop on boss fights if i have the high resolution textures enabled.(which honestly im not seeing a difference except on fenris's weapon handle if he's in the back when we enter the docks or lowtown). Is there any chance your issue is steeming from this being a amd/ati optimized game? similar to how crysis was optimized for nvidia cards?

How much frames are you generating with your card?

[Irx]

DA2 uses a lot of heavy image precessing effects, problem is, the game still looks like **** and they are barely noticeable due to details sterile locations, low poly models and bad design - thanks to outdated consoles for that.

So yes, enabling 2010-era techs on a 2005-era console port barely improve the visuals and just wastes fps.

[vania z]

Notick
You just dont understand what you are talking about.  

Modifié par vania z, 24 mars 2011 - 09:34 .

[JironGhrad]

Haha, the real problem here is that NVidia uses too few Cuda shaders as compared to Radeons' shaders. The games with the AMD/ATI logo actually do run better with a proper number of shaders...

[Valaskjalf]

Just turn of AA . at 1920x1080 you shouldn't need AA

[mikey3k]

This game is horribly unoptimized on nVidia hardware. Bioware worked directly with ATI/AMD on the DX11 features no surprise that it runs much better on ATI cards. Bioware is supposedly working on a patch to improve DX11 performance so we shall see if they come through.

JironGhrad wrote...

Haha, the real problem here is that NVidia uses too few Cuda shaders as compared to Radeons' shaders. The games with the AMD/ATI logo actually do run better with a proper number of shaders...

Haha, um no. ATI worked with Bioware on this game so of course it would run better on ATI cards, just as some games where nVidia works directly with the dev team, nVidia cards run better.

Modifié par mikey3k, 24 mars 2011 - 09:24 .

[mjordan79]

Valaskjalf wrote...

Just turn of AA . at 1920x1080 you shouldn't need AA

That's the popular belief. In reality, 1920x1080 (1920x1200 in my case) needs AA just like other resolutions.
The resolution is higher but also the diagonal is bigger. In fact, the aspect ratio is the same. That means the pixels have the same dimension across all resolutions / diagonals. As a consequence, AA is needed just like is needed with smaller monitors / resolutions.

[JironGhrad]

mikey3k wrote...

This game is horribly unoptimized on nVidia hardware. Bioware worked directly with ATI/AMD on the DX11 features no surprise that it runs much better on ATI cards. Bioware is supposedly working on a patch to improve DX11 performance so we shall see if they come through.

JironGhrad wrote...

Haha, the real problem here is that NVidia uses too few Cuda shaders as compared to Radeons' shaders. The games with the AMD/ATI logo actually do run better with a proper number of shaders...

Haha, um no. ATI worked with Bioware on this game so of course it would run better on ATI cards, just as some games where nVidia works directly with the dev team, nVidia cards run better.

That's actually not true. NVidia claims that Cuda shaders are better... but there aren't any benchmarks that put ATI's hardware behind straight across at comparisons (Geforce GTX 590 vs HD6990 is a hardware generation straight comparison). ATI's GPUs have approximately 3 to 4x the number of Stream Processors that an NVidia GPU has. Go do a side-by-side comparison anywhere that sells GPUs (like Newegg). ATI's top-end GPU models (the xx70 and xx90s) also are clocked higher than the NVidia models (x80s and x90s) in practically all cases. The majority of instances where the NVidia logo is displayed as "better" are due to the PhysX engine performances... and if you add a PhysX stand-alone to a Radeon GPU you get massively better performance vs a standard NVidia card. Running all maxed out on the original DA: O, my HD4890 saw a 20 fps increase (~135 fps in FRAPS) with the PhysX card vs without it.

[vania z]

JironGhrad
Actually, nvidia is clocked higher because they use double frequency for cuda cores.
Nvidia has higher tessellation performance and better handle branching.
Amd has more raw math power, which is rather hard to use. That is why super computers use nvidia for calculations.

[JironGhrad]

vania z wrote...

JironGhrad
Actually, nvidia is clocked higher because they use double frequency for cuda cores.
Nvidia has higher tessellation performance and better handle branching.
Amd has more raw math power, which is rather hard to use. That is why super computers use nvidia for calculations.

The problem with this argument is that in game benchmarks, NVidia consistently scores lower than ATI when all the settings are maxed out. Try fact checking again, since ALL GPUs run at clock multipliers over the base bus speed. That's why they list RAMDAC on specs... because the base memory speed is fixed at 400Mhz and then you run it against a bus multiplier. Even the CPU's are bus speed multiplied, try looking at your BIOS.

This isn't even much of a real argument since NVidia rebranded their 8x00 series GPUs for two consecutive years without changing the hardware performance specifications at all, while ATI made incremental changes. There may not be a huge difference between HD2800 and HD3800 series cards, but they're still more advanced than the same generation (Geforce 2xx series) GPUs, and ATI continues to hold onto that lead in the place where performance actually shows... the benchmarks. As a side note to all this, my Radeon HD4890 actually out performs my good friend's GTX 560 in DX 10 mode. So it's not just the DX11 coding...

[mikey3k]

Again, ATI worked directly with Bioware hence why their cards are better optimized. The lead designer confirmed they worked with ATI directly. Just as Dirt 2 ran better on ATI cards because ATI worked directly with Codemasters for DX11. Just like in some other games (games WITHOUT physx) ran better on nVidia cards because nvidia worked directly with the developers.

And btw, ATI has rebranded cards as well (for example Radeon HD 4230, HD 4250, HD 4570, HD 4580, HD 4730 and finally HD 4750 were rebadged 3xxx series cards), both companies do it to save money and make profit. And the fact that you brought up something that happened, what 3-4 years ago? Update your talking points. You didn't even dispute his point about nVidia having better tesselation.

You sound like someone who believes that everything ATI does is good, everything nvidia does is bad. I buy the best performing card at the time I am looking to buy a card, regardless of the brand.  Some games ATI is better in benchmarks and some games nvidia is better. One brand does not always come out on top and anyone who says so is being intellectually dishonest or is a fanboy.

Modifié par mikey3k, 25 mars 2011 - 06:11 .

[VFTWV]

I'd say DX11 makes it run slower, I get average 30-40 fps on AMD phenom 1 9650 x4 2.4 ghz, 4 gig DDR2 RAM and Sapphire 6850 on max settings Dx11 1920x1080 with no AA.

[Valaskjalf]

mjordan79 wrote...

Valaskjalf wrote...

Just turn of AA . at 1920x1080 you shouldn't need AA

That's the popular belief. In reality, 1920x1080 (1920x1200 in my case) needs AA just like other resolutions.
The resolution is higher but also the diagonal is bigger. In fact, the aspect ratio is the same. That means the pixels have the same dimension across all resolutions / diagonals. As a consequence, AA is needed just like is needed with smaller monitors / resolutions.

mmm well people shouldn't be buying low res large displays

my 1920x1080 screen is a mere 21.5 inches diagonal. highest pixel density i could find that isn't on a laptop

[Cobaltleo]

in response to mjordan79 i havent fraps'ed it yet but im not getting any clipping, hitching or stutter at all on my setup. im using a 19" 4:3 aspect ration monitor @ 1280x1024 all settings maxed out. on my 40" 1920x1080P tv with the computer connected via hdmi cable i only get slight hitching with the high texture patch enabled. but to be honest im really not seeing any difference with or without the high textures except when you first enter a room and the camera is zoomed in on the who ever is in the back of the group. for example fenris;s sword handel looked low textured when i first entered his house. sry bout the delay in responding

[Cobaltleo]

oh and all seetings maxed out on the tv too. it might actually be running better with windows 7 though. last time i had the computer plugged up to the tv i was running windows vista 64. will give it a try tomorrow.

[Playarn]

I can run this game on the highest settings...
AMD Phenom II X4 955 3,2GHz (CPU)
Radeon HD6950 2GB (Graphics)
As I said, I can run the game on highest setting without problems, with DX11.
I havent reed everyone's comments but it may be your computer thats messed up, My friends NEW computer cant even run World of Warcraft... and as what I've heard isnt it so high req for "WoW". so its probably something wrong with hes computer. I have no idea, but u may have the same "problem" so to speak.

[Cobaltleo]

ok im now on my 1080p tv with hdmi cable. I downloaded fraps and ran from gamelon house to sundermount right outside the dalish camp and fought the spiders there. Enabled directx 11 Resolution set to 1920x1080 Everything maxed out and high resolution textures on got 17-20 fps in sundermount. With the high res textures turned off i actually lost a frame lol 16-21

[Cobaltleo]

i know will cause your system to take a frame rate hit in and of itself but im not sure of how much but if you figure at least 5 fps then im doing pretty good. i got a friend that has a system thats a step up from mine. Ill have him try it this weekend. and report back if anyone is interested. he's got the 6870 ver of my card and the amd 6x cpu. and 8 gigs of ram.