Resident Evil 5 View previous topic :: View next topic  


NFOHump.com Forum Index - PC Games Arena
Page 8 of 51 Goto page Previous  1, 2, 3 ... 7, 8, 9 ... 49, 50, 51  Next
mao5




Posts: 416
Location: Nanjing,China
PostPosted: Fri, 17th Jul 2009 10:43    Post subject:
JBeckman, can you fix the fps drops in Area 3 on ATI graphics cards when DX10 and AA eabled?
Back to top
JBeckman
VIP Member



Posts: 34981
Location: Sweden
PostPosted: Fri, 17th Jul 2009 12:42    Post subject:
Didn't a poster on the previous page say it was due to how Crossfire worked?
(Assuming you use Crossfire either via two cards or a X2 card.)

It'll perhaps be fixed with the 9.8 Catalyst drivers, bit of a flaw with only one driver release per month as the 9.7 drivers are more or less complete by now, 9.7 might help a bit though with performance in general. Smile

Other than that don't force AA and modes like Edge Detect (Or Tent) or Adaptive AA will take quite a hit on how the game is running but the console version is at 30 fps or so (Options even allows game to be locked to this or 60) so the game runs quite well on PC even at higher settings in my opinion.
Back to top
accountbroken
Banned



Posts: 226
Location: an italian living in the north
PostPosted: Fri, 17th Jul 2009 13:09    Post subject:
JBeckman wrote:
Didn't a poster on the previous page say it was due to how Crossfire worked?
(Assuming you use Crossfire either via two cards or a X2 card.)



Well he said so but i do have a single core hd4870 and having exactly the same issue.
Back to top
Ricky78




Posts: 400
Location: Pizza's country
PostPosted: Fri, 17th Jul 2009 13:15    Post subject:
metatron83 wrote:
Those with Nvidia cards should install the new forceware (190.xx), specially if you have SLI. It gives a big boost in performance.

I get an average of 55 FPS with all max, vsync on, no AA @ 1024x768 on an Athlon X2 7750 overclocked to 3.0GHz and an ATI HD3850 256mb.
AA 4x makes the frame rate go down to 30, most likely due to the lack of enough memory.

Disabling vsync makes the game go up to 100 FPS in some scenes, and since this is a port it looks like it's playing in fast forward Laughing

PC games shouldn't have that behavior. They should run smother, not "faster". But since it's a port from a console, where games are programed with a fixed FPS, we get this "fast forward" effect when going above 60 FPS without vsync enabled...


It runs faster than the normal game because it's a benchmark, this happens also with the Street Fighter IV benchmark, but not in the final product.
With the vsync on you get a false result.
Back to top
Cedge




Posts: 1480
PostPosted: Fri, 17th Jul 2009 13:25    Post subject:
metatron83 wrote:
But since it's a port from a console, where games are programed with a fixed FPS, we get this "fast forward" effect when going above 60 FPS without vsync enabled...


You assume wrong, and/or don't know what you are talking about. Console games are not really programmed with a "fixed FPS" any more so than PC games. They're on fixed hardware, sure, but that doesn't guarantee a fixed framerate.

And yes, the "fast-forward" behavior only exists in this benchmark version (because, ya know, it's a benchmark). It's fixing the animations to a certain number of frames, rather than a certain amount of actual time, and then seeing how fast it can render that.

The actual game will run as you would expect it to: smoother, not "faster."
Back to top
metatron83




Posts: 152
PostPosted: Fri, 17th Jul 2009 17:18    Post subject:
Ricky78 wrote:
It runs faster than the normal game because it's a benchmark, this happens also with the Street Fighter IV benchmark, but not in the final product.
With the vsync on you get a false result.

Vsync on is what prevents the "fast forward" effect, so in a way it fakes the result, yeah. But having 100 fps and a game running off caffeine is not a real measure of performance either...

And the variable benchmark is an actual gameplay session, not just a fixed scripted animation. The only thing scripted is the route the main char takes around the areas since you don't control it, but the enemies behave different in every run.

The same goes to Cedge. That one benchmark is not a fixed animation. And as it was said before, console games are programed to be fixed at 30 or 60 fps.
Try running a PSX emulator and don't lock the frame rate. You will get a ridiculous fast forward gaming session at +200 fps Razz
Back to top
Cedge




Posts: 1480
PostPosted: Fri, 17th Jul 2009 17:26    Post subject:
metatron83 wrote:
Vsync on is what prevents the "fast forward" effect, so in a way it fakes the result, yeah.

No. Vsync only "prevents the effect," because vsync always lowers the framerate. If you had an insanely badass computer, you could no doubt get the game to "fast forward," even with vsync.

Quote:
And the variable benchmark is an actual gameplay session, not just a fixed scripted animation. The only thing scripted is the route the main char takes around the areas since you don't control it, but the enemies behave different in every run.

Arg. I don't think you understand what I'm trying to explain to you.

Quote:
The same goes to Cedge. That one benchmark is not a fixed animation.

But the animations in it are.

Quote:
And as it was said before, console games are programed to be fixed at 30 or 60 fps.

They are limited to 30 or 60hz due to the nature of televisions.

Quote:
Try running a PSX emulator and don't lock the frame rate. You will get a ridiculous fast forward gaming session at +200 fps Razz

Which is exactly what is happening with this benchmark.
Back to top
metatron83




Posts: 152
PostPosted: Fri, 17th Jul 2009 18:07    Post subject:
Cedge wrote:
No. Vsync only "prevents the effect," because vsync always lowers the framerate. If you had an insanely badass computer, you could no doubt get the game to "fast forward," even with vsync.

I said vsync is what prevents the effect, not that it's the only thing it does...
And yes, it locks the frame rate by syncing it with the monitor's refresh rate.
Also, having a bad ass computer should not make a game run like that. having a shitload of fps should mean more frames rendered per second of gameplay, not an accelerated gameplay effect.

Anyway, let's leave it here since we either don't understand each other points or simply have different vision or conception on the matter. We could keep arguing for the whole day...
Back to top
Cedge




Posts: 1480
PostPosted: Fri, 17th Jul 2009 18:17    Post subject:
metatron83 wrote:
Anyway, let's leave it here since we either don't understand each other points or simply have different vision or conception on the matter. We could keep arguing for the whole day...


How about you compare the SF4 benchmark, with the actual game, and you will realize that the point you are arguing is completely senseless. The "fast forward" behavior is ONLY present in the benchmark. Everything else this discussion besides this fact is irrelevant. Case closed.
Back to top
LeoNatan
☢ NFOHump Despot ☢



Posts: 73196
Location: Ramat Gan, Israel 🇮🇱
PostPosted: Fri, 17th Jul 2009 18:21    Post subject:
SF4 is a bad example, because when running the game at 30fps it runs in "slow motion" as opposed to skipping frames. It would be the same if it ran faster than 60fps.
Back to top
tombata
Banned



Posts: 868
PostPosted: Fri, 17th Jul 2009 21:55    Post subject:
Wel it runs on full settings at 20 fps for me and on medium at around 31 and it's playable Surprised
Back to top
Aeon




Posts: 8700
Location: Netherlands
PostPosted: Fri, 17th Jul 2009 22:13    Post subject:
Twenty playable? I think 35 is playable...

Intel i7 6700K, RTX470 Super, Kingston HyperX Fury 32GB
Back to top
human_steel




Posts: 33269
PostPosted: Fri, 17th Jul 2009 22:33    Post subject:
Well, everyone has standards depending/constructed on the basis of their current rig.Very Happy
Back to top
JBeckman
VIP Member



Posts: 34981
Location: Sweden
PostPosted: Sat, 18th Jul 2009 11:18    Post subject:
Decided to try one last thing, adjusted the options.ini file and renamed the EXE, worked pretty well.




Everything is as high as possible but there's no AA, SLI=OFF was briefly adjusted to SLI=ON but it only hampered performance (Seems it's not a generic multiGPU option.) and the difference between 32-bit and 64-bit is that HDR=Low was set to HDR=High
(Not much else to adjust.)

EDIT: No perfomance didn't improve, variable benchmark plays different on each run.
Back to top
mao5




Posts: 416
Location: Nanjing,China
PostPosted: Sat, 18th Jul 2009 16:13    Post subject:
but still no aa
Back to top
Yondaime
VIP Member



Posts: 11741
PostPosted: Sat, 18th Jul 2009 17:12    Post subject:
⁢⁢


Last edited by Yondaime on Mon, 2nd Dec 2024 16:01; edited 1 time in total
Back to top
JBeckman
VIP Member



Posts: 34981
Location: Sweden
PostPosted: Sat, 18th Jul 2009 17:22    Post subject:
Quote:

But still no aa


Actually that was in the first result I posted, quite a hit during scene #3 (Smoothened out a bit after renaming the EXE as seen in the above image.)

http://www.nfohump.com/forum/viewtopic.php?p=1013070#1013070

Anyway the "score" or rather framerate isn't that important though renaming the EXE can help performance further for ATI hardware but it's still more than playable and it looks quite good although not that different from the console edition from what I've heard.
(Aside from perhaps the higher screen resolutions and AA possibilities.)

http://www.pcgameshardware.com/aid,690313/Resident-Evil-5-PC-vs-Xbox-360-and-DirectX-9-vs-DirectX-10-graphics-comparison/Practice/
Back to top
stoled




Posts: 13
PostPosted: Sat, 18th Jul 2009 17:28    Post subject:
it is coming to pc you can download the pc benchamark tool @
http://www.nzone.com/object/nzone_re5_downloads.html

looks like im slow today Razz

well I have yet to fully run the benchmark...but its time for a new vidcard I think

<h>I has a signature<h1>

Last edited by stoled on Sat, 18th Jul 2009 17:35; edited 1 time in total
Back to top
xExtreme




Posts: 5811
Location: 43 6C 75 6A 2D 4E 61 70 6F 63 61
PostPosted: Sat, 18th Jul 2009 17:33    Post subject:
ur pretty late ?
Back to top
stoled




Posts: 13
PostPosted: Sat, 18th Jul 2009 17:36    Post subject:
xExtreme wrote:
ur pretty late ?
lol yea im dead this morning
Back to top
LeoNatan
☢ NFOHump Despot ☢



Posts: 73196
Location: Ramat Gan, Israel 🇮🇱
PostPosted: Sat, 18th Jul 2009 17:37    Post subject:
stoled wrote:
looks like im slow today Razz

just today? bench was released 2 days ago Laughing
Back to top
stoled




Posts: 13
PostPosted: Sat, 18th Jul 2009 19:46    Post subject:
iNatan wrote:
stoled wrote:
looks like im slow today Razz

just today? bench was released 2 days ago Laughing
really Confused
I only found out yesterday Razz
Back to top
Yondaime
VIP Member



Posts: 11741
PostPosted: Sat, 18th Jul 2009 19:50    Post subject:
⁢⁢


Last edited by Yondaime on Mon, 2nd Dec 2024 16:01; edited 1 time in total
Back to top
mao5




Posts: 416
Location: Nanjing,China
PostPosted: Sat, 18th Jul 2009 20:01    Post subject:
JBeckman wrote:
Quote:

But still no aa


Actually that was in the first result I posted, quite a hit during scene #3 (Smoothened out a bit after renaming the EXE as seen in the above image.)

http://www.nfohump.com/forum/viewtopic.php?p=1013070#1013070

Anyway the "score" or rather framerate isn't that important though renaming the EXE can help performance further for ATI hardware but it's still more than playable and it looks quite good although not that different from the console edition from what I've heard.
(Aside from perhaps the higher screen resolutions and AA possibilities.)

http://www.pcgameshardware.com/aid,690313/Resident-Evil-5-PC-vs-Xbox-360-and-DirectX-9-vs-DirectX-10-graphics-comparison/Practice/


would you tell me the renamed name? maybe I'll tried it again.
Back to top
JBeckman
VIP Member



Posts: 34981
Location: Sweden
PostPosted: Sat, 18th Jul 2009 20:49    Post subject:
It's the MT Framework engine so I simply used DevilMayCry4_DX10.exe (DX9.exe would also work for DX9) though Crossfire isn't utilized much regardless but the framerate will be a bit more stable. Smile
Back to top
anti-flag




Posts: 246
Location: Brazil
PostPosted: Sun, 19th Jul 2009 04:46    Post subject:
omg just ran the benchmark here and this game looks amazing!
really really better than re4 pc...
is there much difference between dx10 and dx9 versions? running winxp here =/
Back to top
tombata
Banned



Posts: 868
PostPosted: Sun, 19th Jul 2009 04:55    Post subject:
I wanna fuck this girl with the main character! The one with the sexy ass! But she is just a texture Sad
Back to top
madmax17




Posts: 19374
Location: Croatia
PostPosted: Sun, 19th Jul 2009 10:09    Post subject:
Heard it tears really bad on the xbox version, microsoft should patch that shit like you press the guide button and you can change vsync off and on at will, I can hardly play Splinter cell: DA there is so much tearing, no tearing on the pc though.
Back to top
Yondaime
VIP Member



Posts: 11741
PostPosted: Sun, 19th Jul 2009 13:08    Post subject:
⁢⁢


Last edited by Yondaime on Mon, 2nd Dec 2024 16:01; edited 1 time in total
Back to top
Glottis
Banned



Posts: 6313
PostPosted: Sun, 19th Jul 2009 13:14    Post subject:
Yondaime wrote:
The main difference I see is that the PC version is so much clearer and sharper, must be the textures.

Main advantages over 360 version are: High res, high antialiasing, anisotropic filtering, high fps, no tearing. Textures are the same I think.

C2D E6750 @ 3.2Ghz, 4GB 800MHz DDR2 4-4-4-12, GeForce GTX 260 c216 OC 896MB, 3.2TB, Windows 7 Ultimate x64

Xbox 360 Elite, PS2 Slim, Xbox
Back to top
Page 8 of 51 All times are GMT + 1 Hour
NFOHump.com Forum Index - PC Games Arena Goto page Previous  1, 2, 3 ... 7, 8, 9 ... 49, 50, 51  Next
Signature/Avatar nuking: none (can be changed in your profile)  


Display posts from previous:   

Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB 2.0.8 © 2001, 2002 phpBB Group