The online racing simulator
How altering my graphics settings has changed my whole outlook on gaming.
Being pretty much a computer noob, I had never fiddled around with any of my computers performance settings. Following a post about how to optimise NASCAR 2003 computer graphics performance, by a fellow vMax Nascar driver, I decided to take a look at my graphics card.

My Computer Specification is:

Intel Pentium 4 Processor 3.4 GHZ
1GB RAM
360GB Hard Drive
GeForce 6200 TM 256MB
19 inch TFT Monitor

Following his advice on how to maximise settings, I changed my settings from the default "look what the application says" setting to the following:

Quote :Antialiasing settings: 8xS
Anisotropic filtering: 16x
Image settings: High quality
Color profile: Not available
Vertical sync: Manually, turned Off
Force mipmaps: None
Conformant texture clamp: On
Extension limit: Off
Trilinear optimization: Off
Anisotropic mip filter optimization: Off
Anisotropic sample optimization: Off
Gamma correct antialiasing: On
Transparency antialiasing: Supersampling
Tripple buffering: Manually, turned Off
Negative LOD bias: Clamp

Having done this my FPS has dropped from around 90-100 to 30-50. But the game looks so much better, it's like having a new game. I will definitely be doing this for other applications if only to check out the graphics.

I just wish that information on things like this where more readily available for normal people like me. People don't understand what they can do with their systems.

I would like to know peoples opinions on what I have done and whether my system should be running this, or running better? I seem to have a hell of a lot of processes running at the moment even though they aren't taking up much CPU power.

Could someone explain in layman's terms what I have done?
What FPS should I be looking to run in LFS?
Should I change my settings?
Quote :Antialiasing settings: 8xS
Anisotropic filtering: 16x

i personally feel this is overkill for just about anything.
Antialiasing (AA) is the smoothing or blurring of lines so that they don't appear jagged.
Ansiotropic Filtering (AF) is the sharping of textures as the move off into the distance.

having both set at their maximums is severly taxing your video system.
i find that in most instances an AA setting of either 2xS or 4xS (usually 4, but depending on the game, this could cause too much of a performance hit) is more than enough to smooth out the lines, and an AF setting of either 4x or 8x provides good sharpening of details in the distance without over sharpening. The lower settings also have an added benefit of better frame rates.
Yes, that's what I had before. I heard a rumour that the eyes can only detect differences up to 60 frames per second. Anything above that is undetectable?
60fps is closer to the minimum for fluid motion than it is to the maximum perceived motion. 24 frames per second being just about the minimum for fluid motion. also, don't confuse frames with fields. for traditional tvs, a frame is nothing more than 2 alternating fields (odd and even). computer monitors, lcd screens, and progressive display tvs don't use fields, they just draw the frame straight down the screen.

but lets forget all that and lets just say for the sake of arguement that the human limit is 60fps. a game running at 200fps is going to have "smoother" game play than the same game running at 60fps because the user inputs are being updated into the game enviornment more often...even if you can't "see" a difference, you'll "feel" a difference.
I think that if Your monitor supports high resolution, then turn it to cca 1600*1200 and turn off the AA , or lower it to 2x . It will look nicer, and maybe run a bit faster.
Quote from MataGyula :I think that if Your monitor supports high resolution, then turn it to cca 1600*1200 and turn off the AA , or lower it to 2x . It will look nicer, and maybe run a bit faster.

i disagree. in most instances a smaller resolution with AA and AF will look better than a larger resolution without...performance is similar.

1600x1200 0xAA 0xAF
http://www.tweaktown.com/popImg.php?img=AAAF18.jpg

800x600 4xAA 8xAF
http://www.tweaktown.com/popImg.php?img=AAAF17.jpg

sure, the image is smaller, the quality is better...jaggies are a huge detractor to image aesthetics.
Well, it depends on the size of the monitor. For example, on my 19" monitor it is really better to run LFS on a bigger resolution than 1024*768 , because on a big monitor U will see a huge difference between 800*600 with AA&AF and 1600*1200 with no AA&AF. For taking screenshots, it is different, because in my situation, i will see the picture taken @ 1600*1200 with no AA&AF on a 1024*768 res desktop with all the jagged lines, but actually U wouldn't notice it ingame , but U will see the pic taken @800*600 with AA&AF on a 1024*768 res desktop with no jagged lines, and U will think, hey, that looks good, but ingame, it isn't so good , imho (it is hard to say what i really mean,pls forgive me, if it isn't understandable , i need to learn english more :o )
no, i definately understand your point. but i think the smoothness/crispness that is shown in the still screen shots definately comes across in game. i agree you don't really notice the jagged lines in game, but but it still lacks a certain quality to it.

but i think this is an issue of personal preference, and we aren't going to convience each other to think differently

btw, your english is very good
@glyphon
Thank You !
#11 - Jakg
ok, just did a test using those setting with full aa and af, and one using the standard settings with no aa or af, pic one is alot nice methinks!

EDIT - Ok, jpeg compression probably wasnt my best idea!
Attached images
AAAF.jpg
Raw.jpg
glyphon, examples like that make no sense as they are not the same visual size
yeah i have a 19 inch lcd monitor. i run 1280x1024 with 4xAA and 8xaf sometimes 16xaf... i would use temporal but it kills framerate on certain parts of some tracks like kyoto ring

sempron 3000+ slightly oc'ed to 2.2ghz
x800gto bios flashed to x800xl....
768mb ram at 190fsb.


lfs looks very bad to me with no aa or af.
8x FSAA looks so much better than 4x. Less than 4x and you can still see pixelisation, so it's pointless. 4xAA and 8x AF is the minimum for me. I agree, it really does transform games.
Quote from Blowtus :glyphon, examples like that make no sense as they are not the same visual size

you completely miss the point...the point being that a smaller resolution with AA and AF enable can look better than a higher resolution without AA and AF enabled.
Quote from glyphon :you completely miss the point...the point being that a smaller resolution with AA and AF enable can look better than a higher resolution without AA and AF enabled.

That may be true, but it's slightly misleading when one picture is half the size of the other. Double the size of the 800x600 image and then compare them. I think thats what Blowtus was trying to say
/stop hijacking and arguing in my thread...
Quote from Decibel :That may be true, but it's slightly misleading when one picture is half the size of the other. Double the size of the 800x600 image and then compare them. I think thats what Blowtus was trying to say

that's not how it works. if i take that 800x600 jpeg compressed image, and resize it to 1600x1200, its going to look like ass. to compare them side by side, the larger one would have to be reduced.

you can discard information from an image to make it smaller, and still have good quality, but you can't do the reverse.

when you are playing a game at 800x600, it will fill up the whole monitor screen the same as 1600x1200. for me, 800x600 is 19" and 1600x1200 is 19". my monitor doesn't magically resize for the larger resolution.
/sighs
here you go...

i resize the 1600x1200 down to 1280x960 to match the version with AA and AF. the "smaller" resolution still looks better.

even with out it being resized there are things to look for about the image quality...

the fuel line running across the screen. in the 16x12 version its very jagged. and floor of the hanger becomes noticibly blurred in the distance. in the 12x9 version, the fuel line has been smoothed out and the hanger floor retains its details in the distance.
Attached images
16to12.jpg
1280.jpg
Quote from Chris_Kerry :/sighs

its not really highjacking...its all a discussion about how all the different variables can effect the image that you see onscreen. its a good thing
Quote from glyphon :
you can discard information from an image to make it smaller, and still have good quality, but you can't do the reverse.

That's not true, i see guys in those CSI shows do it all the time...

CSIboss -Can you enhance that detail in the back there ?
CSIlab - Ah, ya sure... hehe, looks like (insert witty pun-filled nonsense)

Heck, the other day i saw this guy enhance something like 4 pixel wide into a full screen hi-res video !

Ok, ok..

Sarcasm Mode: OFF
Quote from Fonnybone :That's not true, i see guys in those CSI shows do it all the time...

haha, yeah, the do that, and it bugs the carp out of me. they must be using a pentium x2 7300+, with octal-pipeburst multi caching, at least 16gb of pc 25000 ram, and a gf x93000XLT PE video card with 48 texture pipes and shader v.10.3 to be able to do that, which is well beyond what those of us without NSA double ultra secret highlevel security clearance have access to...that or they must be magicians :gandalf:
#24 - Jakg
the thing is that people still think that it's possible! you can just about make out an upside down sheet of paper on a b&W cctv camera in the back of moving car thats 4*4 pixels, he enhances it so you can read the text and then it is in colour!
Quote from glyphon :when you are playing a game at 800x600, it will fill up the whole monitor screen the same as 1600x1200. for me, 800x600 is 19" and 1600x1200 is 19". my monitor doesn't magically resize for the larger resolution.

That was my point
Most people viewing the sample images wouldn't have resized them to fit the screen, thus the 800x600 image still looks as though it contains as much detail as the 1600x1200 image, yet it only contains a quarter.

I'm not arguing about which looks better and i'm not tying to pick a fight with you Glyphon. I admire the fact that you have provided sample images to back up your claims. My only concern was that there may be a bias inherent in the way the samples were presented.

Sorry Chris, here's you thread back ....
1

FGED GREDG RDFGDR GSFDG