Upgrade warning (disappointment over dollars spent / performance increase)

DyerWolf said:
I have had my new 256 mb eVGA Geforce 6800 GT in for over a week, replacing my old 128mb GeForce4 ti 4800 SE. I got it on sale for $328 from MicroCenter.

Immediately noticed that my 3dMark03 and AquaMark benchmark scores increased significantly. Thought - wow this is great! I wonder what my games will look like?

The problem is, they looked the same.

I already had the graphics pretty much maxed out in FarCry, the Doom3 and Halflife Demos and was experiencing good quality game play. The only change I made to the graphics was to push FarCry water quality to Ultra - which was the only setting not already maxxed... The change in cards didn't affect my gameplay experience - perhaps a barely noticeable FPS increase. I say this, because while the FPS numbers may have increased, without the numbers being displayed, I really cannot see any functional difference.

What differences did I see? I can now run the 3dMark05 benchmark (the old card was not DX-9 compatible) but even with my new system, two of the tests run at 4 FPS (the second showing of the balloon thing and the second space battle...). Big whoop. It is not a game.

My warning is this: If you have a good processor and card - don't upgrade now. There are no games out there that need this level of processor power for the price they are charging.

I play DesertCombat (mod for BF 1942), FarCry, JointOperations Typhoon Rising, the two demos mentioned above, Rome Total War (which I won't open again until the next patch comes out) all of which can be fairly demanding of GPUs and CPUs. However, with no real increase in performance over what I had before, I caution anyone against rushing into a new purchase just because the reviewers and others are excited about the cards. None of these games need DX-9 functionality, none of them stressed my old card to unplayable framerates (EVEN AT MAX GRAPHIC QUALITY), and no current game is worth $300 plus just to get a few extra framerates...

So what advantage did I get? - Bragging rights. Also, I am future proof for a few years. That's it. Guess I won't have to buy another card for 3-4 years. I will keep the card I bought, but strongly recommend people looking to upgrade to either wait for the prices to drop significantly, or for the games to improve significantly. Without either happening, there is no reason for anyone to drop dime on a new card.

Just my thoughts.

Sys:

AMD Athlon 64 3000+
MSI K8N Neo Platinum NForce3 250 Gb
1Gb PC 3200
256mb eVGA Geforce 6800 GT

old: Geforce4 ti 4800 SE (Which is a kick ass card for the $$$)


LMAO!! WTF!

You have to be kidding

gg


-fro
 
WOW this has turned into a bomb fire. Let me start by thanking everyone that didn't in some way mock DyerWolf for his opinion you are the best among us. With that aside let me get to what I have to say about my opinion of all this. First DyerWolf I truly understand where you are coming from and let me say there is no way you are going to get through to any of these "high gamers" it just isn't possible. Second is it worth going from you ti4200 to a 6800 gt. The answer is simply no. Why? Well thats simple as well. You enjoy your games and that is what they are ment for (okay they are ment for making money, but you know what I mean hopefully) and that is awesome. Some people don't understand the concept of skipping a generation or two of graphic cards to save money. Plus there is no reason your ti4200 can't play the most current games, which with most of the current ones they are designed for the last generation high end models like the 9800xt and the FX5950. Games haven't even started to take full advantage of the new generation of cards yet except for games like Farcry that have a patch to supposibly take advantage of these extra features. So how would I know about value versus performance? I know because I build computers for all different types of gamers who don't care for the know how of building a computer. I have to look at every situation from everyones different point of view all the time. I can't make a person who can only afford a 9800pro go out and buy an x800 xt because they simply don't have the money even though in my head I am saying "Get the x800 you won't regret it, it would be a better choice!" Oh and one last thing about this whole nOOb calling issue. Since nOOb is used so much even a nOOb would know how to call a person a nOOb so calling a person a nOOb is getting kind of nOObish. Just something for all you to think about. Hey you read this so it has to be something you had to think about.
 
DyerWolf said:
The second point is merely a warning to the uninitiated: very, very few games require or even offer DX-9 functionality, fewer still can take advantage of AGP 8x or need 256 mb of video memory.

Very few games even offer DX-9 functionality?! If by "very few" you mean the hundreds of games out for the PC that now support it and its many different lighting/water effects, then yeah. Almost ALL new games support and take advantage of DX9, i.e. they look significantly different on a DX8 only card.
 
DyerWolf said:
I have had my new 256 mb eVGA Geforce 6800 GT in for over a week, replacing my old 128mb GeForce4 ti 4800 SE. I got it on sale for $328 from MicroCenter.

Immediately noticed that my 3dMark03 and AquaMark benchmark scores increased significantly. Thought - wow this is great! I wonder what my games will look like?

The problem is, they looked the same.

I already had the graphics pretty much maxed out in FarCry, the Doom3 and Halflife Demos and was experiencing good quality game play. The only change I made to the graphics was to push FarCry water quality to Ultra - which was the only setting not already maxxed... The change in cards didn't affect my gameplay experience - perhaps a barely noticeable FPS increase. I say this, because while the FPS numbers may have increased, without the numbers being displayed, I really cannot see any functional difference.

What differences did I see? I can now run the 3dMark05 benchmark (the old card was not DX-9 compatible) but even with my new system, two of the tests run at 4 FPS (the second showing of the balloon thing and the second space battle...). Big whoop. It is not a game.

My warning is this: If you have a good processor and card - don't upgrade now. There are no games out there that need this level of processor power for the price they are charging.

I play DesertCombat (mod for BF 1942), FarCry, JointOperations Typhoon Rising, the two demos mentioned above, Rome Total War (which I won't open again until the next patch comes out) all of which can be fairly demanding of GPUs and CPUs. However, with no real increase in performance over what I had before, I caution anyone against rushing into a new purchase just because the reviewers and others are excited about the cards. None of these games need DX-9 functionality, none of them stressed my old card to unplayable framerates (EVEN AT MAX GRAPHIC QUALITY), and no current game is worth $300 plus just to get a few extra framerates...

So what advantage did I get? - Bragging rights. Also, I am future proof for a few years. That's it. Guess I won't have to buy another card for 3-4 years. I will keep the card I bought, but strongly recommend people looking to upgrade to either wait for the prices to drop significantly, or for the games to improve significantly. Without either happening, there is no reason for anyone to drop dime on a new card.

Just my thoughts.

Sys:

AMD Athlon 64 3000+
MSI K8N Neo Platinum NForce3 250 Gb
1Gb PC 3200
256mb eVGA Geforce 6800 GT

old: Geforce4 ti 4800 SE (Which is a kick ass card for the $$$)



You've proved what I've been saying for longer than I can remember ... Some people just get upgrade crazy.. My 9800NP plays EVERYTHING I throw at it with no problems.

HalfLife 2 and Doom3 were both billed as games that would FORCE me to upgrade because they would be "unplayable" .. Not even close to a lag issue with my rig...

Now.. when we get per-pixel-rendered games rather than triangle/polygonal games we have now, maybe it'll be justified. Or when I decide to run halflife-3 at 3200x2400 or something.. but as it stands now, My lowly little 9800NP is MORE than enough.


and to all those people who say he's got some sort of driver issue, then his fps wouldn't have gone up, would it?

he's saying the even though he's got more fps, the games feel the same..

that's cause anything more than 30fps is unneccessary. this is a BIOLOGICAL fact.. we can't tell the difference anyhow.
 
Laforge said:
that's cause anything more than 30fps is unneccessary. this is a BIOLOGICAL fact.. we can't tell the difference anyhow.

not true, I would like to see rock solid proof that the human eye/brain cannot see any more than 30fps. There was a study conducted years ago. A test batch of fighter pilots could spot and identify an aircraft flashed on the screen for 1/200th of a second. Just because your eyes can't tell the difference doesn't mean mine can't.
 
Woblin_Goblin said:
not true, I would like to see rock solid proof that the human eye/brain cannot see any more than 30fps. There was a study conducted years ago. A test batch of fighter pilots could spot and identify an aircraft flashed on the screen for 1/200th of a second. Just because your eyes can't tell the difference doesn't mean mine can't.


just because it was on the screen for 1/200th of a second doesn't mean it didn't land in the 1/30th of a second "framerate" our eyeballs function under.

That same government that did that test also says this..

"When an image strikes the retina the nerve impulses last approximately
1/25 sec. This is why motion pictures appear to move continuously, even
though the number of frames/sec. is not extremely fast. "


http://www.ime.usp.br/~hitoshi/framerate/bigeye.mpg

showing a strobe effect on an iris.
 
Ok, sorry for the old thread, but I gotta give a big old EFF-U to the original poster....

I just got done installing my gf6800gt (evga) in my system. Didnt do drivercleaner, just removed old card and ran some quick benchmarks....

Old D3 benchmark gf5600 -- 18.7 FPS (demo1)
New D3 benchmark gt6800gt -- 94 FPS (demo1)

I couldnt tell IQ since I didnt have time to see the demo.

Dollars well spent ($320) You better friggin believe it

Ok, sorry for the flame, troll-post, whatever.... the adrenaline is getting out of my system.
 
Well I can see FPS increases up to at least 60. That doesn't mean I can look at a machine running 66fps and say hey your getting 66fps. No it means that I can tell when a game goes from above 60fps to 35fps. I can tell real easy. So I don't buy the BS about not seeing anything above 30.

As far as the older cards not being able to play new games. It's BS and we know that. It's not that the old cards can't display the same eye candy, (in some cases they can't) but they can't do it as fast.

I can certainly tell the difference between 1024x768 and 1600x1200 in a game. I can. As far as IQ goes, it's all about the AA and AF. That one is harder to notice but once you've noticed it, it's annoying. It will bug you forever. Your eyes lock in on the jaggies and it just irritates you. Some games I can't honestly tell the difference between 1280x1024 and 1600x1200, but I can tell when AA is on or off. That's where your eye candy comes into play.

User's of Geforce FX series cards that play HL2 know what it's like for the eye candy to not turn out right. As do Quake III users. Remember ATi rendered the sky properly and nVidia didn't? That's a huge IQ issue. The older cards in some instances can't even remotely approximate what the new ones can do.

New games like Farcry for example that support HDR and PS3.0. If you've never seen that, your missing out. No other cards can do that except the 6800GT and Ultra at playable frame rates. Not in anything above 1024x768 at least.

So there is a difference. Now and in the future. If you're happy playing UT2004 on a Ti4200 at 1024x768 then great. Don't buy high end cards then. I saw a huge difference going from a 9600Pro to a X800Pro and a 6600GT to a 6800GT.

But it DOES make a difference and if you increase your resolutions and add the AA and AF it will make a difference.
 
Meh, I went from 9800pro to 6800 ultra and I sure noticed difference... My comp got hotter! :D hehe but really, I had to oc 9800pro a lot to pull through things that 6800 just ate for breakfast. Besides, I am now able to play at 1600x1200 res and I am not ready to give up that :cool: 9800 struggled in FarCry and new gen games that utilize DX9 a lot. The original poster had to use DX8 path in all new games , so speed-wise GF4 was the same as GF6800. THe thing is, GF6800 does more with new games than GF4 can ever dream about ;) If he doesn't see the difference. well the upgrade ain't any good for him. He'd have best bang for a buck sticking to GF4 - but it's not for everyone.
 
Could we just end this thread now? Because we all know that we got a performance boost, but he didnt and can leave it at that...
 
I don't think the computer enthusiasts are finished flaming him.

EDIT: Oh and, isn't the "ti4800" an AGP 8x version of the ti4400?
 
the problem with the whole 'eyes cant see above x fps' arguement is interesting because we are talking about fps AVERAGES here.


if the game is averaging 30fps on a benchmark that means that around half the time, the card is performing UNDER 30fps.. I think 70fps avg is a minimum for a multiplayer game because you won't get many dips below 30 if your avg is 70.
 
DyerWolf said:
Originally Posted by DyerWolf
I have had my new 256 mb eVGA Geforce 6800 GT in for over a week, replacing my old 128mb GeForce4 ti 4800 SE. I got it on sale for $328 from MicroCenter.

Immediately noticed that my 3dMark03 and AquaMark benchmark scores increased significantly. Thought - wow this is great! I wonder what my games will look like?

The problem is, they looked the same.

run FarCry at 1024x768, disable AA, set AF to 8x, then add:

"e_vegetation_sprites_distance_ratio 100.000000" "r_hdrrendering 7"

to your farcry shortcut and see if your ti4800 could do that

should look like this if you have it set right:
http://www.reflectonreality.com/images/hdr/fc1.jpg
http://www.reflectonreality.com/images/hdr/fc2.jpg
http://www.reflectonreality.com/images/hdr/fc3.jpg
 
I went from an oc'd 4400 to my 6800gt-oc and can tell you there was a difference.. maybe some games to some people it just isn't noticeable but there is a difference between the cards and I can say that the difference is enough to make it worth the cost.

I play cs 1.6 and source and in 1.6 with my 4400 oc'd and everything turned to its lowest setting i would get smoke lag...anytime smoke was around it would drop my frames from 100 to 25 or 30.... with my 6800gt-oc, i have everything at it's highest setting and run a constant 100fps no problems no matter what is going on and how many smokes i chuck out...
with cs source, i ran my 4400 with everything turned down...averaged 50-60fps and the framerate you could see get hindered by all the surroundings and movements... with the 6800, its a constant 100fps and again, everything is cranked and nothing slows it down...

I suppose it depends on the person viewing the information and the way they feel the game should play that makes upgrading worth it or not.

If you didn't find the upgrade worth it, sorry, it is a helluva card and I personally think the upgrade I did was well worth the cash.... if you have the ability to play cs 1.6 and then source with those cards, it is a good test of what the difference is that makes one card superior to the other.....
 
I am not one to say my old 4200 is as good as my new 6800, nor has any one here I think, but man my old 4200 sure did play games nice!! :cool:

FarCry2004-10-1323-29-23-89.jpg
 
Regarding the rediculous arguement about "human eyes cant see more than x fps..."

Load a DVD up. It should play at ~ 30 fps. This is film speed (very roughly). Now hit 2x FF on the DVD player. This is roughly 60 fps, or closer to videotape speed. Now think, doesn't program shot on videotape look different than when shot on film? This is but one of thousands of techniques filmmakers use, the visual difference between film and videotape.
To all who think your vision is limited to 30 fps; I offer to sit you in front of 2 of my PC, one running Quake3 at 30 fps, and one running it at 60 fps, all other quality settings the same. You WILL see a difference between 30 fps and 60 fps, and it has nothing to do with screen refresh. Just try it. Try the Quake example, and the DVD example, you'll see the diff.
 
MAngelo said:
Regarding the rediculous arguement about "human eyes cant see more than x fps..."

Load a DVD up. It should play at ~ 30 fps. This is film speed (very roughly). Now hit 2x FF on the DVD player. This is roughly 60 fps, or closer to videotape speed. Now think, doesn't program shot on videotape look different than when shot on film? This is but one of thousands of techniques filmmakers use, the visual difference between film and videotape.
To all who think your vision is limited to 30 fps; I offer to sit you in front of 2 of my PC, one running Quake3 at 30 fps, and one running it at 60 fps, all other quality settings the same. You WILL see a difference between 30 fps and 60 fps, and it has nothing to do with screen refresh. Just try it. Try the Quake example, and the DVD example, you'll see the diff.

I do not under stand your DVD example here. Most TVs only run at 30 FPS so what am I to be seeing when I hit the fast forward? The video looking twice as fast? Maybe you got a cool HD TV that can show video noninterlaced.

But if you take a pre-rendered scene from the movie Final Fantasy and run it on your HD TV at both standard 30 FPS and de-interlaced 60 FPS will it look different? Not really because the source material was shot for only 24 FPS then every 6th frame is doubled up when they convert it to work w/ TV refresh rates so the resulting frame additions then doubling will make the scene look the same to your brain.

This is when we can start talking about what the human eye can and can not see. Set your PC monitor to the lowest refresh rate it can go. Maybe 60 or so. Use it like this for a day or two. Then crank up the refresh to 80, 85 or 120 if you can. Even if you can not "see" the difference w/ the naked eye, like I can, you can feel it after a full days work. On the higher refresh rates you will still have fresh eyes after work, but at 75 or lower I get a headache every time!

Another quick test to see this for your self, works best w/ a TV or cart monitor, is to hold your hand infront of you while looking at the wall and shake it, your hand not the wall. Now hold your hand the same way infront of your TV/monitor and shake it. See how this 2nd time you see ghosting of your fingers in the monitor? That is your eyes picking up the missing visual data it was getting when you where just looking at the wall, which has an unlimited refresh rate. :cool:

No quick put your hand down before all the people in your office start wonding about you even more. ;)
 
Did you do the experiment, or are you talking theory?? It works a lil different than you think.
if your worried about refresh, try it on a PC w a DVD and a good monitor, set monitor refresh high as poss, then run the movie at normal and at 2x, you'll see at 2x it looks more like videotape, which was my assertion. Videotape clearly looks different than film, whether you view it on a TV, PC monitor, whatever. And the difference you see between the two is the speed at which it is shot. :)
HTH


This may help http://www.editorsguild.com/
 
real gamers want more detail at playable framerates. not just higher framerates.

how about this question?
why did you buy the card? you knew you had playable framerates and your detail was at max. and you dont game at higher resoultions.


thats why i think you are stupid. because you bought it without knowing what you needed, wanted, or got.
 
i've upgrade from ti4200 to 9500@9700 to 9800se@9800pro to 6800nu.. i still have 9500@9700 n i can tell the big diff between 9700 n 6800..
ti4600 to 6800gt is a big jump..
 
Try playing Painkiller:Battle Out Of Hell with dynamic lighting enabled if you want to stress your card. It's awesome!
 
asmielia said:
Did you even bother turning on AA or AF? That makes a huge difference and any card pre-Radeon 9700 had serious trouble turning those on without a major penalty.

I think you need to re-evaluate how you're comparing the cards. Look at this chart:
Tom's Hardware VGA Chart: Far Cry

Go to the second chart on that page. At 1024x768 with 4xAA and 4xAF in Far Cry, the difference between your new 6800GT and a Ti4600 (roughly your card) is 53.4 to 12.6!!

I don't care who you are, anyone would notice that difference and would enhance your gameplay experience substantially.


a ti4800 is the same as a ti4400.....
 
i noticed it did the same thing with farcry for me, the max settings werent really max settings, so i went to the configure farcry shortcut and i went threw the advanced settings, noticed everything was on close to lowest settings. just manually select the highest settings, and it should be great
 
LOL 30 fps normal and 60fps fast forward, the images are being forced by 2x as fast....most vcr's start skipping the images by small amounts u wouldn't notice....its just a bit different that putting out 2x the fps normal speed...

How we perceive fps differs based on situation. In high action in a larger area(18" monitor within 1 1/2 feet away is 'big'), our eyes can go as low as 16fps, because theres so much to take in, so much going on. Our brains are VERY parallel in how they work, which is great for things like speaking and stuff. I read(and w/o sources I look like a moron) that the human brain operates at the rough equivilent of a few mhz. (Computer-boxes these days work at a few billion hz...most communications in serial).

Sure you can compare yourself to pilots, but they can see 50-70 and these are people TRAINED to be perceptive. Normal gamers are only trained sort of... We can't perceive more than 35-45 tops in average game play where you're turning around, running around, looking for things. IF you were standing still, not doing anything no action you might see 50-60...dead still 100% focused.

They did this study on people where they put them infront of a white wall for a long time and randomly for 1/4 second flashed the light off(so it went black) the person noticed. Then they used the opposite colors, and they noticed. if it was 1/4 or less w/e they did...that means they can see rediculously high fps.....but when they were forced to do other things or move around, or pay attention to details, they couldn't see as many. The conclusion for fps in eye sight is that dips and sudden drops in FPS are far more important than high or low often. EG your eye will notice a drop of 20 from 50 to 30 pretty well, and from 30-16 quite well. But if your system is low end(ti) it'll dip more as games throw more complex things at the card. Otherwise it could very well render everything and get it up fast enough you wont notice....but there will be more dips in older hardware in newer games. Thats all there is to it
 
You must not have your settings correct or you have the drivers installed wrong. I remember the day I went from a ti4200 too a 9800pro and it was a huge difference, much less going from a ti4200 too a 6800gt lol.
 
Back
Top