This is how badly optimized TF2 has become.

[J1M]

LOL wtf you can't tell the difference between 30 and 60 fps?

Maybe you are having micro seizures like on house.

 

[desocupado]

He's right. There's fuck all difference even between 30 and 60. And thats with as blatant a display as possible. You'd never discern the fps difference between two different games, or even the same game doing different things. Even if you've got better eyes than me (Or your brain is shittier at filling the blanks), more than 30 is still fucking meaningless mechanically. Human reaction time is about 1/10th of a second at best anyways. You're not going to be fucked over because you noticed something 30 milliseconds later due to a lack of fps.

Same kind of poser that thinks they can taste the difference in purity of bottled water.

It's not about reaction time, you fucking idiot, it's about fluidity of the image.

OF FUCKING COURSE I can notice difference between games or the same game. That was the fucking reason I fired FRAPS when playing Risen. Because performance was fine, but at some point shit got slightly jerky, and I got curious too see if the performance was dipping in some points. And it was.

 

[Johannes]

And it isn't just about having a nice fluid looking game, it affects your mouse accuracy too.

 

[Black]

Oh yeah, I remember when I switched from 30 fps (because I also believed that dumb myth) to 100 fps in counter-strike. The difference in aiming and fluidity of gameplay was something new.

But I've got the feeling that this bs won't die simply because some (if not vast majority) of console games run at 30fps. Sad peasants

 

[Black Bart Charley]

He's right. There's fuck all difference even between 30 and 60. And thats with as blatant a display as possible. You'd never discern the fps difference between two different games, or even the same game doing different things. Even if you've got better eyes than me (Or your brain is shittier at filling the blanks), more than 30 is still fucking meaningless mechanically. Human reaction time is about 1/10th of a second at best anyways. You're not going to be fucked over because you noticed something 30 milliseconds later due to a lack of fps.

Same kind of poser that thinks they can taste the difference in purity of bottled water.

It's not about reaction time, you fucking idiot, it's about fluidity of the image.

It is most certainly about reaction time. It is just relative to your opponent. If he is running with 100 fps and you with 30, and you both spot each other at the same physical time and have the same reaction time. He will fire first...

 

[desocupado]

He's right. There's fuck all difference even between 30 and 60. And thats with as blatant a display as possible. You'd never discern the fps difference between two different games, or even the same game doing different things. Even if you've got better eyes than me (Or your brain is shittier at filling the blanks), more than 30 is still fucking meaningless mechanically. Human reaction time is about 1/10th of a second at best anyways. You're not going to be fucked over because you noticed something 30 milliseconds later due to a lack of fps.

Same kind of poser that thinks they can taste the difference in purity of bottled water.

It's not about reaction time, you fucking idiot, it's about fluidity of the image.

It is most certainly about reaction time. It is just relative to your opponent. If he is running with 100 fps and you with 30, and you both spot each other at the same physical time and have the same reaction time. He will fire first...

We're talking about fractions of a second here, the lag between his mouse and his computer should be bigger than that. Really, that situation you are supposing should never happen, because the time frame for the syncrony (sincrony? How do you spell that?) is 1/30 of a second.

Second, shouldn't the game be syncronous? I mean, what difference does it make the rate at which your video card updates the screen? At worst you're going to see something 1/30 of a second later than another guy I guess.

Third, even if that's true, that surely isn't the issue for me, and I would guess not the majority of people either.

 

[Destroid]

Most LCD monitors only run 60hz and their pixel response times won't be fast enough to accurately produce 100fps anyway, not to mention the ~10-30ms input lag on a modern LCD display.

 

[Black Bart Charley]

Well, if the monitor is 60 hz than its almost negligible cause you cant go over 60 fps. But 100 fps are most certainly an advantage. Lcd, mouse, net lag or any lag dont play a role in a direct comparison since you both have the identical one. We are comparing fps here. 1/30 of a second is 33 ms, man. Back in the day(tm) people ran CS in 16 bit for more fps. So you hang behind the other dude by ca 20 ms. It can be an advantage. I play SF3rd strike with a dude and his character has a low sweep that has 250 ms runtime. I can parry it safely everytime in a testing environment, with my reaction time of 150 on a good day. In a fight though, it gets much harder. If we were to play on 2 different monitors, I'd have 20 ms less time to react which would crap all over my success rate of 50% or so. And that low sweep is a launcher for a bigger ass whooping...

sf is capped to run at 60 fps =)

Third, even if that's true, that surely isn't the issue for me, and I would guess not the majority of people either.

WELL IN THAT CASE MASTER GOD EMPRAH I APOLOGIZE

 

[Unkillable Cat]

Speaking of TF2... their latest money-grabbing scheme is selling 3 exclusive hats for charity, the relief aid for Japan, to be precise. One for $8, one for $18 and one for $100.

Even though they state that all profits from the sale goes to the relief efforts, I am having a very hard time believing that. I'm having an even harder time believing anyone will fall for it.

But the amount of in TF2 now is humongous.

Also, the Meet the Medic video is almost done. At least I don't have to play the game to enjoy that.

 

[S_Verner]

No. The scam is that Valve is going to try and claim the hat-money as a tax writeoff.

 

[Damned Registrations]

30 FPS is 33 ms/frame. So if you're comparing running at 30 FPS to 300 FPS (LOLZ guys 10 times betterz!) you're talking about a potential difference of 30 ms, but on average the difference will only be 15 ms. But since both monitors are likely capped at 60 FPS (17 ms/frame) the actual average difference is 8ms. If you're playing online, your fucking ping is probably changing by more than 8ms from minute to minute. And it's probably in the neighbourhood of 100-250. On top of that network delay, your physiological delay is going to be at least 100ms. So you're talking about an 8ms delay making a difference when your delay is already varying between 200-220 based on internet connection, having to blink once in a while, and general human variation. And this is being very conservative. I gave up playing online fighting/fps games because I'd have a fit of rage every time some douchebag from third worldia joined a game and the latency became 600ms.

So your response time to anything the enemy does is already ~210ms. So from a 'two people round the corner who fires first?' angle, it's irrelevant. The guy with the better reflexes, or who had a view of the other guy 30ms earlier because he had a slightly longer gun barrel, which vary a lot more than 8ms.

So how about on something like parrying in a fighting game? Now you have a specific window of time to act in. It's not even a matter of being before the other guy any more. Now you have to react within a relatively long timeframe (~250ms) but execute your reaction precisely within the last 66ms of that. Which makes the fps completely irrelevant, since you're predicting that window based on however fast the move is, and it takes the same amount of time whether it has 300 frames or 10. Except fighting games are locked to specific frame rates anyways, because the game needs to know where the hit boxes are during each frame, and it won't work if one guy is missing half the frames.

And this is with 30fps on the low end. Between 60 and 100 FPS, the difference is 3ms on average. What the fuck were you going to do 3ms before the other guy? That's a third of a frame even at 100 FPS. The game isn't going to register a difference that small.

Oh yeah, it'll take sound about 3ms to travel 1 meter btw. Hope you're wearing headphones or you'll get raped by sound delay n00b!

 

[MetalCraze]

FPS is not the main problem.
The main problem is FPS changes.

When it stuck at 35 FPS you won't even notice that it's "low" (which 35 FPS isn't)

It's when it suddenly changes from 35 to 50 that you go FFFFFUUUUUU

That's why all games on XBox360 are locked at 30 FPS.




Then again if one baaawws that 72 FPS is low he is clearly retarded. Chances are one's shitty monitor can't show more than 60. And he won't tell the difference between 50 and 60 either.

 

[Johannes]

The guy with higher fps in the corner scenario will be marginally faster, not because he sees it few ms earlier, but because his click registers that few ms faster.

 

[Damned Registrations]

That's a valid point. Having FPS capped at 40 would be much more useful than having it flop around between 35 and 50.

Of course, this would just give developers an excuse to make the framerate lurch between 25 and 40, cause hey, it's 40 most of the time, and the graphics look way cool in our trailer this way.

 

[S_Verner]

What dark sorceries did the TRIBES devteam use to make their game playable on 18kbps modems?

Why can't modern devteams be as good at making netcode as the TRIBES team?

 

[MetalCraze]

The guy with higher fps in the corner scenario will be marginally faster, not because he sees it few ms earlier, but because his click registers that few ms faster.

Your ping to the server will be higher than the difference between your stupid FPSes

That's a valid point. Having FPS capped at 40 would be much more useful than having it flop around between 35 and 50.

Of course, this would just give developers an excuse to make the framerate lurch between 25 and 40, cause hey, it's 40 most of the time, and the graphics look way cool in our trailer this way.

I wouldn't mind games implementing customizable frame limiter where a player will be able to set any number of frames he wants.
Because if your videocard runs the game no lower than 35-40, but sudden jumps to 60 annoy you - limiting it to 40 will be the best solution as FPS will indeed seem much smoother.

 

[Black Bart Charley]

30 FPS is 33 ms/frame. So if you're comparing running at 30 FPS to 300 FPS (LOLZ guys 10 times betterz!) you're talking about a potential difference of 30 ms

Are you talking to me? I was comparing 30 vs 100. Thats where the 20 ms came from. 33-10 = 20

But since both monitors are likely capped at 60 FPS (17 ms/frame)

I had a crt just 2 years back. But thats irrelevant, the points was a fundamental one.

And it's probably in the neighbourhood of 100-250.

I have 10 ms to all german servers and all other germans with fastpath too.

On top of that network delay, your physiological delay is going to be at least 100ms. So you're talking about an 8ms delay making a difference when your delay is already varying between 200-220 based on internet connection, having to blink once in a while, and general human variation. And this is being very conservative.

Its all irrelevant. We are making a direct comparison here.

I gave up playing online fighting/fps games because I'd have a fit of rage every time some douchebag from third worldia joined a game and the latency became 600ms.

I quit because playin Guy vs Gen in SFA3 with 5 vs 40 ms ping was not representative of my skillz since in person we had a decent 10:10 ratio but online I lost 2:8

So from a 'two people round the corner who fires first?' angle, it's irrelevant. The guy with the better reflexes, or who had a view of the other guy 30ms earlier because he had a slightly longer gun barrel, which vary a lot more than 8ms.

Probably, but since its 20 ms, it could stack up in a high skill Quake 3 railgun match. Or CS for that matter. Dont know about TF2.

So how about on something like parrying in a fighting game? Now you have a specific window of time to act in. It's not even a matter of being before the other guy any more. Now you have to react within a relatively long timeframe (~250ms) but execute your reaction precisely within the last 66ms of that. Which makes the fps completely irrelevant

100 fps

30 fps (its supposed to be 30 ms in the picture)

Except fighting games are locked to specific frame rates anyways

I wrote that already. But -> fundamental issue.

Your ping to the server will be higher than the difference between your stupid FPSes

You are a fucking moron. Go play Kotor again.

 

[Ruprekt]

Racofer I have a new PC.

It's time you upgraded too. It is glorious.

 

[racofer]

Why would I upgrade if my PC can run it maxed out with high fps all the time?

 

[Destroid]

For fighting games you absolutely need the minimum latency possible, I couldn't imagine playing a fighter online with the usual 50ms or so ping we have to servers here.

But what you describe BBC is also a worst case scenario, the average delay should be roughly half the frame update rate.


Personally, I've played most games with less than 40fps and often going down into the 20s and with Vsync on because I hate frame tearing, but it doesn't impact my performance much because I'm a dreadful sniper (thus single pixel accuracy doesn't matter to me) and my play style isn't much effected by the minute delay.

 

[TheWesDude]

can the human eye discern much over 32.5 frames-per-second?

no, as long as they are drawn instantly and consistently without any kind of delay.

unfortunately as we moved away from analog displays ( CRTs ) and more twords digital displays ( LCD/Plasma ) it introduces whats called "frame delay" which can at times be upwards of 1.75 of your FPS your monitor is displaying and what your graphics card is pushing out.

to get a totally flawless display on a digital display at 32.5 FPS, you would need to be drawing 1.75 times that on your GPU, which means you need:

32.5 * 1.75 = 58.6 FPS your GPU is pushing out.

so to not be able to tell any kind of difference on a digital display drawing 32.5 a second ( hz ) your GPU has to be getting at least 58.6 images.

to get an accurate 60 FPS on a digital display, your GPU would have to be pushing:

60 * 1.75 = 105 FPS

so to get a "true" 60 FPS ( hz ) on a digital display, your GPU has to be drawing 105 FPS.


it has NOTHING to do with your eye "seeing" a difference between 33 FPS and 60 FPS, its because of the inherent problems with using a digital display.

when you use an analog display, you have to be drawing not 1.75 times the attempted display FPS, you rather need 1.10 times the display FPS comming out from your GPU.

its not the FPS, its the shitty digital display methods that cause the problem.

 

[Destroid]

Link please.

 

[MetalCraze]

You are a fucking moron. Go play Kotor again.

Me?

Seriously saying that 5 or 10 or 20 ms make all the difference in the world in a multiplayer game is the most retarded thing to come from you.

Or did I hurt your "I'm leet player I can react within 0.005 seconds to whatever happens around me" dumb ego?

 

[Norfleet]

Oh yeah, I remember when I switched from 30 fps (because I also believed that dumb myth) to 100 fps in counter-strike. The difference in aiming and fluidity of gameplay was something new.

Myth. Most games themselves do not actually run at that many cycles per second. If the game were truly running at 100 fps, you'd be getting 100 updates a second. If the graphical frame rate exceeds the game's processing frame rate, the benefit you derive from this is precisely zero, because it's going to be the same frame getting pointlessly rerendered, as the game itself does not generate new information to render in that time. Even if you are a robot and thus unencumbered by human perceptual limitations, it won't help you to reprocess the same thing over and over. Your belief that changing the frame rate somehow improved your skill was purely a fallacy: Your skill improved because you had experience in the game. The fact that you changed the frame rate to a higher rate changed nothing at all, but you mistakenly attributed your improvement to this change instead of simply experience.

So your response time to anything the enemy does is already ~210ms. So from a 'two people round the corner who fires first?' angle, it's irrelevant. The guy with the better reflexes, or who had a view of the other guy 30ms earlier because he had a slightly longer gun barrel, which vary a lot more than 8ms.

You wanna know the real answer to the "who fires first"? It has nothing to do with the speed of light, the speed of sound, the gap between when a frame is rendered and reaches you vs. when it actually occurred...the guy who shoots first is the paranoid fucker who was going to shoot anyway, and indeed, had already shot by the time you realize he was there at all. Network latency is maybe 150-300ms. By the time you see ANYTHING, it ALREADY OCCURRED IN THE PAST, 150-300ms ago! This means if some paranoid fucker opens fire as he turns the corner, by the time you even SEE him, YOU ARE ALREADY DEAD. You died a quarter of a second ago, before the frame even reached your computer. No matter how fast you are, you cannot change the past. You're already dead. You turned a corner and caught a bullet in the face that was already in flight, either through the air, or through the tubes, before you turned that corner.

That guy? He wasn't reacting to your presence at all. He was reacting to the POSSIBILITY of your presence, anticipated that you WOULD be present, and pulled the trigger on the hunch that you might be there. This is what separates GOOD players from dead players. He didn't shoot you because he saw you turn the corner. He shot you because he figured you WOULD turn the corner, even though he didn't see you do it.

How do I know this? Because I've done precisely this. Back in the old days text gaming and modems, when frame rates didn't really exist, I could out-quickdraw someone on a high-speed link with only my shitty modem link, even though he used client bot triggers and I was manual. How did I outshoot someone with a quarter of my latency and a reaction time no human could match? Simple: I didn't wait for him to appear to respond to him. I shot on gut instinct, knowing the guy would walk into it. ALL TRULY GOOD PLAYERS DO THIS.

 

[Multi-headed Cow]

http://www.teamfortress.com/hatless/

So did they fix everything?