 |
Quote: |
 |
|
|
 |
Originally Posted by bcruise |
 |
|
|
|
|
|
|
|
|
|
I disagree that there's input lag - lack of responsiveness is a little closer, but it's not the whole story. Instead I believe that it's the animations that give off the impression of things being delayed. And I can use the pump fake example to show it, albeit in a different way, with the help of my PC version.
For the purposes of this test, the shot meter is a raw display of how the game is reading my controller input. If I just tap the shoot button on a pump fake it doesn't prove much because the player will just go through his gather motion animation and bring it down. BUT....if I hold it as long as possible without shooting, then the outcome becomes VERY dependent on my precise input (letting off the shoot button) and you can really see that you have more user control than you think. And, with a real-time display of my controller inputs on the screen, I can demonstrate that.
(muted for copyright - not a big deal, considering I don't need sound for this)
I mess around for a bit at the start to show there's no funny business going on with the inputs, and start actually trying to do this around :33. So....the key to seeing this is the meter. Watch it start to move and then cut off as I cancel the shot into a pump fake at the last possible moment - the first good one is at :36 and there's a string of several in a row at :56. And at the same time keep an eye on the controller input. It's pretty much spot on IMO - the split second the input stops and that button goes off, so does the meter. Later in that second stretch of successful ones, my timing got a little thrown off and the pump fake starts started looking delayed because I was pressing Square before the previous one ended. That's a great example of what people see and consider "input lag".
I actually find the "animations cause input lag" claim to be a very familiar one - The Show is constantly berated for that too. But I feel like this approach results in a much more realistic-looking game - if animations were directly tied to our exact controller inputs at a given time, the game would look like a herky-jerky mess. I'll never fault SCEA or 2k for taking that direction, but I do understand why it draws the ire of gamers who want split-second reactions.
JMO. I was bored so I though I'd toss in my two cents.
If anyone knows another way to judge split second reactions within the game that isn't tied to animations, let me know and I'll try it.
|
|
 |
|
 |
|
The pump shot test is not a very good test in my opinion... Tho I must say its awesome that you took the time to put the video together ...
If there is a way for you to test this in live game action verses the CPU AI you may find truth our complaints ... Like one of the posters said a few quotes up.... The AI triggers the input lag aka lack of responsiveness or whatever you guys want to name it lol during actually live gameplay.
Do the same test if you want with an avg team vs a contender SS/ sim. 12 min qrt.. However you are on PC so not sure what you'll get .. Here's a link of a video I made a few weeks ago.. It highlights several problems with the game
..
http://youtu.be/_gGHVvLU2u0
1. You will notice the AI intentionally pulls you into the scores table to get an unfair advantage on fast break numbers. First I thought is was me, but after a while of testing its not me this is hardcoded into the game...
2. At 0:12 sec. You will see I have a possible fast break opportunity and as I make the outlet pass to Devin Harris .. Aka lead pass. He comes back to the ball as if I'm making a lead pass for him to come toward the passer Dirk... The result gives the CPU AI time to recover on defense.
3. At 0:37 sec. You will John wall running a simple double screen and roll play.. This is very easy to read however as I know the roll is coming I try to move Zaza Pachulla back into the paint to defend the roll.. At exactly 0:36 sec and you will see the AI ignores my commands on the left stick and makes Zaza jump toward Gortat. I don't get control of Zaza until after the layup is made.. You can see the block command I inputed trigger late after the layup is made.. This is a prime example of what me and others have been complaining about.. You guys can't really see it maybe because it's more of a feel than something visual you guys can see easily.
4. At 0:45 sec you will see very stupid defensive rotation programmed in that the user has to compensate for. You will see me switch to Parsons because I know the AI is about to pull him away from the guy he's guarding leaving an open uncontested 3pt. Since I negate this attempted cheat by the AI the devs have gone a step further and now the AI pulls Wesley Matthews a good defender off his guy so they can and will get the open shot lol.
5. At 0:51 sec you can see the magnetic pull into the scores table again .. More obviously you can see me trying to negate this by angling Parsons inwards... And still when the AI decided last to do so I go flying into the scores table thus giving them a 3 on 1 ... Fast break .. Well more of a 4 on 1 since my guys are slow as hell.
Then there are highlights you can skip as they are not relevant to the topic
6. At 2:25 sec.. You will see me run a play I call to get the ball to Dirk quickly when he's hot. I make the icon pass LT out of the double team to Zaza and the ball floats slowly like a lob pass allowing the defense to recover and contest the dunk. It's clearly not a lob pass input.
Sent from my iPhone using Tapatalk