corduroygt
Banned
Perhaps someone needs to make an analog foot controller for foward, back, right and left (basketball size foot trackball?).
Ask and you shall receive
Last edited by a moderator:
Do FPS fans want to go motion-based direct-aim?
- Thread starter Shifty Geezer
- Start date
Ask and you shall receive
So you're essentially voting against options for people since it wouldn't be mandatory? That sounds brilliant.
Prophecy2k
Veteran
I voted "Yes" as i don't think tradpads should die for FPS use, but i do believe that motion controls can bring something fresh and new to the experience... not just for aiming but also for stuff like throwing grenades etc (as has already been said).
I'm wondering however if a game could work which uses head-tracking on the PSEye for controlling the viewpoint rotation, cursor pointing with the move, and lateral (forward/backwards/strafing) with the Nav controller?
I'm sure it'd take a bit of getting used to, and you may get a bit of neck ache, but so long as the head-tracking gives you enough effective "dead-zone" i'd be curious to see how well it would work.... it'd certainly "feel" pretty immersive and natural for the player ;-)
I'm wondering however if a game could work which uses head-tracking on the PSEye for controlling the viewpoint rotation, cursor pointing with the move, and lateral (forward/backwards/strafing) with the Nav controller?
I'm sure it'd take a bit of getting used to, and you may get a bit of neck ache, but so long as the head-tracking gives you enough effective "dead-zone" i'd be curious to see how well it would work.... it'd certainly "feel" pretty immersive and natural for the player ;-)
Yeah, that too. Even more so in 3D I think by the way. But this morning I was thinking about what sort of additional things you could do that now no-one is even thinking about, like maybe a granade in the shape of a discus or boomerang, that you can throw, arc and float over or around terrain/corners/hills etc. There have to be a tonne of new things possible here ..
Someone on GAF made and posted this video, which is fairly relevant to at least part of this discussion as it gives you a very close up look of the relation between the controller movement and the reticule movement in Socom (and thus also the lag for this game).
Well, everybody has network lag that fluctuates. Sometimes you win, sometimes you lose in terms of latency. The only one you can really avoid is display lag, by carefully selecting your tv. The control lag inherent in the game is a wash, because everyone has it, and it's unavoidable. Now, if there's a difference in lag between Move and the standard DS3, and you can choose which one you'll use, then that's another place you can control your lag. But even with a slight difference in lag, there may be other advantages in the Move control method that would make you better in some situations, and disadvantages that would make you worse off in others. It would be interesting if you could see stats in Killzone3 or Socom, that would show you how Move players compared to DS3 players in terms of online success (K
or other metrics).I believe Move lags about 6ms more than DS3. However, the actual in-game input lag depends on how the game is written. It is possible to have no lag difference between the 2 of them. For a good comparison, we may need to see them side by side.
IMHO, the problem with Move when compared to DS3 is stability. DS3 analog stick is mounted. When left alone, it will remain stable. Move is a floating device. Need to keep you hand stable for fine aiming. The game may also smooth your shaking hand motion. So you may lose a little accuracy.
The advantage is it should be more versatile since it allows free form movement.
I have seen a few posters saying they have gotten used to (and prefer) Wii style controls for FPS. The Socom devs mentioned that during their tests, some prefer DS3, others prefer Move.
In the future, we may see a Move gamer given extra roles in a MP FPS game. It would be interesting to see how a pure DS3 team fare against a mixed Move and DS3 team.
IMHO, the problem with Move when compared to DS3 is stability. DS3 analog stick is mounted. When left alone, it will remain stable. Move is a floating device. Need to keep you hand stable for fine aiming. The game may also smooth your shaking hand motion. So you may lose a little accuracy.
The advantage is it should be more versatile since it allows free form movement.
I have seen a few posters saying they have gotten used to (and prefer) Wii style controls for FPS. The Socom devs mentioned that during their tests, some prefer DS3, others prefer Move.
In the future, we may see a Move gamer given extra roles in a MP FPS game. It would be interesting to see how a pure DS3 team fare against a mixed Move and DS3 team.
But would it even make sense to let DS3 and Move users to play together given any advantages/disadvantages either way, at least for ranked matches? Keep things fair for everyone.
Hopefully DigitalFoundry will do a nice comparative piece on the differences in lag and other things between the DS3 and Move once the actual games are released.
Hopefully DigitalFoundry will do a nice comparative piece on the differences in lag and other things between the DS3 and Move once the actual games are released.
Good question, and I believe that's one of the main reasons most devs don't support moues and keyboard on PS3 and Microsoft doesn't allow it at all. For a huge game like Halo, you have enough people to fracture the userbase. For smaller titles, you really can't afford to split your community. I'm not sure how many people play the Socom games online, but I don't think it's a really big number relative to other titles.
What are you basing this on?
Doesn't move include PSEye (possibly running @ 60hz) + image processing. :/
Yeah, lag for buttons and the internal gyros should be different than the pointer, which is based on Eye tracking those glowing orbs.
A recent Sony comment that Move was 22ms latency versus 16ms for the DS3, or whatever it was. I don't think that figure includes processing though as Sony have said you can introduce more lag adding smoothing, which I guess is an interpolation of movement between several frames.
There should be very little additional lag for the Move in theory. In practice, a little more lag could be added to smooth out the cursor movement, as otherwise you do indeed see every slight tremble of your hand and the environment reflected in the cursor, which can be annoying. But it's difficult to know whether or not you're seeing a difference here between Move and what it would look like when using a DS3, taking into account display lag, render pipeline, etc. You don't usually see your finger's movement into relation with an on-screen cursor in these cases, as your mind translates most of that accordingly (though it does matter of course - it will still feel more direct if there is less lag, even unconsciously).
However, you have to take into account that after a little practice you should be able to take a lot more shots on targets in your field of view with a lot more accuracy as with the DualShock 3, even when the latter keeps the amount of assist it already has.
Should the lag be a real factor and caused by making sure the reticule doesn't jitter, I think developers should perhaps even consider not compensating, and not shoting a reticule instead. You don't actually need one when you aim at your target directly.
Actually, the pointer isn't primarily driven by the glowing orb. The orb is primarily necessary for more accurate Z detection.
The glowing orb isn't used at all for pointer controls. There is no way for the Eye to tell which way a near perfect sphere is pointing, since it's relatively the same on all sides
I'd assume the pointer controls are based off of a combination of the gyros, accelerometers, and magnetometers. This explains the lag, as it must use a combination of all of this technology to accurately determine which way the device is pointing.
Well it is lag, as all lag is there for a reason. My guess is it's smoothing out motion based on a few frames, in essence drawing a path between several control points taken as samples of the Move direction per frame, so an occasional judder doesn't send the pointer flying. Contrast this with Wii's calibration that showed the cursor jittering around the screen without any smoothing, and you can see why it's wanted, but also the lag that's introduced. I can't think of a way to de-jitter the cursor without lag and without reducing sensitivity; you could ignore all slight variations from the current position, kind of snap-to the current point, but that'd feel odd.
My guess is it's smoothing out motion based on a few frames, in essence drawing a path between several control points taken as samples of the Move direction per frame, so an occasional judder doesn't send the pointer flying. Contrast this with Wii's calibration that showed the cursor jittering around the screen without any smoothing, and you can see why it's wanted, but also the lag that's introduced. I can't think of a way to de-jitter the cursor without lag and without reducing sensitivity; you could ignore all slight variations from the current position, kind of snap-to the current point, but that'd feel odd.If you read the articles, they would say something like this regarding the 22ms number: "This game highlighted the slight lag between actions and on-screen reaction, Sony says this is 22 ms, and it is a little noticeable."
Dr. Marks also has an interview with Kotaku regarding Move's stats: http://kotaku.com/5502008/the-possi...f-playstations-move-according-to-the-inventor
So...
Camera data comes in @ 60fps, sensor comes in at much higher rate (probably similar to SIXAXIS). The camera input should be the bottleneck (16.7ms). For image processing, 1 SPU chewed through 1280x720 MLAA in 20ms, 640x480 is 3 times smaller. Assuming the light ball image processing is as intensive as MLAA, 1 SPU would take 20/3 = 6.7ms. So camera input should take no worse than 16.7 + 6.7 = 23.4ms. In reality, I suspect the light ball recognition should be much faster than MLAA (e.g., There's no need to blend colors).
So I think 22ms already accounts for camera input + internal sensors.
It's like what the article says, 22ms is the "lag between actions and on-screen reaction", assuming zero display lag and other lags.
However, the game will add its own interpretation on top (e.g., smoothing, gesture recognition). Smoothing (raw) input should be negligible. Gesture recognition will be hefty (very hefty) because we need to wait for the user to complete the gesture.
EDIT: We can further optimize the performance by ignoring/interpolating camera input when absolute positioning is not (absolutely) required.
In addition, these values may be updated on the fly while the game simply reads/samples them as and when needed, at its own rate.
Yeah, I'm thinking about how that would work and remember how the wiimote works. You need to measure how far the remote/wand is off x,y axis relative to the tv to know where the controller is pointing. Wii uses the IR emitter bar for relative positioning.