Overclock.net banner
1 - 13 of 13 Posts

Xcrypt

· Registered
Joined
·
6 Posts
Discussion starter · #1 ·
So ps/2 keyboards have less input lag because they use an 'interrupt system'

I know that ps/2 mice use less cpu cycles (although supposedly nowadays barely noticeable), but is there also any difference in input lag compared to USB mice (even if it's only a couple ms)?
 
USB 3.0 doesn't rely on polling either, so it's also better than USB 2.0. It has less CPU overhead as a result, and inherently less input lag, even when USB 2.0 is running at 1000hz. It does however still use a driver, so PS/2 is still more direct to the CPU, and therefore theoretically a more responsive interface. HOWEVER this is never the case in reality. Because it actually has slower throughput, it's slower and less responsive than USB 3.0 and even 1000hz USB 2.0 in practice. It's more noticeable on certain keyboard keys that transmit extra data, or multiple key presses, and it's infinitely slower than even USB 1.1 for mice, which I think use 100hz reporting rates in W7 anyway. I'm probably not using the correct terminology but this is the gist of it. PS/2 is also much less capable of delivering power. Without a doubt, USB 3.0 is the best current interface for KB and Mice, while usb 2.0 is second best (2.0 can even be programmed for NKRO). Will be some time before manufacturers bother with 3.0 though.
 
No mouse can use usb 3.0 now, they work in it the same as in 2.0

But In reality usb 3.0 is much more laggier than usb 2.0

If developers make good drivers and controllers then its gonna be good.

125hz in usb 2.0 > 1000hz in 3.0
 
You're misinformed.

The 3.0 standard is fully backwards compatible so the only reason a 3.0 port would be laggier than a 2.0 port is if it's a 3.0 port on a third party controller - and there are many of those out there. Any Sandybridge mobo with 3.0 ports, or any Ivy mobo with more than four ports, or Haswell with more than six ports use third party. And those are best-case numbers on the best consumer chipsets. There isn't even one LGA 2011 chipset with native 3.0 support.

No mice "use" 3.0 because no mice are 3.0 - both mouse and port need to be 3.0.

Developers don't need to make "better drivers" since 3.0 drivers are packaged in Windows.

There is no "1000hz" in 3.0 - the standard abandons polling once and for all, apart from for backward compatibility.
 
Quote:
Originally Posted by dukeReinhardt View Post

You're misinformed.

The 3.0 standard is fully backwards compatible so the only reason a 3.0 port would be laggier than a 2.0 port is if it's a 3.0 port on a third party controller - and there are many of those out there. Any Sandybridge mobo with 3.0 ports, or any Ivy mobo with more than four ports, or Haswell with more than six ports use third party.
I was talking about how it is now on my and other people's chipsets that noticed the delay from it.

Quote:
only reason a 3.0 port would be laggier than a 2.0 port is if it's a 3.0 port on a third party controller
Quote:
And those are best-case numbers on the best consumer chipsets. There isn't even one LGA 2011 chipset with native 3.0 support.
So your saying that very few boards actually have non third party controller?
Quote:
Without a doubt, USB 3.0 is the best current interface for KB and Mice
Will be when kb and mice will use it. Right now theres lots of 3rd party controllers as even you said, so its worse than 2.0, because of the delay from 3rd party controller =D
Quote:
No mice "use" 3.0 because no mice are 3.0 - both mouse and port need to be 3.0.
Why did you say this I dont even know.
Quote:
Developers don't need to make "better drivers" since 3.0 drivers are packaged in Windows.
I actually needed driver for usb 3.0. Idk if windows has drivers for 3rd party controllers. And my drivers caused a little bit of delay.
Quote:
There is no "1000hz" in 3.0 - the standard abandons polling once and for all, apart from for backward compatibility.
I wasnt comparing non existent usb 3.0 mouse with usb 2.0 mouse.
 
I wasn't trying to be confrontational... your wording suggested you didn't fully understand the USB 3.0 spec so I set out to correct what seem like misconceptions. According to what you said just now, it is laggy on your chipset (technically it should be "motherboard" since your chipset has no 3.0 standard), and you need drivers. This is because z68 has no native 3.0 support. My native 3.0 ports don't need extraneous drivers as I said - they're included in Windows. The reason I said few mobos have native 3.0 is because I was trying to demonstrate why you and others might think "USB 3.0" is the cause of lag, when it has no inherent design flaws that make it laggier than 2.0. 3.0 is still the best standard, and the fact that in practice not everyone has native 3.0 ports doesn't undermine its strengths. I know you weren't comparing non-existent things but I was talking about interface comparisons, and that's also why I added "Will be some time before manufacturers bother with 3.0 though." at the end of my post. When you replied in a manner which seemed to bemisinformed, and disparaging of the 3.0 spec, I just wanted to correct you before someone got the wrong idea.

If you're saying that in practice A. no such 3.0 mice exist and that B. not many people would be able to take advantage of them then I would fully agree with you. I just hope that 3.0 peripherals will be more common in future. In fact two of the next generation of consoles use 3.0 peripherals, which I'm very pleased to see. It seems Sony and MS felt the benefits of 3.0 would justify the cost in implementing the more expensive standard.
 
1 - 13 of 13 Posts