Urban Terror Forums: Bugs and Suggestions list discussion - Urban Terror Forums

Jump to content

 Login | Register 
Advertisement
  • (15 Pages)
  • +
  • « First
  • 4
  • 5
  • 6
  • 7
  • 8
  • Last »
  • You cannot start a new topic
  • This topic is locked

Bugs and Suggestions list discussion Rate Topic: -----

Discussion here

#51 User is offline   mitsubishi Icon

  • Account: mitsubishi
  • Country:
  • Joined: 28-February 10
  • Posts: 13,481

Posted 04 June 2010 - 01:39 PM

Turn the personal vendetta mode off.

#52 User is offline   DeMS Icon

  •   QA member   
  • Account: dems
  • Country:
  • Joined: 28-February 10
  • Posts: 93

Posted 04 June 2010 - 03:46 PM

I think the 'raw' input mode should be also in the requested features :

http://forums.urbant..._1

#53 User is offline   slackin Icon

  • Account: slackin
  • Main tag: GlaD-
  • Country:
  • Joined: 28-February 10
  • Posts: 3,888

Posted 04 June 2010 - 10:08 PM

View Postmitsubishi, on 04 June 2010 - 01:39 PM, said:

Turn the personal vendetta mode off.


My vendetta button is stuck!
[img]http://i188.photobuc...n35/urt/sig.jpg[/img]
/*----------------------------------------------------------------------
All normal mapped(bump mapped) and specular mapped .|^|.
Just a few of many, and more to come.
----------------------------------------------------------------------*/

#54 User is offline   MaJ Icon

  •   verified user   
  • Account: maj
  • Country:
  • Joined: 07-May 09
  • Posts: 189

Posted 05 June 2010 - 02:15 AM

View PostDeMS, on 04 June 2010 - 03:46 PM, said:

I think the 'raw' input mode should be also in the requested features :

http://forums.urbant..._1&#entry265021


Depending on what revision 4.2 pulls from ioQ3, there will be a new mouse acceleration mode (coded by ttimo) that mirrors QuakeLive's.
http://bugzilla.iccu...793&action=diff (scroll down for explanation in the code)

Raw input from the Windows API isn't all it's cracked up to be.
Less latency? Maybe....probably 1 or 2 milliseconds.
Better acceleration? That's a personal opinion. Usually you can tweak your OS's settings to feel right in game.


#55 User is offline   DeMS Icon

  •   QA member   
  • Account: dems
  • Country:
  • Joined: 28-February 10
  • Posts: 93

Posted 05 June 2010 - 03:13 AM

View PostMaJ, on 05 June 2010 - 02:15 AM, said:

Depending on what revision 4.2 pulls from ioQ3, there will be a new mouse acceleration mode (coded by ttimo) that mirrors QuakeLive's.
http://bugzilla.iccu...793&action=diff (scroll down for explanation in the code)


Interesting stuff, thanks!

View PostMaJ, on 05 June 2010 - 02:15 AM, said:

Raw input from the Windows API isn't all it's cracked up to be.
Less latency? Maybe....probably 1 or 2 milliseconds.
Better acceleration? That's a personal opinion. Usually you can tweak your OS's settings to feel right in game.


It's not about less latency or better handling, but about more consistency.

If you use mouse 1, you have that movement 'lag' that makes any high frequency (over 125Hz) useless (you can try with m_filter 0 and you'll see mouse movement being jerkier), even lower frequencies do feel the same.

If you use mouse -1, you don't have that movement 'lag' on your pointer, but you're limited in the ratio of your screen resoultion/mouse DPI (higher DPI + lower resolution = hitting the edge of the screen, thus getting lots of negative acceleration), besides making mouse movement completely FPS-dependant.

With 'raw' mouse movement, you wouldn't need to choose either high DPI/low resolution (relative, you can have negative accel at 1600x1200 if you have enough DPI), it brings smooth movements and is consistent in almost whatever FPS scenario.

Not to tell that it could possibly also help removing negative acceleration on sub-unity sensitivity values (try using your mouse with sensitivity 0.02 and cranked up DPI, you will feel negative acceleration) which is inherent to Q3 engine afaik.

Disclaimer : those are the feelings I noticed when testing those settings with several gaming mice, I'm not saying they are technically true and proven.

bullet_loaderAdvertisement

#56 User is offline   MaJ Icon

  •   verified user   
  • Account: maj
  • Country:
  • Joined: 07-May 09
  • Posts: 189

Posted 05 June 2010 - 04:12 AM

View PostDeMS, on 05 June 2010 - 03:13 AM, said:

If you use mouse -1, you don't have that movement 'lag' on your pointer, but you're limited in the ratio of your screen resoultion/mouse DPI (higher DPI + lower resolution = hitting the edge of the screen, thus getting lots of negative acceleration), besides making mouse movement completely FPS-dependant.


I forgot about the clipping issue. I don't observe that issue at 125fps, 1920x1080, and a lower dpi/os sensitivity.

I understand that my conditions aren't always the norm and others might have some trouble with both directinput and position polling that could be fixed with raw.

The only real complaint I have is that it is a rather bit trickier to implement such a feature in linux based operating systems. There is no standard API like Windows has for raw mouse/keyboard capture. Several Xorg modules have such functionality but they get broken every other revision on their SVN. You can flip a coin as to which one is installed and/or working. Maybe SDL has already taken care of this trouble....I'd have to check.

.......although that may not be a good reason to leave a helpful feature out of Windows builds.


#57 User is offline   DeMS Icon

  •   QA member   
  • Account: dems
  • Country:
  • Joined: 28-February 10
  • Posts: 93

Posted 05 June 2010 - 04:48 AM

View PostMaJ, on 05 June 2010 - 04:12 AM, said:

I forgot about the clipping issue. I don't observe that issue at 125fps, 1920x1080, and a lower dpi/os sensitivity.


That's a fairly high screen resolution, a good number of FPS and low DPI. If you have a mouse capable of doing it, try doing this at 5600DPI and lowering your screen size. Negative acceleration is the norm then, not the exception. The same behavior can be observed giving very low values of sensitivity and probably limiting your FPS to a low value.

One of the first things that surprised me when I started playing UrT was the number of people with high sensitivity values, only to find that they are able to play at those values basically because they're forcing negative acceleration without knowing it.

View PostMaJ, on 05 June 2010 - 04:12 AM, said:

I understand that my conditions aren't always the norm and others might have some trouble with both directinput and position polling that could be fixed with raw.

The only real complaint I have is that it is a rather bit trickier to implement such a feature in linux based operating systems. There is no standard API like Windows has for raw mouse/keyboard capture. Several Xorg modules have such functionality but they get broken every other revision on their SVN. You can flip a coin as to which one is installed and/or working. Maybe SDL has already taken care of this trouble....I'd have to check.

.......although that may not be a good reason to leave a helpful feature out of Windows builds.


Being multiplatform does give the developers some extra headaches, indeed.

SDL is evolving everyday, and it looked like in the mailing lists linked on Mitsubishi's post that raw input might be used with it on a newer release (not 1.2, but maybe 1.3 or later), so it looks ok for linux in the near future.

Mac is a very different thing alltogether which I don't want to get in since I've got no first hand experience, though what I've gathered is that it has permanent acceleration on the mouse.

As for Windows, the number of outdated/obsolete standards is just too much. Old DirectInput causes problems, position polling has it's limitations, and there's no sign from any dev to port the input part of the code into newer DI standards (á là CoD4) which ain't affected by those design flaws. Going by the raw mode, instead, does seem like the sensible answer to a long standing problem, it's easier to implement than other methods, it's almost guaranteed that it will work and it will help a vast percentage of UrT players (most of them are on Windows platform).

#58 User is offline   mitsubishi Icon

  • Account: mitsubishi
  • Country:
  • Joined: 28-February 10
  • Posts: 13,481

Posted 05 June 2010 - 09:22 AM

I have never noticed negative acceleration or negative acceleration only when supposedly hitting an edge of a screen. How I understand it is that the DPI may be high but, unless you use WM_INPUT or DirectInput [DirectInput inflicts overheads according to Microsoft] you're going to lose the high definition and scale it down to the Desktop resolution capability [or have delays]. [in windows]

This is how it's explicitly explained in Microsoft's guidance document: http://msdn.microsof.../ee418864(VS.85).aspx

What we basically have is either mousemove which has no overheads but scales down DPI or DirectInput [which keeps the DPI!] but it has overheads. [in windows]



See, that's why I say let's read the forum; this is repetition.

This post has been edited by mitsubishi: 05 June 2010 - 09:25 AM


#59 User is offline   YAY5637 Icon

  • Account: yay5637
  • Main tag: <PWC>
  • Country:
  • Joined: 28-February 10
  • Posts: 633

Posted 06 June 2010 - 12:55 PM

As a suggestion
What about auto-autojoin?
winning team joiners are sometimes very annoying on pubs...

#60 User is offline   mitsubishi Icon

  • Account: mitsubishi
  • Country:
  • Joined: 28-February 10
  • Posts: 13,481

Posted 06 June 2010 - 01:23 PM

Already exists.

  • (15 Pages)
  • +
  • « First
  • 4
  • 5
  • 6
  • 7
  • 8
  • Last »
  • You cannot start a new topic
  • This topic is locked

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users

Advertisement


Copyright © 1999-2024 Frozensand Games Limited  |  All rights reserved  |  Urban Terror™ and FrozenSand™ are trademarks of Frozensand Games Limited

Frozensand Games is a Limited company registered in England and Wales. Company Reg No: 10343942