ET commands.

What are they doing and what value should be the best ?

1) m_filter

2) cl_packetdub

3) r_swapinterval

4) pmove_fixed

(i dont know which one is forced in cb config)
Comments
14
4) pmove_fixed

lets you jump a tiny bit higher, good for trickjumping

This command isn't forced and can be 1 if you wanna have some help with trickjumping. Consider it's seen as 'lame' to use this in higher level trickjumping.
cl_maxpackets 40
Parent
im sure u know what packetdup is :D
m_filter = smooth mouse
cl_packetdup = how much packets you want to duplicate
r_swapinterval = dno
pmove_fixed = you jumped a bit higher with 1, but its disabled in cb configs now afaik, ++it had more spread if enabled :/
pmove make u jump higher, but u cant hit rly good with it
r_swapinterval 1 = like you wanna lose fps
r_swapInterval - controls vsync and defaults 0 which is vsync off. Set to 1 to reduce tearing problems, you may lose FPS as it will be capped to monitor refresh rate.
cl_packetdup - range is 1 to 5 and is 1 as default. if a packet gets lost then a 'backup' command may still be received. Set to 1 as default which is recommended if you have packet loss. If you have an excellent connection with very little or no packetloss at all then this could be set to 0 and cl_maxpackets possibly raised. However, as with all settings experiment to see which is best suited for your connection.
1) m_filter - Smoothens mouse, I think it gives neg. accel, raziel says its bugged..

2) cl_packetdup - Duplicates packets, put to 2

3) r_swapinterval - Caps your fps to your refreshrate, HYPE says its bugged..

4) pmove_fixed - some trickjump-thingie, leave to 0
r_swapinterval= when @1 = vsyncs *cough*if u wanna bypas min fps limit change it to 2/3/4/n - it will change you fps to "monitors hz / n" *cough*
need abort's comments!
What are they doing and what value should be the best ?

1) m_filter 0
will average your mouseinputs instead of taking the real values

2) cl_packetdub 1
amount of times a packet get resend if its lost, you could put this value higher but i guess it only makes you more hittable.

3) r_swapinterval 0
This feature will buffer rendered images untill the display is ready to show them. Putting the output in a buff will add a delay to the images you see. Unless you use 200hz I dont recommend using it. Its usually turned off inside the video drivers, overriding the ET value, but you can put it at 0 to be sure. If you want a more stable image, try using tripple buffering but it will also add some delay to the images you see on screen. By default opengl renders 2 images in advance, tripple buffering will render 3 frames ahead. This is done to prevent tearing aka '2 frames at the same time' aka 'annoying motherfucking line in middle of your screen which shows where the old and new frame meat". Abort uses triple buffering, this must be the reason why he owns.

wiki:
Triple buffering attempts to provide a speed improvement over double buffering. In real life applications, this often involves trying to abstract the graphics drawing operations from being synchronized with the monitor's refresh rate. Typically this involves frames being drawn at a rate lower than or higher than the monitor's frame rate (a variable frame rate) without the usual effects this would cause (namely flickering, shearing and tearing).

summary: swapinterval prevents tearing at the cost of fps and reactiontime.

4) pmove_fixed
afaik this cmd is outdated, it was used to fix the fps depended fysics ET inherited from Q3base.

(i dont know which one is forced in cb config)
so m_filter 1 can make neg accel ?
Parent
No, it cant. m_filter will take the last mouseinputs and calculate the average movement over 2-3 (?) frames. This is repeated every frame. M_filter is kinda like vertical sync, it creates a delay. Old mouseinputs are used. Unless you have played with this option for ages, I recommend to keep it off.

It might reduce neg accel by averaging out the mouseinputs. Neg accel in Q3 (not hw neg accel older mice had - mx500 8bit datapath @ 125hz) is caused by your mouse hitting the side of your screen. Averaging out the mouseinputs, could reduce the amount of times your xhair hits that virtual wall (aka neg accel)
Parent
pmove 1 !!
Back to top