This site is a testing version, but all data is shared with the live forum.


Raised This Month: $ Target: $400
 0% 

66.7 FPS vs 144 Hz


Post New Thread Reply   
 
Thread Tools Display Modes
Author Message
Potato Uno
Veteran Member
Join Date: Jan 2014
Location: Atlanta, Georgia
Old 05-26-2015 , 12:13   66.7 FPS vs 144 Hz
Reply With Quote #1

[Slightly off topic but still relevant to servers]

So I was looking at some 144 Hz monitors and was thinking about gaming at 144 frames per second. But as a server owner, I remembered that TF2 servers (mvm ones to be exact) only run at 66.7 frames per second. So if the amount of data thrown from the server to the client is 66 frames per second, if someone is using a 144 Hz monitor in TF2, where does the other 78 frames per second come from? Do people with 144 FPS have any advantage over those playing at 60 FPS, despite getting the same amount of data from the server?

Does the game just duplicate each graphical frame twice before the next server tick kicks in?

Thanks!
Potato Uno is offline
miyoko9299
Member
Join Date: Apr 2015
Old 05-26-2015 , 12:18   Re: 66.7 FPS vs 144 Hz
Reply With Quote #2

It offer a big advantage, unless your client fps is also locked to 66.7 using 144 would not really benefit or harm you in this case but if you as client can run 150+ frames easily then you will will notice a much smoother game play compared with an older 60hz.

Besides comparing fps to hz is not really a good practice like this quite i have saved;

- fps = Frames Per Second. Paint the specified number of whole frames in one second.
- Hz = Hertz, which means a signal is crossing the "zero point" that many times per second. The old term that was replaced by Hertz was Cycles Per Second.

So, in a way fps = Hz, but that's like saying an orange is equal to a tangerine.
This also a very un-detailed way to explain this subject, pardon my english.
__________________
I'm a user of google translate, i do not speak English fluently.

Last edited by miyoko9299; 05-26-2015 at 12:19.
miyoko9299 is offline
Impact123
Veteran Member
Join Date: Oct 2011
Location: Germany
Old 05-26-2015 , 12:46   Re: 66.7 FPS vs 144 Hz
Reply With Quote #3

Your client's fps aren't directly bound to the tickrate of the server. Your game also won't duplicate frames when it's not receiving data, otherwise it would effectively freeze during a timeout. Think of server fps as world / player updates.
__________________

Last edited by Impact123; 05-26-2015 at 13:20.
Impact123 is offline
Mitchell
~lick~
Join Date: Mar 2010
Old 05-26-2015 , 13:40   Re: 66.7 FPS vs 144 Hz
Reply With Quote #4

Tickrate - Server-side
FPS - Client-side
Monitor Hz - Physical-side.

(I say physical because you need at least a dual-link DVI cable for over 60hz)
Mitchell is offline
Potato Uno
Veteran Member
Join Date: Jan 2014
Location: Atlanta, Georgia
Old 05-26-2015 , 13:47   Re: 66.7 FPS vs 144 Hz
Reply With Quote #5

Consider the following table:

Client ---------------- Server
Frame 1 (of 144) --- Frame 1 (of 66)
Frame 2 (of 144) ---
Frame 3 (of 144) --- Frame 2 (of 66)
Frame 4 (of 144) ---
Frame 5 (of 144) --- Frame 3 (of 66)

In the above table, frame 1, 3, and 5 of the client have had the data updated from server frame 1, 2, 3.

The server updates all the entity positions (counting players, projectiles, etc), statistics, etc once per server frame. The client updates its graphical data (i.e. the frames on your monitor) twice during that interval. What will the client see in frames 2 and 4, when there has been no data sent from the server since frame 1 and 3 (or 3 and 5)?

The mouse lag will be less, I can clearly see that, but how does it "give you an advantage" (so to speak) in playing a game if you don't get server data equaling your frame rate? In other words, what does the game insert in frame 2 and 4 with no server data corresponding to it? (It still has the same data that it used to draw 1 and 3, respectively.)
Potato Uno is offline
shavit
AlliedModders Donor
Join Date: Dec 2011
Location: Israel
Old 05-26-2015 , 15:23   Re: 66.7 FPS vs 144 Hz
Reply With Quote #6

Quote:
Originally Posted by Mitchell View Post
Tickrate - Server-side
FPS - Client-side
Monitor Hz - Physical-side.

(I say physical because you need at least a dual-link DVI cable for over 60hz)
hm
I don't use a dual-link cable and got this after messing with settings for a while



Got this at 1920x1080
__________________
retired
shavit is offline
miyoko9299
Member
Join Date: Apr 2015
Old 05-26-2015 , 16:08   Re: 66.7 FPS vs 144 Hz
Reply With Quote #7

Quote:
Originally Posted by shavit View Post
hm
I don't use a dual-link cable and got this after messing with settings for a while



Got this at 1920x1080
Standard VGA (DSUB) support up to 70 usually that is odd hz you have, perhaps use hdmi?
Potato 60/120/144 will no make any matter on the server, is only client side what is effected and depend on this individuals fps.
__________________
I'm a user of google translate, i do not speak English fluently.

Last edited by miyoko9299; 05-26-2015 at 16:12.
miyoko9299 is offline
Impact123
Veteran Member
Join Date: Oct 2011
Location: Germany
Old 05-26-2015 , 17:17   Re: 66.7 FPS vs 144 Hz
Reply With Quote #8

You can easily test this if you play l4d / l4d2, afaik their servers are running with 30 tick. I think the answers you're looking for can be found on this page and its links.
__________________

Last edited by Impact123; 05-26-2015 at 17:22.
Impact123 is offline
necavi
Veteran Member
Join Date: Sep 2010
Old 05-27-2015 , 01:38   Re: 66.7 FPS vs 144 Hz
Reply With Quote #9

Quote:
Originally Posted by Potato Uno View Post
Consider the following table:

Client ---------------- Server
Frame 1 (of 144) --- Frame 1 (of 66)
Frame 2 (of 144) ---
Frame 3 (of 144) --- Frame 2 (of 66)
Frame 4 (of 144) ---
Frame 5 (of 144) --- Frame 3 (of 66)

In the above table, frame 1, 3, and 5 of the client have had the data updated from server frame 1, 2, 3.

The server updates all the entity positions (counting players, projectiles, etc), statistics, etc once per server frame. The client updates its graphical data (i.e. the frames on your monitor) twice during that interval. What will the client see in frames 2 and 4, when there has been no data sent from the server since frame 1 and 3 (or 3 and 5)?

The mouse lag will be less, I can clearly see that, but how does it "give you an advantage" (so to speak) in playing a game if you don't get server data equaling your frame rate? In other words, what does the game insert in frame 2 and 4 with no server data corresponding to it? (It still has the same data that it used to draw 1 and 3, respectively.)

The thing you aren't realizing there is that the update you received from the server is delayed by the latency between you and the server. Your client interpolates the position based on the last update ANYWAY, this just means that it will interpolate more frequently, allowing slightly more granularity between server updates.
necavi is offline
Powerlord
AlliedModders Donor
Join Date: Jun 2008
Location: Seduce Me!
Old 05-27-2015 , 10:46   Re: 66.7 FPS vs 144 Hz
Reply With Quote #10

Traditionally, when you talk about Frame Rate, you're specifically talking about graphics.

If the frame rate (measured in frames per second (FPS)) is too low, things may look jerky or blurry. In gaming, you generally want 60FPS or higher. Movies are generally 24FPS although this may change to 48FPS soon.

As for the game server, it is running a timer internally that ticks a fixed number of times per second. Every time this timer ticks, the game server does work. Hence the name tickrate. For reasons unknown to me, these are sometimes referred to as game frames.

TF2 servers generally process 66 game frames per second. It sends partial game snapshots to clients based on client's cl_updaterate (capped on both ends by the server cvars sv_mixupdaterate and sv_maxupdaterate)... although it will never exceed the client's rate (capped at both ends by sv_minrate and sv_maxrate).

Full game snapshots are only sent when a client first joins (including after a map change), if the server detects issues with the client, or if the client manually requests one using cl_fullupdate (the record command does this internally).

The graphics system can have a higher framerate than the game's cl_updaterate because it still has things it can draw on screen to make the game look nice. Projectiles, water effects, taunt animations, etc...
__________________
Not currently working on SourceMod plugin development.

Last edited by Powerlord; 05-27-2015 at 10:47. Reason: process to match the correct tense.
Powerlord is offline
Reply



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 22:16.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Theme made by Freecode