capping your fps.........


(Dawg) #21
  1. I hadn’t really considered the benefits of a constant FPS generating smoother gameplay. That’s a good reason to try it out.

  2. Jumping seems to be the most measurable benefit from FPS-capping. See the Coriolis article linked in previous post.

  3. Ping will not be affected by pretty much anything on your computer other than your connection to the internet. Ping is a measurement of how long it takes for data (stored and transferred in packets) to travel from your machine to the server and vice versa. Therefor, the FPS setting should have no effect. Even if one assumes that a higher FPS requires the transfer of more data, that just means that more packets will be sent. Their transmission speed is unaffected.

  4. Makes sense. You are reducing the processing the card needs to perform. Yet another reason to cap, especially on the faster cards.

  5. What is the cap you selected?

I have to think that there is a point where additional frames are of no benefit. I am thinking along the lines of how movies use 24 frames/second (I think its 24 - something around there) because anything more the human eye cannot perceive a difference. Has anyone ever attempted an evaluation of this? It would make for an interesting biology/physiology study for some undergrad. If nothing else, the principle of diminishing returns must apply at some point.

Dawg


([=O=]SMYLER) #22

I did many mathmatical computations, consulted with several pagan gods and sacrificed a chicken in a vodoo ritual and thought that 90 was a nice number. So thats the one I ended up putting in. I put zero thought into my choice andI suppose I was running about on our server last night often taking out 4/5/6 enemy bastards to my one death (this included me using lazy ass airstrikes but I did seem to have the upper hand in many shoot outs).

I hope this thread has helped no one at all apart from my good self and I waste your asses every time I see you, unless your on my team, then I will leave it to me clan member [=O=]Zaton who is known as “TK machine” :angry: . He is also a right cunt but that is another story :banana: .


(SCDS_reyalP) #23

That is not actually quite true. The way ping is measured in quake3 games, higher frame rates will give you a lower displayed ping. Even in the game browser. I have seen this many times with multiple systems sharing the same connection.

My guess is that the time for a frame to complete ends up included in the displayed ping. That may also explain the supposed cl_maxpackets / com_maxfps relationship. Remember, ping is the round trip time, which includes whatever processing is required at each end.


(Hugo Phurst) #24

I think I’m gonna try this out, but what I need to know is what is the console command to show the fps counter please.

Hugo Phurst


(SCDS_reyalP) #25

/cg_drawfps 1


(Hugo Phurst) #26

Cool, v43 v43.

Hugo Phurst


(Dawg) #27

[quote=“SCDS_reyalP”]

That is not actually quite true. The way ping is measured in quake3 games, higher frame rates will give you a lower displayed ping. Even in the game browser. I have seen this many times with multiple systems sharing the same connection.

My guess is that the time for a frame to complete ends up included in the displayed ping. That may also explain the supposed cl_maxpackets / com_maxfps relationship. Remember, ping is the round trip time, which includes whatever processing is required at each end.[/quote]

Good point. I was thinking about that when I made the previous post which I why I hedged a little and said ‘pretty much’. I suppose that there needs to be a differentiation between ping (being packet travel time), and ping (being Q3A ping measurement). For sake of simplicity, we should assume we are talking about Q3A ping.

Then the question becomes, how is Q3A ping measured?

Also, I can see how a variable named max_packets could be used to speed processing (by varying depending upon the reliability of the connection), but I would imagine that a higher FPS cap would actually slow the connection since there is more overhead processing being done in the client-server connection before the data is forwarded to the vid card (which is a major factor in getting a good FPS).

I dunno… Without more information or insight into the actual client/server protocols and what not, all we are really doing is guessing. Ahh… but there’ no harm in that… :slight_smile:

Dawg


([=O=]SMYLER) #28

Sacrifice the chicken and stick the number “90” into the maxfps bit. We all know of the connection between voodoo and 90 and cottage cheese (of course (as cottage cheese is the one to please)).