[Resolved] Trying to Reach 1000+FPS

[hpk7331]

Server OS: Windows 7 ultimate 64bit
Processor: AMD phenom II 965 (Quad 3.4ghz)
Ram: 8gb

Already tryed:

HiResTimer
Fast Multimedia Timer
srcdsfpsboost

Dont get me to the mark im looking for.

ive only been able to reach 600-800 FPS.

Any idea's would help. Thanks ahead of time.

[Nisd]

You aint going to get near 1000 fps unless you do like this fellar.
http://forums.srcds.com/viewtopic/15263

[hpk7331]

yea i noticed that but i couldnt find the programs to do so.

if you can show me where to start, i can be off, well thats if your
up for it

Edit: One other thing, i was capable of reaching 800-950 without modding of any sort.

[Nisd]

Yes, normally you can reach around 900 fps.

But the thread I liked to is only a prof of concept, he have not released the source code, nor is it stable.

[Beaverbeliever]

Your eye can't even tell the difference between 200 fps and 100000 fps, so I don't see any point.

[Monk]

He means servers 'FPS', not clientside.

[hpk7331]

Thanks Monk. Yes server side, just trying to up the server FPS some. if someone can get me start that would be great. if not, its cool.

[BloodAngel]

The only realistic way of doing it, is to use a Linux distro and a lib to boost over 1000fps, although it doesent improve anything (IMO), if you are wanting to do it then read this: http://wiki.fragaholics.de/index.php/BEpingboost.c


But yeah, your not going to break 1000fps on windows any time soon,

Christy

[helio]

just remember its not how high the fps is, its how stable the server is that counts

[Terrorkarotte]

Even more important: Good peering/connection not overloaded box and decent hardware.

[hpk7331]

Thanks guys, looks like Linux is the only way to go.
now for the windows not having 1000fps stable, meh i got close enough,
900 to 1000 stable. there is ways to get it up there, but im going to go linux thanks guys.

[Monk]

(09-14-2010, 09:07 PM)BloodAngel Wrote:  The only realistic way of doing it, is to use a Linux distro and a lib to boost over 1000fps, although it doesent improve anything (IMO), if you are wanting to do it then read this: http://wiki.fragaholics.de/index.php/BEpingboost.c


But yeah, your not going to break 1000fps on windows any time soon,

Christy

Not to beat a dead horse, but there is still debate to what FPS still does inside the engine. There was a thread on hlds_linux (valve mailing list) about it and a large amount of people agreed that stable/high fps doesn't do anything except suck up large amount of CPU. (I agreed with them, obviously, because of the lack of empirical evidence)

Of course, your mileage may vary, just keep an open mind about everything.

[hpk7331]

(09-15-2010, 05:14 PM)Monk Wrote:  

(09-14-2010, 09:07 PM)BloodAngel Wrote:  The only realistic way of doing it, is to use a Linux distro and a lib to boost over 1000fps, although it doesent improve anything (IMO), if you are wanting to do it then read this: http://wiki.fragaholics.de/index.php/BEpingboost.c


But yeah, your not going to break 1000fps on windows any time soon,

Christy

Not to beat a dead horse, but there is still debate to what FPS still does inside the engine. There was a thread on hlds_linux (valve mailing list) about it and a large amount of people agreed that stable/high fps doesn't do anything except suck up large amount of CPU. (I agreed with them, obviously, because of the lack of empirical evidence)

Of course, your mileage may vary, just keep an open mind about everything.

I have proven that a higher FPS can make a difference than a default 300 fps from nuclearfallout. im hosting a server publicly and people have found that they register better with a lil higher ping at my server than one of nuclearfallouts servers because i have my fps doubled to 600 fps. makes clients register better server side, i keep getting compliments over and over about how they register better on my server.

[Monk]

I'm sorry, but that doesn't qualify for empirical evidence. Just because people say so doesn't mean anything. You have the follow setup:

- 2 different physical locations
- 2 different network return paths (asymmetrical routing)
- 2 different latency results
- 2 different pieces of hardware
- 2 different OS's (linux vs windows (i think nfo uses linux?))

So it's very possible at least 2 out of those 8 is the reason for things behaving differently than on "x" server compared to "y" server.

I am not flaming you, I am just trying to get you to look at things objectively, instead of being on this FPS-is-good-for-everything kick since there's

a.) No documentation on what FPS does
b.) No documentation on what else tickrate does
c.) No Source code for VALVes Lag compensation / interpolation / extrapolation mechanics

[helio]

(09-18-2010, 12:33 PM)Monk Wrote:  a.) No documentation on what FPS does
b.) No documentation on what else tickrate does
c.) No Source code for VALVes Lag compensation / interpolation / extrapolation mechanics

Further proof that theses things like "FPS" and "Tickrate" are a conspiracy among GPS to charge more