FPS in CS (take 2, no flaming)

Discussion in 'PC Gaming' started by Zedric, Apr 23, 2002.

  1. Zedric

    Zedric NTFS Guru Folding Team

    Messages:
    4,006
    Location:
    Sweden
    Ok since people thought it funny to flame my thread to pieces I will have to try again... NO FLAMING THIS TIME! ok?




    Old thread is: http://www.xp-erience.org/forum/showthread.php?s=&threadid=9223&perpage=15&pagenumber=1








    Now then...

    I've seen you all object to the high FPS I was talking about. I assure you I get 400-600 in CS (1024*768*32). I will ask my brother about the uncapping and the fps measurement command (when you execute the command your character spins one lap and the console displays the FPS and the time it took to spin). The 2500 screenshot sounds faked, I know, but it is authentic according to my brother. he saw it in action. There may however (and IS probably) something wrong with his game since 2500 is just ridiculus.

    The point with getting 600+ in CS is not to get better gameplay in CS, anything beyond 60 is just wasted. The point is that it's a measurement of the overall graphics performance of the computer. If I get "low" FPS in CS (say 100) it's probably that it will be much lower in newer games. If I get high FPS (say 600+) it's probable that newer games will run smoother.



    Please don't flame the thread again, I don't want to rewrite it another time!
     
  2. Qumahlin

    Qumahlin Moderator

    Messages:
    2,006
    OK lets be serious, I haven't really seen CS do 600fps, obviously V-sync would have to be disable or else you'd have to have a 600mhz refresh rate (impossible) and on the system you described it is definitely not right, I have a athlon XP 1900+, 1 gig DDR, and a Geforce 4 TI 4400 OC'ed to beyond TI 4600 lvls and I guarantee I couldn't even score that.

    You have to remember a prand new p4 2.4ghz, with a geforce 4 TI 4600 is only capable of about 350FPS in quake 3....now try explaining getting DOUBLE that on a system that pales by comparison..and the main reason that quake3 doesn't get beyond the 350fps on such a system is due to the rest of the components of the computer such as mobo and such don't have the bandwidth to acheive more...so this isn't a flame but ask yourself doesn't it seem odd that on your system you were scoring double that of a top of the line system?

    But also using CS as any type of measure of computer performance is also flawed as CS really isn't advanced at all and is merely a mod. the FPS rating alone could be wrong across the board

    Just go download yourself a copy of 3dmark and you'll have a good idea of your computers performance for current and future games.

    I'm sorry if any of this seemed like a flame, but i'm just trying to use the quake 3 and such examples to put it into perspective.
     
  3. r3n3gad3

    r3n3gad3 Guest

    Yeah i agree totaly, i mean, although as you have said previously... the human eye can not see more than 60fps - so anything like 2500 is kinda a waste of resources dont ya think? :)

    On another note (baring in mind Zedric is referring to a screenshot)... didnt anyone think about the screenie being edited? i mean, its not exactly hard to change 250 to 2500 is it? (i'm only using 250 as an example...lol, wouldnt it be lovely 2 get 250fps all the time :p) hehe
     
  4. dijital

    dijital Guest

    this is not a flame. its just not possible to get frame rates that high with current hardware. once again, the human eye couldnt distinguish the difference even if it were possible.
     
  5. dejav00

    dejav00 Guest

    ...

    What flame?

    lol
     
  6. r3n3gad3

    r3n3gad3 Guest

    yup - n e 1 know the highest fps ever recorded?
     
  7. Zedric

    Zedric NTFS Guru Folding Team

    Messages:
    4,006
    Location:
    Sweden
    Qumahlin:
    Compared to the flamers you are cool as ice ;)
    However you talk about Q3. CS and Q3 are completel different when it comes to the kind of performance it will require from a computer. CS is based on HalfLife and that engine is getting old. I'm pretty sure there aren't any curved surfaces in it. Of course i cannot guarantee the accuracy of the built in CS FPS meter (obviously) but 400 is not that way of really.

    400*1024*768 = 314572800
    Considering the GF3Ti200 runs at 175 MHz this means roughtly 2 pixels per clock. If I remeber correctly (I may not) this is something the card can do.

    I'll ask my brother about those commands and post ar 3DMark score when I have the time.
     
  8. oDin

    oDin xp m0nk3y

    Messages:
    368
    Location:
    vancouver, bc
    in a half life based game...i belive 99

    oDin.
     
  9. wyrlwyn

    wyrlwyn Guest

    the number you are reading is the TRIS COUNT!!!! the number of polygons drawn in a second are not the same thing as the number of frames a second!!!!! the fps can not exede 99 in half life. the number of polygons in a view depend on the map, how it was made, and how it was compiled, 2000 or even 3000 is not uncommon, but the tris count if NOT YOUR FPS!
     
  10. oDin

    oDin xp m0nk3y

    Messages:
    368
    Location:
    vancouver, bc
    imagine 2000fps though....how nice would that be eh

    edit: anyone know how many fps the human eye can see??

    oDin.
     
  11. dejav00

    dejav00 Guest

    ...

    I think it has been decided that we can only tell the difference up to 60, after that...

    Note: We are lame, our eyes are only capable of taking 60 samples a second, my mouse is taking 400 right now...lol
     
  12. wyrlwyn

    wyrlwyn Guest

    well i'm glad we finally settled this dispute....
     
  13. Zedric

    Zedric NTFS Guru Folding Team

    Messages:
    4,006
    Location:
    Sweden
    No wyrlwyn it's not tris count. That would be higher if you look out over an open area of the map and lower if you looked into the corner of a wall. This is not the case as I remember.

    Ok as you all have stated 100 FPS is impossible. This is true if you look at the netstat. If you, on the other hand, take up the console and type "timerefresh" (or maybe "time_refresh", not sure) the character spins 360 degrees and the console displays the FPS during that time. The 400 I was talking about was measured in de_dust at the starting area (down the stairs). Again I cannot guarantee anything, but the console says 400 FPS.
    When I get home I'll get you a screenshot.

    So run the timerefresh and report back. Who is wrong? I'd like to know. My brother is getting anoying ;)
     
  14. dijital

    dijital Guest

    @oDin,

    i believe its either 60 or 72. i'm not sure though.
     
  15. wyrlwyn

    wyrlwyn Guest

    it depends on how the level designer made it and what he used to compile... generally, in a hallway, a player would get about 2000, in crossfire, looking out over the courtyard, youd get higher then taht
     
  16. Zedric

    Zedric NTFS Guru Folding Team

    Messages:
    4,006
    Location:
    Sweden
    Ok here's the screenshot. It's AUTHENTIC, I made it myself. Besides, who would acctually take the time to fake that many numbers?

    Oh and the 3DMark 2001 score. I ran a benchmark and it came out 5989.
     
  17. dijital

    dijital Guest

    but look at your netgraph in the bottom right corner? that is more realistic.
     
  18. Zedric

    Zedric NTFS Guru Folding Team

    Messages:
    4,006
    Location:
    Sweden
    Maybe so, but it's also capped to 100.
     
  19. dijital

    dijital Guest

    set the cap higher then lets see what you get. it'll only go as high as your monitor can support or your gfx card. one of the two will be the limiting factor.
     
  20. goten133

    goten133 Guest

    wtf...

    wtf... u people talking about 200-300 fps u guys are on crack hl engine only supports up to 99...dont believe me? switch ur rez at 800x600 or even lower