Counter Strike: Source FPS Issue


Recommended Posts

On my new iMac I seem to have problems getting CS:Source above 60fps in mac OSX, I get ~120fps in Half-life 2: Episode Two and Half-life 2: Death Match so I know I can get these higher scores in CS:Source too at the same settings, yet it seems to be stubbornly locked at 60fps.

Settings:

* Recommended Setting

Resolution: 1920 x 1080*

Model Detail: High*

Texture Detail: Medium*

Shader Detail: High*

Water Detail: Reflect All*

Shadow Detail: High*

Colour Correction: Enabled*

AntiAliasing Mode: 2x MSAA*

Filtering Mode: Anisoptric 2x*

Wait for Vertical Sync: Disabled

Motion Blur: Disabled

Field of View: 75.00*

Multicore Rendering: Enabled*

Fps_max=300

System:

CPU: Intel Core i3 3.06Ghz

RAM: 4GB

HDD: 500GB

Sound: Onboard

Graphics: ATI Radeon HD 4670 (256mb)

Link to comment
https://www.neowin.net/forum/topic/1006022-counter-strike-source-fps-issue/
Share on other sites

  On 19/06/2011 at 13:30, Mr_Edwardo said:

Yep, this is Vsync. Must still be on somehow in the AMD drivers.

+1

Locked at 60 FPS means VSync is enabled somewhere.

I'm not trolling I've just never understood and even googled prior to posting; correct me if I am wrong: the human eye can only view 30fps? If that is correct then why do games shoot for/boast about (CoD and you're old engine) 60fps, I know my MacBook Pro runs games I play @ higher frame rates, but is there really a benefit? People say the game runs smoother, yes obviously - but is it even noticeable to the human eye if we're physically limited to 30fps?

  On 20/06/2011 at 18:28, Alladaskill17 said:

I'm not trolling I've just never understood and even googled prior to posting; correct me if I am wrong: the human eye can only view 30fps? If that is correct then why do games shoot for/boast about (CoD and you're old engine) 60fps, I know my MacBook Pro runs games I play @ higher frame rates, but is there really a benefit? People say the game runs smoother, yes obviously - but is it even noticeable to the human eye if we're physically limited to 30fps?

Yes, the eye most definitely can see a difference between 30fps and above. Here's a demonstration of that:

http://www.boallen.com/fps-compare.html

It's all fairly complex: http://en.wikipedia.org/wiki/Frame_rate#Visible_frame_rate

  On 20/06/2011 at 19:04, Mr_Edwardo said:

Yes, the eye most definitely can see a difference between 30fps and above. Here's a demonstration of that:

http://www.boallen.com/fps-compare.html

It's all fairly complex: http://en.wikipedia.org/wiki/Frame_rate#Visible_frame_rate

Thanks for the links, checking them out.

  On 20/06/2011 at 23:32, Alladaskill17 said:

and that is why I said correct me if I am wrong :)

No worries. ;)

I think we're limited somewhere around the 90-100 frames per second. John Carmack said there's a very little difference between 60 and 120 fps, and you'll only see the difference if you know what it is (from a technical / programmer's point of view).

Anyway, getting the games to run at a standard 120 fps would be our limit, no doubt (there is no need to go above that for graphical reasons, only for calculations done in background). :)

  On 20/06/2011 at 23:46, KavazovAngel said:

No worries. ;)

I think we're limited somewhere around the 90-100 frames per second. John Carmack said there's a very little difference between 60 and 120 fps, and you'll only see the difference if you know what it is (from a technical / programmer's point of view).

Anyway, getting the games to run at a standard 120 fps would be our limit, no doubt (there is no need to go above that for graphical reasons, only for calculations done in background). :)

Perfect! That is basically what I wanted to know - what is our 'max' (or closest estimate) of viewable/recognizable FPS and what is just wasted processing. Thank you.

Depending on the scenario, the human eye can distinguish the contents of an individual frame displayed for 1/10000th of a second.

Of course, that doesn't matter when it comes to computers, since most monitors are physically limited to 60fps or so (Which is why it's pointless to render faster than that). You'd only ever want to disable v-sync for benchmarking.

  On 21/06/2011 at 05:38, The_Decryptor said:

Depending on the scenario, the human eye can distinguish the contents of an individual frame displayed for 1/10000th of a second.

Of course, that doesn't matter when it comes to computers, since most monitors are physically limited to 60fps or so (Which is why it's pointless to render faster than that). You'd only ever want to disable v-sync for benchmarking.

Enabling V-sync messes with the mouse input. Also, when playing online restricting your FPS to 60 on a 100 tick server would limit the packets you are sending/receiving to 60 (probably not noticeable but a small difference in hitreg). Having a higher FPS improves the smoothness and responsiveness of the game (even if your monitor can only display 60fps, all that will happen is a bit of tearing). In general it is better to keep v-sync off and ideally the FPS would be above 100 all the time for the smoothest possible experiance.

I hate all this discussion about the human eye can only see this...blah blah...those guys don't have a clue what they are talking about.

I think the tick rate has been set to a constant value as of recent CSS updates.

FPS rate is separate though.

If you cannot see the impact of 30 vs 60 FPS, you might need to get that checked - sounds like a lack of sleep or some disease or age or something.

Yeah, CSS, TF2, L4D and L4D2 are limited to 66 ticks a second (L4D/2 operate at 30 ticks a second by default)

The thing is though, that's how often the server thinks, not how often it talks to clients (You can render at 400fps, but it'll only send 20 updates a second to the server unless you play around with things)

To be honest, I always keep v-sync on for the sheer fact that most games' framerate fluctuates greatly depending on the amount of action on the screen. I -notice- the difference between, say, 65 fps and 90 fps. When the framerate changes so frequently, I feel like I'm speeding up and slowing down. Not to mention the tearing on the screen. V-sync + triple buffering or bust for me.

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • I think that the last game I bought on day-one was something like Ghost of Tsushima. I've come to realise that I don't have enough time (or sometimes inclination) to game as much as I used to, and I've got plenty of games to keep me occupied if I do decide to play. So I don't see the need to buy a game when it first releases - I wait until it goes on sale. If I'm really excited about a game and the price is under £50 then I would consider it, but anything over that price point gets ignored until the price comes down.
    • For the foreseeable that is your choice. I'm interested tom try one, my wife was very pleased because one she is anti-social and driver chat annoys her and two more seriously there is a long history of drivers abusing women, it's rare, but it happens and more than it should. Sometimes she needs to get a late taxi and she says it may make her feel safer.
    • 5800x3d was chopped because it was too good, why bother releasing a new mid range.
    • Was it bad, sure, but I got it on PC so it was nowhere near as bad as the PS4/XBO versions. Besides, I spent around 200hrs in there and have yet to play the expansion for it. All in all I think I got my monies worth. Maybe I was lucky and didn't run into anything game breaking like others. Either way, $60 is my limit for a game, the point is I'm not going over that no matter what the title is or how much I want to play it. Thank god I don't suffer from FOMO.
    • spwave 0.9.0-1 by Razvan Serea spwave is a cross-platform audio editor designed for research and advanced analysis. It supports a wide range of audio formats, including WAV, AIFF, MP3, Ogg Vorbis, FLAC, ALAC, raw PCM, and more via plug-ins. spwave offers precise editing tools such as zoom, crop, fade in/out, gain adjustment, and region extraction. It enables detailed spectral and phase analysis and supports unlimited undo/redo. Users can drag and drop files, edit metadata, save labeled regions, and view multiple synchronized waveforms. Internally, spwave processes audio in 64-bit precision, ensuring high accuracy. It runs on Windows, macOS, and Linux, making it a reliable and flexible tool for audio research and editing. spwave has following features: Support for multiple platforms: Windows, macOS, Linux (Motif, gtk), etc. Support for WAV, AIFF, MP3, Ogg Vorbis, FLAC, ALAC, raw, and text files by using plug-ins. Support for many bits/samples: 8bits, 16bits, 24bits, 32bits, 32bits float, 64bits double. Converting the sampling frequency and the bits/sample of a file. Playing, zooming, cropping, deleting, extracting, etc. of a selected region. Fade-in, fade-out, gain adjustment, channel swapping, etc of a selected region. Editing file information that supports comments of WAV and AIFF, and ID3 tag of MP3. Analysis of a selected region using several analysis types, e.g. spectrum, smoothed spectrum, phase, unwrapped phase and group delay. Undoing and redoing without limitation of the number of times. Waveform extraction by drag & drop. Opening files by drag & drop. Autosaving of selected regions (you can do this by drag & drop also). Saving positions and regions as labels. Viewing some waveforms and setting regions synchronously. Almost all processing is 64 bits processing internally. Supported Formats: Read/Write: WAV, AIFF, AIFC, CAF, MP3, Ogg Vorbis, FLAC, ALAC (.caf, .mp4), WMA (Windows), APE, AU/SND, PARIS, NIST, IRCAM, raw PCM, text. Read-only: MPEG-2 Layer 3 MP3, RMP files with VBR support. With 64-bit internal processing, autosave capabilities, and synchronized multi-view waveform editing, spwave is a solid tool for anyone handling complex audio editing or acoustic research. spwave 0.9.0-1 changelog: Implemented CQT spectrum and CQT spectrogram (beta version). Implemented piano-key display for spectrum/spectrogram view. Implemented indication of musical note name in cursor information for spectrum/spectrogram view. Fixed a bug that spectrogram view after zoom-in with large factor sometimes freezes. Fixed a bug that scroll and zoom-out in spectrogram view after zoom-in with large factor do not work correctly. Fixed a bug that spectrogram view provides sometimes wrong time information. Fixed a bug that plugin errors sometimes cause a crash. Fixed a bug that the color of grid lines is wrong in printing. Optimized layout of spectrogram view for printing. Enhanced the function of waveform cropping from label information. Fixed a bug that some items in the preference dialog related to labels do not work. Added some items related to the region label in the preference dialog. Fixed a bug that drawing selected region in the log-frequency axis does not work correctly. Added partial support for the dark mode of Windows (the menu bar and the menus). Fixed a bug that the cursor to indicate current calculation position of spectrogram is sometimes not shown. Changed drawing of cursor information into that with white background so as to make the information legible. Fixed a bug that moving to the head by scrolling the overview display sometimes fails. Added feature of alignment of the view region between spectrum view and spectrogram view. Download: spwave 64-bit | spwave 32-bit | ~3.0 MB (Freeware) Download: spwave ARM64 | 2.9 MB Links: spwave Home page | Other OSes | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
  • Recent Achievements

    • Apprentice
      Wireless wookie went up a rank
      Apprentice
    • Week One Done
      bukro earned a badge
      Week One Done
    • One Year In
      Wulle earned a badge
      One Year In
    • One Month Later
      Wulle earned a badge
      One Month Later
    • One Month Later
      Simmo3D earned a badge
      One Month Later
  • Popular Contributors

    1. 1
      +primortal
      595
    2. 2
      ATLien_0
      277
    3. 3
      +FloatingFatMan
      181
    4. 4
      Michael Scrip
      148
    5. 5
      Steven P.
      111
  • Tell a friend

    Love Neowin? Tell a friend!