Oolite framerates

[Shipbuilder]

Moderator: Split from topic "Suggestion to increase missile damage"

What version of Oolite, and what is your approximate frame rate? (Shift-F)

(And yes, do you have any defensive tech beyond the Shield Boosters and Military Shield Enhancement)

Just out of interest what is considered poor, average and good frame rates ?

[cim]

Just out of interest what is considered poor, average and good frame rates ?

Anything below about 20 is poor. It'll be difficult to have fine control of your ship, the AI will have similar problems, and beam weapons won't work well.

60-100 is good. (i.e. the game is probably being limited by the refresh rate of your monitor, rather than intrinsic performance bottlenecks)

There has been quite a bit of work done in trunk to both improve the game performance at low frame rates, and to optimise the individual frames for speed, so when 1.77 is released it should give much better frame rates than 1.76.1

[Cody]

I get a pretty rock-steady 60fps, even though my monitor's refresh rate is 75 (I've never been able to get it higher than 64fps). <scratches head>
That is in full-screen mode, though - it's possible that the fps could be higher in windowed-mode? (I see no increase in fps between present trunk and 1.76.x, btw).

[CommRLock78]

I get a pretty rock-steady 60fps, even though my monitor's refresh rate is 75 (I've never been able to get it higher than 64fps). <scratches head>
That is in full-screen mode, though - it's possible that the fps could be higher in windowed-mode? (I see no increase in fps between present trunk and 1.76.x, btw).

Funny - my monitor's refresh rate is 60 Hz and I get a fairly stable 75 FPS

[Cody]

I get a pretty rock-steady 60fps, even though my monitor's refresh rate is 75 (I've never been able to get it higher than 64fps). <scratches head>
That is in full-screen mode, though - it's possible that the fps could be higher in windowed-mode? (I see no increase in fps between present trunk and 1.76.x, btw).

Funny - my monitor's refresh rate is 60 Hz and I get a fairly stable 75 FPS

In full-screen mode? Maybe I'll have to try re-setting my monitor to 60!

[cim]

I get a pretty rock-steady 60fps, even though my monitor's refresh rate is 75 (I've never been able to get it higher than 64fps). <scratches head>
That is in full-screen mode, though - it's possible that the fps could be higher in windowed-mode? (I see no increase in fps between present trunk and 1.76.x, btw).

Monitor refresh rate generally depends on resolution, so if you set your desktop resolution lower you might get a higher frame-rate, but windowed/full-screen shouldn't affect it. (Or possibly your graphics card thinks the monitor can only do 60 due to some driver issue).

At any rate, if it's rock-steady, rather than fluctuating up and down wildly, it suggests that the actual frame calculations are not the bottleneck, and waiting for some form of synchronisation is. I get a solid 60 FPS with trunk, but the actual per-frame calculations and drawing only takes about 1/200 of a second most of the time, the rest is waiting for the hardware to be ready ... I only get 30 FPS - fluctuating between 15 and 60 - with 1.76

[CommRLock78]

I get a pretty rock-steady 60fps, even though my monitor's refresh rate is 75 (I've never been able to get it higher than 64fps). <scratches head>
That is in full-screen mode, though - it's possible that the fps could be higher in windowed-mode? (I see no increase in fps between present trunk and 1.76.x, btw).

Funny - my monitor's refresh rate is 60 Hz and I get a fairly stable 75 FPS

In full-screen mode? Maybe I'll have to try re-setting my monitor to 60!

In both full-screen and windowed mode I get 75 - although I'm seeing the occasional 77 FPS in full-screen and the occasional 70 FPS in windowed (so full-screen appears to be just slightly better). What's your monitor's optimum refresh rate? Edit: (I have my monitor set for it's optimum resolution and refresh rate)

[Cody]

Interesting - even back on 1.72 and my old ATI HD3650 I got a rock-steady 60fps.
It's the rock-steady bit I like - but I may have to tinker some, methinks!

[Zieman]

There has been quite a bit of work done in trunk to both improve the game performance at low frame rates, and to optimise the individual frames for speed, so when 1.77 is released it should give much better frame rates than 1.76.1

Does the trunk have some extra checks & stuff running compared to 1.76.1 ?
I ask this because I got lower framerates on trunk (5524) when I last tried it, than on 1.76.1.

[cim]

Does the trunk have some extra checks & stuff running compared to 1.76.1 ?
I ask this because I got lower framerates on trunk (5524) when I last tried it, than on 1.76.1.

Non-deployment builds (as the current trunk nightly builds are) do have some quite expensive checks going on. Exactly how expensive varies quite a bit with your graphics card and driver combination, but they can be substantial. A deployment build of trunk, on the other hand, has fewer checks than the released 1.76.1, so should run faster in that regard.

[another_commander]

Split from "Suggestion to increase missile damage" in Suggestions, as it seemed that there were two parallel discussions going on.

Regarding framerates, one must be very careful when making statements like "1.76.1 runs faster than trunk" or similar. In reality, benchmarking and performance comparisons are extremely complex tasks. I find it very unlikely that someone would be able to define a minor speed improvement in trunk (or a performance regression for that matter) without very well set-up and executed benchmark or without profiling, for those with direct access to the source. I can easily say that trunk runs slower on my system sometimes than 1.76.1, but does this mean anything? Certainly trunk will run slower than 1.76.1 if the test run on trunk was done with 120 entities in the universe and 90 collision checks happening and the test run on 1.76.1 was done with 40 entities and 1 collision check ongoing. In order to be able to directly compare, one must set up a test run which will be exactly the same on both versions and will stress the system in the exact same way and still, things like available computer memory, cache hits etc. should be taken into account when benchmarking between different runs.

So, let's not jump to conclusions about the speed differences between trunk and older versions. There are indeed performance optimizations in trunk at the moment and they should be visible if a proper benchmark is run between the two versions. Would they affect the game in a way that playability is different? I don't really think so. If there are cases where there is a visible performance drop between versions, then my best guess is local driver issues (see recent Tricky's framerate horrors with Nvidia's multi-threading driver setting) and not changes in code performance.

[DaddyHoggy]

For those who are getting reported frame rates from the game than their monitor refresh rate is set at - you could do a couple of things - download FRAPS and run it as a background process - just to confirm the frame rate generated by Oolite itself and the gpu thinks it's running at agree - also check whether Vsync is turned on on your graphics driver - if it is (and your monitor is correctly reporting its refresh rate to the gpu) then if your gpu is running faster than the refresh rate of your monitor it will basically cap out at the refresh rate of your monitor (reduces screen tearing) - so if your gpu can consistently render Oolite faster than this (say 60Hz) you will see (for example) a "rock solid" 60FPS from the game. (Or at least you should!)

[Eric Walch]

I mostly have a framerate of 60 in full screen and 100 in windowed mode with 1.76.

With trunk, I noticed some oddities. When hanging around at zero speed, I see the counter at a stable 99 for a time. Than it becomes jumpy and than it stays stable at 50 for some time. I would not expect that 50 to be so stable, like it is capped at that moment. In trunk (debug build) frametate goes up to 99 in both windowed and full screen.

And, as another_commander writes: you hardly can't compare values by just looking. It not only depends on entity count, but also on how close they are together, or in which direction the player looks. e.g. how many ships and subentities have to be drawn. Generally I see a drop in fps when having a big part of the planet on screen.

Back to the original question: The values for a decent framerate given by cim feel right. For anything of 30 upwards, the player stops noticing differences. For values below 20 the game starts playing worse and the player should think of changing the settings, like shader settings or detail.

A value of 4 is the absolute minimum. When the performance drops below that value, Oolite starts slowing down the clock. It still draws 4 frames per second, but it are no longer real-time seconds anymore. However, the game is un-enjoyable at that rates and is only acceptable for short moments, like an explosion.

[CaptSolo]

Normally, low 60's, but far from rock steady. When the scanner get's real busy it can drop down to 10, which makes the game almost unplayable. But I've been playing this way so long I have learned how to fix it. Lucky for me, I've been a good boy this Christmas!

[Cody]

I mostly have a framerate of 60 in full screen and 100 in windowed mode with 1.76.

Is that something to do with having a Mac, Eric?

Lucky for me, I've been a good boy this Christmas!

That sounds promising, amigo... a whole new machine, perchance?