Can you see the difference between 30fps and 60fps?

[Tierborn 197]

@kevinfirth,

I remember seeing a YouTube "blind frame rate test" for gamers a few years ago. The video showed frame rates of 30, 60, and 90 (blind test).   They asked which was smoother.  Most could not tell the difference.  Then, the test went on to tell the testers that 60fps was 30fps and 30fps was 60fps and like clockwork, they all thought 30fps (told it was 60fps) was smoother.

Edited April 3 by Ray Proudfoot
Long quoted post removed

[Lonesome Cowboy Burt 175]

Still most are avoiding that watching a video is not the same as interacting with a game, FPS has an affect on smoothness of moving objects on which that that interaction occurs that is helped by framerates beyond 30fps.

I.E. any input part way through a frametime wont be recognised and acted on till the next frame is generated, its not very important in a flight sim, but expert shooter players can certainly tell the difference and use it.

[Lucky38i 1,110]

I guess people could just use UFO test and really let it speak for itself https://www.testufo.com/framerates#count=3&background=stars&pps=1200

If some were to try something besides MSFS to do some kind of fps comparison test the difference is very much apparent. I can guarantee everyone here has atleast 60hz screen to see the difference.

30fps vs 60fps isn't some status myth about who can afford the best tech (because 60+ fps is very much attainable in a well optimised game without the need of dropping a lot of cash). If you go outside of the sphere of poor optimisation that harbors most flight sims and try out other games, adjusting settings to retain low 1% averages at 60 vs 30fps there is very noticeable difference. Saying it's a placebo is a pretty outlandish statement when this discussion has been debated to death already, for years at that.

Some forget that it's not just FPS but also average frametime, you may be attaining 60fps but your 1% lows are absolutely atrocious, providing the feeling of stutteriness and chopiness that you don't see at 30fps. It's why if you genuinely think there's no difference it's most likely because you experienced this feeling of choppiness. I highly suggest to those curious to run the testufo to see a vsynced test of 60fps and 30fps, and if you monitor supports it, 120fps and 144fps, assuming you've ensured you're running that refresh rate on your machines.

[pstrub 998]

10 minutes ago, Lucky38i said:

I guess people could just use UFO test and really let it speak for itself https://www.testufo.com/framerates#count=3&background=stars&pps=1200

If some were to try something besides MSFS to do some kind of fps comparison test the difference is very much apparent. I can guarantee everyone here has atleast 60hz screen to see the difference.

I was just looking looking for this link but you beat me to it 😉

Anyway, I don't understand why some people can't accept that there are big individual differences in a) perception of smoothness, b) tolerance to lower fps (as long as they're stable), and c) the need for low latency, especially with head tracking and VR.

In addition to that, the picture of moving objects on many monitors (LCD / TFT type, not on OLEDs) looks blurrier when the framerate isn't high enough. This improved quite a bit for me after I started to use frame generation.

[threexgreen 3,160]

10 hours ago, turbomax said:

I suffered from below 30 fps frustration in VR (which I use exclusively) on my previous RTX 3090, but now absolutely enjoy mostly 50-60 smooth fps in above 4K in VR after I "forked out" the admittedly ridiculous 4090's price

I did upgrade my CPU, but I stayed with my 3090. Does a 4090 make that much of a difference in VR? Do you fly GA or airliners out of big airports?

[s0cks 135]

51 minutes ago, Lucky38i said:

I guess people could just use UFO test and really let it speak for itself https://www.testufo.com/framerates#count=3&background=stars&pps=1200

Wow that is a cool demonstration. It's interesting how different it looks in your peripheral compared to directly.

[Sethos 4,541]

1 hour ago, Lucky38i said:

30fps vs 60fps isn't some status myth about who can afford the best tech (because 60+ fps is very much attainable in a well optimised game without the need of dropping a lot of cash).

I think that idea is still prevalent in the flight simulation community thanks to FSX and P3D. They ran so poorly even on the highest end hardware, that "lock at 30FPS" mentality just became so ingrained, that especially a lot of the old guard simmers have this idea that anything beyond 30 is just some superfluous excess that isn't necessary, like they are stuck in 2006. Meanwhile, the rest of the world and every other simulator understood that more frames is just better, a more pleasant experience and we actually went well past that mythical 60FPS over a decade ago.

 

Then again, the P3D holdouts are still buying thousands of dollars worth of the best hardware on the market and are still chasing an even smoother 30FPS in 2024.

Edited April 3 by Sethos

[Cpt_Piett 4,836]

12 hours ago, MrBitstFlyer said:

This video reminds me of the post about turbulence being so bad that the sim was unflyable, yet when video evidence was provided there was virtually no turbulence at all.

Ah yes, the turbulence thread! I thought about bringing that one up, glad you did 😅 That really went on for ages, wonder if it died of natural causes or got mod-locked? 

But anyway, you’re right, the difference in my video is subtle and certainly not enough to “win” this case. Which was not my intention. After all I should recuse myself from this discussion as clearly I’m suffering from the “4090 owner syndrome”. 

11 hours ago, s0cks said:

Videos probably aren't the best at showing it, as codecs will mess with how it appears compared to in person. Even so, I can see the difference there.

Exactly! I swear the difference was much more pronounced in the sim, with 30 fps being a lot more jerky even with slight panning movements. And writing this I can already hear the sighs of disbelief. I ended up chasing my own tail making the video (nVidia overlay settings, codec settings etc), should have just picked up my iPhone instead…

By the way lately I’ve sacrificed performance for visual quality. E.g. even though I can get to 120fps in some situation, I don’t see the need for that and instead lock at 70 and crank up the TLOD. Conversely, if I can’t maintain 60+ FPS, I turn down everything until I can get a stable at least 65 FPS.

For the last few days I’ve even turned off AI traffic to be able to fly into major airports in the A300 at 70 FPS with a completely flat frame time graph. It’s an obsession, yes, but so is a lot of aspects in this hobby. 

6 hours ago, JetCat said:

Most important factor is having the graphics card saving as much power as it can and the whole computer running ultrasilent - below 50 watts and with no fans if possible for the graphics card, and below 50 watts for the CPU with idling fans on the cooling tower.

You would be absolutely appalled at how my PC runs then 😅 I’ve got six 120mm fans running at full speed at times to cool the CPU and motherboard. And I won’t even disclose the energy consumption and the fan noise of the 4090. Nor that I’m using two power supplies; 750W+850W 😆

12 hours ago, MrBitstFlyer said:

If the comparison was made in normal flight, and not rotating like a fairground ride, the differences would be even less perceivable.

Not sure why your last comment got me a bit upset… But why don’t you make a video demonstrating why there is allegedly a minor difference between 38 and 60? 

*putting on my armor and sharpening my broadsword*

Edited April 3 by Cpt_Piett

[Ixoye 1,841]

2 hours ago, bobcat999 said:

Interesting!  Why 97fps? 

I use V-Sync so it will lock automatic 3 fps lower than my monitors refresh rate (100hz)

[kerosene31 349]

I bought an F-16 off the MSFS sale, and I had downloaded a freeware Nellis AFB (which is very nice, but very FPS heavy).  I take off and my FPS (normally in the 50-70 range, depending greatly on where I am and what I'm flying) dipped down into the 30-40 range.  I didn't even have the FPS counter on, but I instantly noticed the massive drop. 

This isn't a placebo, I didn't even have my shift-z stats mod displayed, and I wasn't even thinking about FPS.  All of a sudden a nimble fighter jet was handling like someone strapped busses under the wings.  It feels like there's now more delay between the controls and what happens on screen (I'm guessing that is actually mostly an illusion, I doubt it really impacts your inputs, but it feels like it does). 

And no, this wasn't just the initial drop in FPS (which is clearly going to make things choppy), but I stayed consistently at 30-40 around that immediate area.  At that point I did turn on my FPS counter and things were not erratic, but pretty consistent. 

For my eyes, 30 fps will never, ever be smooth.  Now, can you fly a heavy commercial airliner at 30fps?  Of course, a fighter plane is an extreme example.  Obviously the goal is to fly an airliner realistically and smoothly, so you are a lot less likely to notice the FPS.  I remember back to the old days of MSFS getting 20 fps on final approach into busy airports. 

[Christopher Low 5,398]

I do not actually know what framerates I am getting in MSFS, because I never check them. All I know is that the performance is extremely impressive for a CPU/GPU combo that were released in January 2017 and June 2015 respectively!

[ErichB 3,651]

I always thought it was nonsense that you could see the difference. Now that I finally have a system which can do 60fps,  with ease,  in all situations, I've realised that you definitely can see the difference.

Edited April 3 by ErichB

[Noel 4,563]

16 hours ago, bobcat999 said:

Noel.  I am not sure what people are going on about here.

I only "went on about" your comment comparing your 30, over his 30, nothing more, nothing less.  Further, if you look at my screen, when panning slowly or rapidly I don't believe you would call that "jerky".  It may be with how you're setup, but not here.  And yes I acknowledge the clear downside of lower frame rates in terms of objects at point blank range being effectively 'stretched' as it were with faster panning actions.   Back in FSX/P3D days I struggled to maintain a rate of 30, and THAT 30, is nowhere near as quality as the current 30 with ultra low FTV.  I can tell the difference between 30, and even 34. Once at 38 or above, sterling FTV, going up beyond that is visible still (a little less "stretch w/ panning"), but doesn't add enough to put up with artifacts out of FG for me.

Edited April 3 by Noel

[Cpt_Piett 4,836]

2 hours ago, Sethos said:

I think that idea is still prevalent in the flight simulation community thanks to FSX and P3D. They ran so poorly even on the highest end hardware, that "lock at 30FPS" mentality just became so ingrained, that especially a lot of the old guard simmers have this idea that anything beyond 30 is just some superfluous excess that isn't necessary, like they are stuck in 2006. Meanwhile, the rest of the world and every other simulator understood that more frames is just better, a more pleasant experience and we actually went well past that mythical 60FPS over a decade ago.

I think this is a very valid point. In fact I locked FPS at 30 for many years. If anyone back then had tried to convince me of the benefits of 60 FPS I would have laughed at such a preposterous statement! 

[Noel 4,563]

I think a lot of the discussion on this centers around lack of clarity about what is being said.  The perception of smooth animation at lower frame rates is intimately tied to frame time variance.  At the low rate of 30 when FTV is near perfect, the perception is ultra smooth animation, even with panning--slow or rapid.  That, however, does not address the other part of lower frame rates, as I described it, "stretching" near field objects during panning--the more rapid the pan, the more evident the stretch. These are two completely different elements of visual performance as it were.  I've always maintained there is an inverse relationship between FTV and rate:  the higher the rate, the less critical FTV perfection becomes.  Early on I stated "brute force" could overcome poor FTV, and I fully believe that is what is happening with many or most Frame Generation users, except for myself and CptPiett and a few others, who adopted my method of improving FTV.

All that being said, am I all cozy and happy with perfect FTV at a rate of 30?  No, because of the other element involved, that stretching that happens with near field panning objects.  Is it a show stopper like stuttering is?  Not at all--panning is still excellent in itself, jerk-free.  I find perfect FTV at a rate of 36 or 38 is approaching perfection as it's improving the 2nd issue sufficiently, for me.   When my new 7800X3D is built I hope to be able to move that up to around 45 to 50 at which point I'll be done with working on performance per se.  Frame Generation turned out to be net loser here and there is a possibility the new CPU will help with that, we'll see in a few weeks.