A Frame-Time Analysis of P3D v3 -- Effects of CPU Affinity, Frame Lock, and HT

[BrettT 63]

Thanks Steve for the explanation.  I like knowing how things work and have enjoyed following your discussions on HT and masking etc

[AviatorMoser 353]

With Unlimited, each next frame is calculated based on the time it took to render the last frame, so more or less never correct and generally results in stutter, and poor panning fluidity.

 

Pretty much sums up my experience with unlimited frames right there.

[SteveW 1,792]

Thanks Steve for the explanation.  I like knowing how things work and have enjoyed following your discussions on HT and masking etc

Appreciated.  :smile:

[DylanM 1,069]

Pretty much sums up my experience with unlimited frames right there.

To play devil's advocate - this is the opposite experience of mine. Adding fixed frames (with combinations of HT / AM / vsync on and off) is the cause of stutters on my system. Fixed frames, even when combined with a lower FFTF, generated more long frames than unlimited (with a default FFTF).

 

This scaled between my older i7930 / GTX680 rig and my new 5930K / GTX980 one.

[carlito777 712]

To play devil's advocate - this is the opposite experience of mine. Adding fixed frames (with combinations of HT / AM / vsync on and off) is the cause of stutters on my system. Fixed frames, even when combined with a lower FFTF, generated more long frames than unlimited (with a default FFTF).

 

This scaled between my older i7930 / GTX680 rig and my new 5930K / GTX980 one.

 

Same here. Many more stutters with fixed frames... (4770K@4.6GHz + GTX 780)

[SteveW 1,792]

Many more stutters with fixed frames

In that case I would say you need to set a lower fixed fps. What did you try?

[pmb 5,137]

Same here. Many more stutters with fixed frames... (4770K@4.6GHz + GTX 780)

Hi,

 

I was under the same impression. Until I made a quantitative analysis. This is

 

3) HT off, no AM, frames unlimited

 

 

Compare this to the upper diagram 1) in my post #93. Same settings (except fps), same circuit (KSFO), Frames are (much) higher, but jitter is as well. I am still surprized by the result, but have to accept it.

 

Of course your system may behave in a different way.

 

Kind regards, Michael

[AviatorMoser 353]

Indeed Michael,

 

While there may be bursts of higher frame rates, the stability of the rendering is poorer. Some eyes may not pick this up, and therefore it many not matter at all, but my eyes are very sensitive to such jitter.

[DylanM 1,069]

I am very sensitive to stutters and long frames (many years of audio & video editing make it almost a curse...), and unlimited is clearly better on my rig. Again, not saying that the opposite is not true given the significant differences in our systems, just that there is no cure-all for everyone.  

 

I'm actually very glad you undertook this analysis, as it highlights some great levers to reduce / eliminate stutters - something that was the thorn in my side back in FSX. 

 

p.s. have you also experimented running with vsync & triple buffering on/off?

[hesynergy 67]

Yes, beware the locked setting generates look ahead frames and requires more or less double the fps showing at all times at Unlimited, to do it justice.

Sure sounds like you know what you're talking about… Anybody else in the scientific simulator community care to chime in?

 

Chas

[AviatorMoser 353]

I am very sensitive to stutters and long frames (many years of audio & video editing make it almost a curse...), and unlimited is clearly better on my rig. Again, not saying that the opposite is not true given the significant differences in our systems, just that there is no cure-all for everyone.  

 

p.s. have you also experimented running with vsync & triple buffering on/off?

 

I must apologize. I must have sounded a bit pompous saying that. I don't think my eyes are better than others, and I do believe others when they say frames unlimited gives them the smoothest results. I just wonder why there are opposite results. Too many variables to consider, I'm afraid.

 

I did attempt v-sync and triple buffering in the first page of the results, but if I recall correctly, they did nothing to reduce the micro-stuttering,

[pmb 5,137]

p.s. have you also experimented running with vsync & triple buffering on/off?

No, at least not yet. My experience has been, at least internal Vsync rather tends to lower performance, but as we see one should be careful about qualitative statements.

 

Next, you have a host of parameters, internal Vsync, Triple Buffering, external Vsync, Triple Buffering set via NI, Standard/Adaptive, combine this with the monitor frequency (30 Hz vs. 60 Hz on my monitor) and different fps settings. This would certainly need some time to perform carefully (and understand the results). No doubt it would be interesting, though.

 

Kind regards, Michael

[jjjallen 37]

To play devil's advocate - this is the opposite experience of mine. Adding fixed frames (with combinations of HT / AM / vsync on and off) is the cause of stutters on my system. Fixed frames, even when combined with a lower FFTF, generated more long frames than unlimited (with a default FFTF).

 

This scaled between my older i7930 / GTX680 rig and my new 5930K / GTX980 one.

Same here.  Unlimited since 2009 and still doing fine.  Make sure that you up the FSX OS Priority.  I get blurries if not using REALTIME priority. 

 

Cheers

jja

[z06z33 763]

I've tried the various settings but I still see the most performance with HT enabled using  AM=224

[radial360 2]

Good Morning all!

 

what an interesting reading! Thanks for all your efforts.

 

I can't get my head around all the information we have here. I am currently running an i7 6700k at 5GhZ. I am running HT OFF and AF14. 

 

What changes would you suggest me to try, so I can post the results back to you guys?

 

HTON AF85? AF84?

 

Thanks a lot.

 

Gio