Jump to content

Noel

Donor
  • Content Count

    5,773
  • Donations

    $30.00 
  • Joined

  • Last visited

Community Reputation

222 Excellent

1 Follower

About Noel

  • Rank
    Member - 5,000+
  • Birthday 02/23/1953

Profile Information

  • Gender
    Male
  • Location
    Arvada, CO
  • Interests
    Golf, music, too many others to mention!

Flight Sim Profile

  • Commercial Member
    No
  • Online Flight Organization Membership
    Other
  • Virtual Airlines
    No

Recent Profile Visitors

6,603 profile views
  1. Good job Marc! It is in complete agreement w/ my anecdotal experience. Unfortunately you were not able to compare vsync to a 30Hz screen which solves the one limitation of RTSS which you mention above, and realtime performance is at least as good as RTSS scanline, and fortunately that option exists for those who don't happen to have a 30Hz capable screen. Hopefully your 'study' will get people using this fabulous way to create ultra smooth flight. I initially thought the main core thread wasn't running warmer using RTSS compared to vsync to 30Hz but I'm prettty sure that is not sure that's the case because you can't rely on the reported core temp to evaluate this if you're using a fan profile to modulate fan speed by core temp. What I find is my fans are spinning much faster using RTSS than w/ vsync to 30Hz.
  2. You're saying when you're in a demanding location like KJFK you're seeing 20-40% CPU load and about the same for GPU load, and have frames at 14-21?
  3. In addition to Shez' outline a couple of other things to get right: 1. Be sure to turn the frame limiter in NCP off. 2. In RTSS hit the Settings button, and on the General tab, select Enable framerate limiter. You won't set a value (of 30, for example) anywhere because when set up like this scan line sync becomes the method to get to 30fps instead of using vsync in-sim, so yes turn OFF vsync in-sim, but keep unlimited on in-sim. You can see I have Scanline sync x/2, where x is entered as '60' to the right, for your screen's refresh rate. I thought that is what this entry was for, but I guess it's not and has to do w/ a scanline offset or something, the '60'. It doesn't seem to matter where I put that but it doesn't hurt it to leave it at 60 FWIW. Here's what I see when it is setup correctly--to the right is what pops up when you hit the settings button in RTSS: If your screen can do 30Hz and you haven't tried it, do. Both work the same to provide smooth stutter-free performance, but w/ direct vsync to a 30Hz screen the CPU core dedicated to P3D's main thread will vary according to demand, whereas w/ RTSS scanline sync that core stays close to or at 100% which isn't all bad, but unnecessarily makes the main thread work harder to get the same result.
  4. Maybe I should have a look under the rock again eh? Well, this belief came from the days of FSX and it applied right on up at last thru P3D 3.4. For years in FSX I used the internal limiter and the results were okay enough to live with it assuming it was as good as it could get given the hardware and sim. Then one day I noticed in a particularly complex scene if I set the sim up with a frame rate limit internally of 30, I would only get 23-24, but if I switched to unlimited, I was getting 30 or more. From there I thought an external limiter might be a way to capture the better frame rate, because for certain a frame rate of 23-4 even if the frames intervals are equal, isn't going to have a very fluid appearance. I managed to find something on a gamer site that worked quite well but I don't recall its name now and it was new to the FS community. It was a contribution I felt responsible for even though I didn't produce the applet. This same reality happened w/ P3D up to 3.4 and on 4 different PCs. The internal limiter clamps total performance in the end, and always did as I say. Maybe it's changed now w/ better hardware, meaning the hardware is sufficient to overcome the penalty incurred w/ the internal limiter. I admit it I haven't tried it since installing P3D 4.5. When I finally discovered vsync>30Hz screen 5y ago that was the end of less than excellent fluid video and that logic applies to both the former last PC and my current recently built one. Hence, I'm a staunch proponent that any method that allows one to stay unlimited in-sim yet limit frames has the best potential to yield consistently best results. For some reason NvCP's limiter created stutters, not so w/ RTSS scan line. After flying 4 or 5 flights using RTSS scanline to a 60Hz screen while quite good in terms of fluidity in the end vsync>30hz screen seems best, and that mainly because the main thread is not hammered on unless...it is. That is, reported % load is commensurate w/ the context, whereas w/ RTSS it's always near or at 100%.
  5. A hundred years ago I had these little errors, changes, weird issues that would crop up every few days and nothing had changed either. Turned out the hard drive the OS was on was slowing failing.
  6. Liquid smooth RTSS scanline sync 60Hz/2. NVCP produced wretched stutters for me and both RTSS and vsync to 30Hz produce fluid stutter-free video.
  7. You're set up this way? In P3D unlimited, vsync off. Monitor refresh 60Hz. RTSS scan line x/2. AND, don't forget to check the box Enable Framerate Limiter in RTSS' settings page, General tab. This is what allows you to turn vsync off in-sim and still have the 60Hz/2 sync to scan line. Are you monitoring your GPU load closely? As you may know you can monitor GPU and CPU loads and temps via MSI Afterburner, so it's a handy way to watch because when you vsync using RTSS as soon as GPU load hits 100% you WILL NOT be able to maintain the 30fps required for that stutter-free video. If your GPU load is not approaching 100% where you see these stutters the next thing to do is dial back the less-GPU related scenery sliders, plus cloud distance, traffic, etc. IOW, for troubleshooting your major stutters, dial these back as needed until your video is golden smooth, then add back what you prioritize you want. I know there are certain features that have an inordinate impact on the main CPU thread and that would be road traffic so consider turning that off if you haven't already. I don't know what hardware you're using but I know I had no fluid video performance w/o stutters, save the rare 'long frame', in a 7 y/o system using P3D 3.4. Performance is identical in my new system, the only thing different is I am running some sliders up more than previously as predicted. Having had the two very different systems, with smooth performance in each, leaves me relatively convinced people ought to be able to achieve the same result. But I know there are going to be exceptions for myriad reasons. For the past 5y I have used unlimted/vsync to 30Hz screen and no RTSS and the result is really the same as w/ RTSS w/ scan line x/2 save the 60Hz screen, AND the fact it's easier to see how 'loaded' the main thread is using the vsync to 30Hz screen leaves a bit more in favor of vsync to 30Hz.
  8. Well, when I said FFTF Dynamic should help the issue of "The fasted loading scenario may be too much work in the background...", you immediately replied, 'If in FPS mode'.
  9. I will have a look and see. Off the cuff, what AM would you predict would be the best balance for me for HT enabled? Right now w/ the mask I am using for 8 core non HT I have Why would frames mode be better than AGL mode? Right now I am playing w/ AGL mode under the belief in general terrain loading demand relates fairly well to AGL and my hypothesis is that the long pauses happen when P3D decides it needs to load a giant terrain tile, decides it too late in the game, and that by picking those off earlier (in a descent for example) this might obviate the rare long frame I get. Because I'm vsync'd via RTSS to 1/2 of 60Hz all the matters is maintaining at least 30 frames. What range would I chose then? 29-32 or something along those lines?
  10. Can you comment FFTF Dynamic should help fine tune this issue I would think.
  11. Nope, as in my sig, Noctua NH-D15S. When ambient is around 21C w/ HT on stays around 65-67C, so w/ HT off around 55-56C.
  12. I will play with it again, HT enabled again, and see. I'm all for longevity in this new build so running cooler is definitely a means to help promote longevity, at least to some degree pardon the pun.
  13. Thanks for that it makes sense! What is that 'task that occupies the remaining percent....'? As you mention the main thread seems to be getting the same amount of work done using RTSS w/ a 100% load, 'same on the CPU more or less', but it's nice to be able to see when I'm overloading the main thread w/ too much work via sliders pushed too far right, and we don't get to see that w/ RTSS and no vsync in-sim.
  14. I've had such good results w/ HT off I'm thinking it is likely not worth the extra 10 degrees Celsius, though core temp w/ HT under load is still under 70C. Since stuttering is no more the only areas I can conceive of improving on is: 1, ability to bring in more objects, i.e. set some scenery sliders a little farther right, and they are already kind of good enough and 2, to see if using more LPs to bring in data will reduce the risk of 'long frames', those 1-2 second hard pauses. Since using FFTF Dynamic I seem to virtually eliminated those already, but not quite!
×
×
  • Create New...