I have run laps on all three configs of the Nord now and i am not seeing any FPS loss on the 24 hour version. It is stuck at 60fps for me, and i even had 50 AI out there with me. The 24 hour version is amazing. All three are, but the 24h...slobber. Overall, the RRE version of the Nord is amazingly done and simply the best you can do without actually going to Germany to be there.
I think no one has claimed that the track can't be run at 60fps. It all comes to system specs, resolution and settings, which information no one here has fully told. But we talk about performance hit here. As I said, if you have enough FPS-buffer above that 60FPS, you take the hit, but remain above it, but the hit is there. Those, who don't have enough buffer, see the drop below 60. I'd be interested to see for example you, who seem to be fine, test both layouts without vsync and see if there is a big difference. If not, it is then related to certain system components or drivers, which make the worse performance. Then we would at least have something to go with here.
Cheers. If you see FPS difference in double digits, it's there Edit: and without knowing your exact in game settings, turn on at least full track details. The moar the better, so GPU gets something to do. AI is not important here
Very much inline with my setup without knowing pretty much anything about your setup, like settings, resolution, CPU etc. I get the minimums with 99 AI around 38 FPS. I have triples, you may have smaller resolution, beefier CPU, who knows and your 6 frames is there. No differences there, I'd say. Besides, there is not much point to test graphical performance much above 50 cars, as maximum visible vehicles is still 52. Everything more than that merely burdens CPU with AI calculations. And as said, cars aren't the problem, it's the track. If 24h is poorly optimised, performance difference is there even if you drive solo.
Just tried the 24hr version of Nord with vsync OFF and it ran at 75-80fps. BUT i didnt like the jaggies and it runs better with vsync on at 60fps.
i7 3770k 4.5 , 1920x1080 , maxed . Yes its something with optimisation , i get similar fps with only 30 cars
Yep, we get pretty much the same performance. You have only one screen, I have three of them, and then there are differences like the CPU, but we are at least on par here and no strangeness going on
I'm not having any problems with fps on the ring, been playing with a 60 cars on track, settings on full and I'm always staying in the mid 70/90's depending where I am on the track, but I do lose about another 15/20 fps when the ring rejoins the GP layout again, which is intresting, as with AC & iracing version you get the same effect, also pcars doesn't have the 24hr layout, but the pit straight also suffers with low fps.
So you get pretty much the same performance (within few FPS or so), with VLN and 24h layouts? I'm not interested in the specific FPS on one layout, but the possible difference between VLN and 24 h. It may well be 300 FPS vs 200 FPS or 60 FPS vs 40 FPS, when it's significant. Or then it's insignificant, let's say 300 FPS vs 270 FPS or 60 FPS vs 56 FPS.
Many of you who post here haven't got your computer memory specs layed out. Just to remind you, that the DX9 API requires the VRAM to be also be copied and be present in system ram at the same time, just laying out that you got a sun for a GPU and a 500 gazillion gigahertz processor with a kiwi on top doesn't give much information to the devs. Its a large track with lots of textures to keep in the memory. Also seeing that some of you have 16gb of ram, can you post your memory usage and commit charge to rule a simple memory problem out easily.
Same for me. Really good work with the track, as always! But the 24h layout is useless for me, even when I lower the ghrapics and grid sizes below what I normally use this layout stutters really bad, and have awfully low fps.. . The other layouts runs smooth with maxed graphics.
I have 12Gb RAM and 2 video cards AMD FirePro D300 (in Win 10 its looks like AMD R9 200 series) with 2Gb VRAM each. Haven't any problem with 24h layout. Only if i start full grid (99 opponents) need to drop settings to Mid from maxed. No difference run it with one graphic card or both
Now before this gets foolish. R3E is 32 bit program and can map 4GB VAS, which is not the same than R3E executable consuming 4GB. I won't explain VAS here, everyone can google. Then there is VRAM, which is memory on your video card, nowadays typically 2GB-6GB. For DX9, you need to map stuff on your VRAM to your RAM, but not all of it. It depends on different factors, but it is a lot less than your GPU's VRAM consumpiton. Then there is this funny DX9 Win 8 and 10 4GB VRAM cap, which means that with those OS's your GPU can use at maximum 4GB of VRAM (of which part of is again mapped to system RAM). This is not fault of the OSs itself, should be handled by the driver, but at least by my knowledge, Nvidia has still this cap for DX9. I can be wrong, but it merely improves the situation. Now, I considered at some point, if my VRAM usage is above that 4GB level on 24h, which would mean that VRAM exceeding that limit would be mapped straight into infinitely slower system RAM (that means GPU is really writing and reading from there). If that would happen, it would definitely explain the slow down. Capping 4GB VRAM would be something to consider for those running high settings AND high resolution. But that is not the case. I couldn't exceed 4GB VRAM in 24h with my regular settings, although I made close to around 3.8GB. To be sure, I then tested lower texture settings in game shaving around 800GB of VRAM to make sure that its not the reason. Result: no effect on performance. And how it is related to system memory, then? As a 32-bit software, R3E can consume at maximum 4GB of VAS and that means that the executable itself can consume considerably less. If that 4GB is full, R3E crashes. The stuff that is mapped from VRAM uses different address space and let's assume that the whole VRAM is mapped, it would be that 3.8GB. That means that much less than 8GB of system memory is used in a terribly worst scenario. To be realistic, I'd say in my case, system memory consumption with both the executable and mapped VRAM is much less than 6GB, because even if rrre.exe would be nearing that 4GB VAS, the program itself would consume perhaps around 3GB, rest is different processes already reserved by OS. Now, I have 16GB of RAM and when I'm now looking at my memory consumption without running the game, its at best around 4GB with several programs and windows open using different memory spaces. Some of that stuff is same that R3E needs for its VAS, that is, the physical memory is already consumed. That means total consumpion is at worst around 12GB, in truth several gigabytes less. I'd most likely do okay with 8GB of RAM. I was in truth concerned about that DX9 cap, but that was not the case. Thanks anyway
hm yeah, fps drop here too. On VLN (50 visible opponents) i have 130 FPS, on 24h layout it's more like 100 FPS. Video Settings all maximum. i7 6700K GTX 970 16GB DDR4 Maybe I try with maximum visible opponents next time.
You can, but opponents aren't the factor here. Of course they have an effect, there are more polygons to draw with more cars and more job to do for CPU with more AI. But for me and many here claiming the performance issue, it is the difference between those layouts that matter. Amount of cars has a small effect.
i7 4770K watercooled 4.7 GHz GPU Zotac 980 GTX GPU clock 1450 MHz, 16 gig main ram. Playing at 1440p(2560x1440) Nvidia G-Sync 144Hz. Using Windows 10 64 bit and Win 7 64 bit Ultimate both same story. Again guys when I test something I am testing it 100%! Check out the crowd in the 24h layout they are bitmaps no real 3D polygons. So there is basically nothing special to calculate. You guys are right about the start in the stadium I get 38 fps gets higher when I get out to 44 fps and when I drive into the Nordschleife it boosts to 48 fps and it stays there for no explainable reason. And yes Grunt is absolutely right about the cars and AI the calculation is mainly cpu related. On all the other layouts 90-100 fps all maxed and even in some sectors boosts to 100-138 fps. The peak was 138 fps. I hope S3S will fix this soon. So far none of the devs have shown up and gave us any feedback concerning this matter! I am a man of principles and honor. I have invested over 150 Euro in this baby. The price politics in this game is very questionable! If we do not get any response on this or it won't get fixed, I won't invest any more money in this game. I am always straightforward and let's face it guys still no day and night cycles and no weather system. Rfactor 2 is my favourite in this regard. Hence even the average Project Cars has day and night cycles and a weather system. To S3S I am just saying I have invested a lot of money in your baby and therefore I demand a certain quality level. The 24h layout should have never been released in this state! It is bad optimized no matter what settings I pick. My dear devs come here to the forum and talk to your loyal customers. Explain yourself whether this issue will be fixed or not. Thank you!