Jump to content

roesti

Members
  • Content Count

    92
  • Donations

    $0.00 
  • Joined

  • Last visited

Community Reputation

71 Good

Profile Information

  • Gender
    Male

Flight Sim Profile

  • Commercial Member
    No
  • Online Flight Organization Membership
    I belong to both VATSIM & IVAO
  • Virtual Airlines
    No

Recent Profile Visitors

1,090 profile views
  1. Hello Ray, I sent you a PM with the file I created a while ago for P3D v5. I don't know if it still works because I haven't flown P3D for a very long time. Let me know if it helps.
  2. Yes, I'm absolutely aware. Like I said, the major reason for me to keep P3D is the Majestic, and given the new installer options, it's possible it will be compatible from day 1 so I can be happy. It would be nice to have the PMDG heavies available as well, but I know the focus of this developer is no longer on P3D and if the planes are not easily compatible (just copy & paste some stuff and adjust folders), maybe there won't be an update. But since I have no expectations and know it might not even happen, I can live with that. I also remember the v5 release with the "invisible walls" bug and certain scenery. Something similar like this can happen for sure and maybe this time it maybe won't be easily fixable with ADE or some exclusion files. We have to see 🙂 But if I'm able to fly the Q400 from time to time and v6 has something to offer which justifies an upgrade from v5, I will get it. If not, v5 won't go anywhere and I will continue using the older version. I agree 🙂
  3. Honestly nothing, I have no expectations. But I'm curious to see what they have to offer to justify a new major version. Eventhough I don't use P3D as often as in the past, there's still stuff like the Majestic Dash-8 (which ironically already got an installer for v6) that will keep this simulator on my SSD. And if there's something new to look forward that I might like, I guess I will buy at least the acedemic version. One thing I will definetly do (as always) is buying the $10 1-month subscription, just for testing purposes.
  4. The whole shop is currently down (click any button like subscription and you get a timeout - edit: back again, now only "Product not found"), so it's clear they're preparing to release something in the near future. This is no surprise, as there were upcoming product IDs in the shop before - then they realized there are actually people interested in finding out those "secrets" and they try to hide it, because it's not ready for the public yet. The v6 beta page was also real, saw it with my own eyes until it got removed. I'm certain the release is very close, maybe a few weeks. I know it's not how it works to predict an exact release date, but if you look at the release cycle of major versions (v1 - v5) in the past years, it's "about time" for v6 to see the light. My guess is it comes in May (by June at the latest). Can't garantuee anything, but this is just my opinion based on past experiences. Nah, I doubt it. I think this is a little bit far-fetched (while still possible, highly unlikely). It's not uncommon to just copy & paste placeholder text if you insert new shop items. I don't think the release is ready and this shop site was not expected to be found (today). Just preparation work, because there are many existing IDs in the shop that are not listed / linked in the shop. It has always been like that in the past, I remember similar stuff shortly before the release of v4 and v5. For me, it seems legit.
  5. It is not neccessary to set full paths for subfolders in the .xml file. If you want to use relative paths, use .\ For example <Category>Sound</Category> <Path>Content\Sound</Path> </AddOn.Component> will work, if you change it to <Category>Sound</Category> <Path>.\Content\Sound</Path> </AddOn.Component> you can even move all the way up and down in the folder structure using things like "..\", which will refer to the parent directory. The relative path is always dependant on where the add-on.xml resides.
  6. Thanks for chiming in. I hope it was clear I wasn't doubting your findings that in your specific case the upgrade did not bring you any gains. Just for this specific situation, the OP won't know if this applies to him as well, if the parameters are not known. It was only a reply to the comment "No gain in P3D" towards me. Thank you for giving your resolution setting. I think the easiest way to compare is just taking screenshots from the options, but you don't have to do this now 🙂 It will always be hard to do meaningful comparisons, because there are so many different systems, settings and add-ons people use. I guess the only way to compare performance in P3D in a meaningful way is without using any add-ons, pre-defined settings and easy recreatable benchmark scenario (maybe some kind of replay). If people are interested in this, I can try to create such kind of benchmark and we can discuss about this on a separate topic. For now (and for everybody who wants to know if an upgrade is worth it or not), I remain recommending going by the way I suggested: Turning down resolution and watch the FPS (of course without frame rate limitation). It will tell you really quick if the CPU is the bottleneck currently or not. If FPS rise on a lower resolution, the CPU is still fine.
  7. Did you read the articles I linked and also the one by F737MAX? You will also find many stuff via google. If you don't want to read it: Basically, UB does not state which kind of "metrics" they use. On one side, they state things like "single core performance" with gains towards Intel while this doesn't reflect with real world tests or benchmarks in actual software / games tested by serious hardware news media. Then they put thing in the score like "more recent" or other stuff, like when they evaluate "slightly better latency" in favor to old Intel processors - at the same time, latency benchmarks by real world testers even contradict these "scores" of Userbenchmark while there's no way to repdoruce the claims of this site. If you are really interested in the topic, please read the two articles I linked. So what is a benchmark what focuses on real metrics? A benchmark I can run myself, which in the best case uses an application or game existing in the real world. So that I, as a user, can draw conclusions for my own situation out of it. UserBenchmark does not give one any of this while also skewing it's own scores in a weird way where some 8 year old processors seem to perform on a level like a newer one for example because of weird "evaluating" of subscores. A single user without saying anything about his settings says "no gains". That's why I wrote it's a shame no settings are known. I say there can be severe gains depending on the settings and I think I explained why there is no absolute way to say if a CPU upgrade might help or not and gave instructions how to test it. Furthermore, I explained that a CPU upgrade rarely helps when using resolution at 4K and above, because most graphics card will limit the performance (today) at this resolution. The 4090 is the first card where you can be CPU limited in 4K (in some games). Benchmarking software or games only work if the settings are known, so that a meaningful comparisons can be made. Otherwise it might happen users spend their hard earned money on hardware they don't need. If I don't know the settings, I can't say if a specific hardware upgrade helps. If user 1 plays in 4K and user 2 plays in 1080p, then user 2 would most likely see an FPS increase with a CPU upgrade, while user 1 does not.
  8. Thanks for linking this topic. It's a bit of a shame that he didn't talk about his settings. Upgrading a CPU can always be zero gains or huge ones. It all depends on settings and that's why I always recommend testing CPU / GPU limits on the users side. It's actually that easy: Disable any AA setting and put your resolution down to 720p. If the FPS do NOT increase, a new CPU will help. If it does increase significantly, then your GPU limits. So in the 2nd case, if the FPS you get at a low resolution is enough for you, you won't need a new CPU. Of course there are more accurate ways to measurements with watching usage of the compoments, but this quick and simple test will give you a quick answer.
  9. I see your point, but benchmarking at 4K is still not "good" for comparing CPUs. The reason is, the higher the resolution, the more the GPU limits. This is my "critic" with Rob's benchmarks on the FSLabs forums, because he does CPU comparisons with an effective resolution of 8K, which isn't quite working. He even mentions there he'd expected a better performance, but if I had seen this thread before, I could've told him in advance this would be the outcome. But I don't want to talk behind people's backs. His work there is still valuable, even if it's only for people having him test this now knowing, that they might not need a new CPU when they run their simulators at this insane resolution like he does 🙂 There's no GPU which doesn't limit in 8K resolution today.. Fortunately with the 4090 released, we now might have the first card not limiting at 4K resolution anymore (in a few scenarios, but not completely). Usually CPU benchmarks are even done in 720p. Why is that so? Because only then we know what the CPU can deliver. The CPU and GPU are always indepdendent. Meaning if my CPU1 can deliver 180FPS, CPU2 120FPS and CPU3 84FPS but my GPU only 30FPS (at 4K), I will have always 30FPS at max. So I don't know which CPU will be faster in the end, but the CPU capability remains the same and comes into play when I maybe upgrade my GPU at later time or want to drive higher frame rates at lower resolution (and get better percentile FPS / 1% Lows). Going mentally to the future, where we might have better graphics cards (at reasonable prices, cough), the CPU still will be good enough (in this specific scenario). So this 5800X3D in my linked benchmark will still deliver 180FPS in MSFS, even at 4K - if we have a GPU that can handle 180FPS @ 4K resolution by this time. It will always give you 180FPS in this scenarios, now and also in 5 or 10 years. But if your today's GPU is too "weak", you won't see these FPS numbers now, sure. I also found an English article about UserBenchmark: https://www.tomshardware.com/news/userbenchmark-benchmark-change-criticism-amd-intel,40032.html It's still not recommended to use. It's a shame it appears on the first place if you google about performance comparisons, but I can only say: Try to avoid it, it's heavily biased towards Intel (and I almost only buy Intel hardware. But being an enthusiast, I know better and like honest benchmarks that focus on real metrics. UB doesn't give us that)
  10. There's no discussion needed because the benchmarks are available already. I just wanted to help the OP, because he specifically asked about the 5800X3D and didn't know about this thread on the FSLabs forum until you linked it. The 7950X falls behind the 5800X3D. Notice the 3D at the end. It's only this processor, which has ~100MB of L3 cache. The normal 5800X (without 3D) is of course slower than the new released 7950X. But the 7950X is already slower in games than the 7700X from the same generation. This is due to how AMDs CPU design works (7950X has two chiplets, the communication between these two chiplets induces an increased latency which doesn't matter in productive workloads but games don't like it). 5950X, 7950X etc. are all great CPUs for productivity workloads, especially if you need all this multicore performance. Games on the other hand like low latency, fast clocks (especially for the main thread, which is still the limiting factor today), cache (and to a small extent fast RAM if the cache limits). P3D, MSFS and X-Plane all fall into the gaming category (eventhough some people don't like to call the simulators "games", but technically they are). Not really. I already said the 13900K is (slightly!) faster than the 5800X3D, but not much (in games! Multi-Core productive Workload, the 5800X3D can't compete with the 13900K of course). And I personally own the 13900K like I said 🙂 But for @CaptKornDog with an existing AM4 platform, the upgrade to the 5800X3D makes more sense in comparison to buy the new intel platform with DDR5. The 5800X3D costs only about 300+ bucks today and he only needs to drop it in to get almost the same performance. This is almost a no-brainer (not saying his current 5900X is "bad", it's still decent). Second Point, in case you don't know: Userbenchmark is NOT a serious website for performance comparison. There are various articles on independent computer news portals about this "benchmarking" page. They heavily favor Intel (with "questionable" scoring algorithms, for example "hugely more recent"), even if it's not the truth. Intel denied they "co-operate" with this site, so it might be some guy who just likes Intel who runs this page. The public opinion in the hardware scene is just: Don't trust userbenchmark for serious performance comparisons. Here's an article from computerbase (a well-known german hardware page) https://www.computerbase.de/2020-11/userbenchmark-umstrittenes-ranking-zen3/ You may need to run it through google translate, but I've already read about this topic on English news sites and hardware forums. I'm happy to discuss about CPU and GPU performance across flight simulation (maybe in a new topic) and Rob Ainscough can chime in if he wants, but I don't think this topic here is a reasonable place for this. I just wanted to help @CaptKornDog with his decision on this specific CPU. In the end, it's all about expectation and we simulation lovers are known to chase after FPS for decades, so there are no "miracles" one should expect. But the 5800X3D is a really decent CPU for simming and I say it from the perspective of an Intel User, using current bleeding edge technology (13900K). Edit: Here is some recent benchmark (13900K is missing, but it doesn't matter) where you see the comparison between 5800X3D, 12900K and the Ryzen 7000 series. Since no website benches P3D, I can't give you that, but I can tell you MSFS performance increases translate almost 1:1 to P3D when it comes to CPU performance. If I had a 5800X3D i'd do a P3D comparison, but I can offer benchmarking P3D on the 13900K for you (with specific defined settings).
  11. But unfortunately not the 5800X3D, which outperforms all CPUs (even the newer 7000 series) across all simulators. 😉 At the same time he is benching P3D with 4K Resolution + 4xSSAA, which results in an actual resolution of 7680*4320 (8K). This is cleary a GPU limited scenario, even with a 4090 and can't tell you anything about actual CPU performance. Maybe in 10 years we will have GPUs that can handle 8K while being CPU limited 🙂 At the moment, we can only dream about this possibility.. This is a actually a good decision. It's the cheapest way to get decent CPU performance without spending too much money. MSFS, P3D and X-Plane all benefit from the large cache. The other advantage is, that RAM performance (speed and latency) is not that important anymore. The downside of the 5800X3D is, in productive (multi-core) workloads, it falls behind your current CPU due to lower clocks. But this is only important if you actually use 12 Threads at max. performance currently, a use-case I guess only 5% of home users have. Current rumors are that the new Zen4 X3D series (ETA early next year) don't have to deal that much with downclocking. But there are no official news from AMD about these CPUs. If it's true, these CPU might be really impressive. I for myself use a 13900K with DDR5-7200CL34. Performance is amazing (I'm CPU limited at 1440p resolution), I rarely drop below 60 FPS in MSFS (Fenix, PMDG or Leonardo planes) and P3D is running around 100FPS+ with my settings. But this platform comes with a high price and is way more expensive than just dropping a 5800X3D on your current motherboard with most likely similar (or not that much slower) performance. One thing you should check is if you are currently GPU limited with your graphics settings (which can be easily achieved using high SSAA settings in P3D), then a CPU upgrade will not really help. Personally with my 3090 and not using 4K resolution, I'm not. But you have to check your GPU usage to be sure 🙂
  12. I only use v5 nowadays. I can't remember when it started exactly, but I think this "Verifying Files" thing has been going on for quite some time now. At least 1 year+ on my side. And I can also confirm I always have to enter the credentials again on ORBX Central, it rarely remembers it.
  13. I also get this message on a regular basis. I don't use P3D as often as in the past, but almost everytime I do, this message pops up and I have to fire up ORBX Central and "verify". I guess that's their "piracy protection". I haven't complained about it and just accepted it, but as always it's the paying customer that struggles with annoyances like this 😉
  14. They still work in v5. The only thing that doesn't work is the automatic scan like in P3Dv4, where it detects the airport and creates multiple views automatically. But I still can create them by hand, set the "availability radius" for example to 5nm once and I will have my views available everytime I load into that specific airport. Another method: If you had your views already set up in v4 and have those profiles, you can copy them over. I agree it was super conveniant with the automatic view creation, but I can live without it, since I adjust the views manually anyway afterwards. If there will ever be a MSFS CP, maybe they reintroduce this feature but I'd be still happy to "just" have the manual way available.
  15. Another great feature from Chaseplane I really miss: Custom Static Views, where you can save multiple views per airport. Set it up once, load / switch to it as you like easily. Basically like multiple Drone Camera Presets and for every airfield you like. There's nothing remotely comparable available in MSFS right now. Yes, I know how to handle the drone camera, I even bought an Xbox controller for it to make life easier (being more expensive than Chaseplane, but money aside, I had to wait months to get one in my country because the controllers were affected by shortages during COVID), still not better. I don't have the time to fly as much as in my past, but if I do, 80% of todays flights are in MSFS. But if occiasionally do a flight in P3D (because I want to fly heavy aircraft like the 747 or 777), I quickly realize how much is still missing from the MSFS camera system. For me, it doesn't matter really matter if these functions are built-in or available with an external software. But they're not available in the base simulator and the API still doesn't allow 3rd parties to come up with an own (optional) solution. People who used tools like Chaseplane only to set up a few cockpit views and bound it to the NumPad may not see a problem, and that's fine. But please stop "gatekeeping" people who prefer to have options by saying "I don't need it, so you won't need it too".
×
×
  • Create New...