Nereid

Why do you think it was added already? It wasn't so it is a bug. Nice to hear. Thanks. I will give it a try as soon as possible.

This is called a request. A missing keybinding is a bug and not a request. Here's your dictionary https://dictionary.cambridge.org/dictionary/english/request So I'm the one who has to facepalm and I'm missing a few hands to do it. But thanks for your silly reply.

If mirrors are on I'm experiencing some flickering on the whole screen every then and now. But only if the mirrors are toggled on and only on the Sinai map. Version 2.8.6.41363, ST (I have noticed it in MT too) It is similar to the issue descibed in but it happens just while looking forward without moving the view and the the whole screen blacks out for just 1/10 of a second or less. (well in some cases it looks like there a black box just on most of the screen and expands shortly after; and it is clearly not a displayport blackout because of this and the short time) I have also tried the solutions described in but they do not fix the problem. This problem (aside from not being mirror related) seems to fit my problem better. Other modules (I have tested the F/A-18) are not affected on the Sinai map and there is no issue for F-14 on the Syria map. It may be a map issue that only shows in the F-14. But because only the F-14 is affected, I'm reporting it here. The Sinai map is quite new, so maybe no F-14 owner has experienced it so far.

Yes, but it is clearly a Tomcat related issue and it is back in 2.8.6.41363 for the Sinai map (at least for me).

There is no keybinding for the altimeter pressure knob in the pilot seat. Version: 2.8.41363, ST (single threaded version)

Is there no keybinding for the Altimeter Pressure Setting Knob on the pilot seat? Or am I simply overlooking it? If there is none, I would call this a bug.

The last time I flew the Mi-24, it was tougher to fly.

I have just tried the AH-64D again after a few weeks of absence from DCS and with the new update. And for me it seems easier to adapt than before the patch. Still not easy to fly, but that is to expect. It's a helicopter, so it shouldn't be easy. Without any practise for three weeks or more, I was quite surprised that I could takeoff and land this beast without too many issues. Takeoff and landing were far from perfect, but that was to expect too. There were some strange behaviours as I was trying to land on a flat roof of a building, but hey I was probably a little cocky even to try this.

I was not talking about any Ryzen. But the long loading times are gone.

Well. It's the other way around. There is an uncage button (called "cage button" but it uncages the instrument) but none for caging. Indeed. A toggle would be nice. But at least they should add a button for caging. But a toggle would spare a button. And some HOTAS systems have encoder dials with a push. So the encoder dial could adjust and the push could cage/uncage which makes it easier to memorize.

Yes. Something like this. But in the AH-64D it seems more hard to compensate for it as in other choppers as far as I remember (and to be clear: I did not mean the countering by applying left cyclic; I meant the sudden drift to the left when airborne). But this doesn't mean it is a bad thing and it should be changed. It just that I need a few attempts to get used to it everytime (!) I try to have some fun with the Apache. And as Admiki said: We are not feeling it in a sim, so we have to rely on visuals or routine. But currently I do like the FM on the AH-64D. On its release in EA and even several months later.... well, not so much. Either it's me or something has been improved a lot.

Indeed. That's why a was imprecise about my feelings about the changes over the last few patches. I think I saw a huge difference between my experiences from last year and my experiences now. But I can not really tell what exactly was different but it felt easier and without surprises in behavior in flight. And I did a quick test with the new update this week. For me it looks more or less the same as last week. But it was a quick test only - ten to twenty minutes, not more and without attitude or altitude hold. And as admiki said: those are not like an autopilot in a jet. The only thing that irritates me (but it did this since the AH-64D was released), is that in a hover takeoff the bird tends to fall to the right - until it is in the air. Then any compensation for this will suddenly drift you to left. But maybe there is an explanation for this. But every day I try the AH-64D I need a little practise to get used to it.

Because of time constraints and real life I could only test this once. But it seems that it loads much quicker now even without fiddling with affinities. The progress bar behaves differently than in ST but it changes in the first few seconds (it never did before), then it stops and after 10 seconds or so it suddenly moves to the end and the loading time was quite normal. But I didn't use a stopwatch to verify this and maybe I'm wrong. I have never disabled HT on my machine.

Did you look at it with the process manager? Because the number of threads that use CPU time cannot be deduced from the utilization of the cores alone. Even a single core can run 10 or more threads (even without loss in performance and even with some GAIN in performance in theorie). The task manager in windows alone will not help you here. BTW: DCS uses a lot more than 10 threads - even in ST. But most of them are just idle.

For me it is stll better than non MT. But there seems to be a difference. In my tests with the AH-64D there is really no difference between non MT and a start with disabled cores. But even with all cores performance is not that much better. But the results are indeed strange. There is a difference for the F-18 though. Problem is: Loading times of 3 to 5 minutes are more than just a nuisance. PS: Maybe they calculate the number of used threads on the number cores and with to many cores there is just one thread "too many" so that a deadlock or resource contention is caused? Debugging such bugs is often hard. But I haven't seen any differences in the threads used with the process manager. Check your fps after doing it please and make a comparison with the non MT version.

I don't think so. I was struggling with some weird behaviors that were hard to get used to for me. They are hard to explain but most often I lost a bit to much control when I was hard turning or slowing down. Sometimes the tail rotor suddenly seemed to do not any affect at all in one direction. But there were other minor effects, that were somewhat counterintuitive. I do not see them anymore.

I have to add that even disabling P-cores seems to fix it. There is a rumor that disabling cores disable MT, but a closer look with the process manager and some performance checks do not confirm this. And if you have to "disable" a core then anybody would go for the E-cores.

I did some tests regarding performance: NMT = Non MT version of DCS MT-8E = MT version of DCS without 8 E-cores MT-4E = MT version of DCS without 4 E-cores MT-2E = MT version of DCS without 2 E-cores MT = MT version of DCS with all E-cores (affinity unchanged) AH-64D, Caucasus, Cold Start, external view only NMT: 80 fps, 25 seconds to load MT-2E: 80 fps, 41 seconds to load MT-4E: 80 fps, 25 seconds to load MT-8E: 80 fps, 25 seconds to loa MT: 85 fps, up to 5 minutes to load So no real performance gain or loss. And in one case I got 90 fps for MT-8E. F/A-18C, Syria, free flight, no flight input (just let the bird fly on its own, slowly descending for about 15 seconds down to the small town/buildings) NMT: 80 fps down to 65 fps, about 30 seconds to load MT-4E: 95 fps down to 80 fps, about 30 seconds to load MT-8E: 95 fps down to 80 fps, about 30 seconds to load MT: 95 to 100 fps down to 80 or 82 fps, about 3 minutes to load Disabling the 8 E-cores seems to have a small negative impact on performance but it is still better than NMT. And the impact is hard to notice and maybe it was just some kind of confirmation bias or pure coincidence. All Affinity changes were made by Process Lasso and no Alt-Tabbing was done.

Why should this be the case? If the code uses threads it will use threads even when the affinity was changed. The OS will just manage the threads differently. Just verified it with the processmanager: After fiddling with the affinities the threads and their cpu usage are the same. More than 10 threads are running using cpu. In the Non-MT version there are just 4 to 5 threads really running. So changing the affinity doesn't change anything. It wouldn't make much sense anyway [and software developers who write code that switches back to single threading when changing affinity should be fired without a second thought...].

You will have to change this each time. But there is a program called "Process Lasso" that does this automatially. Unfortunatelly it doesn't work reliable for me and doesn't set the affinity in some cases. But it shows what cores are the E-cores at least. I have tried this over 10 times so far and every time the loading times were quite normal with the workaround. And without it the loading times are really a pain in the...

I know. But Win-10 works quite well for me. To some extent I stick to the "never change a running system" idea. Anyway. I'm 99% percent sure the E-cores are causing this. 3 tests now and all with loading times around 20 seconds. Before it was ALLWAYS around 90 or more seconds (and sometimes it got completely stuck) Of course it could something like a race condition because of too many cores too. But this is very unlikely in my opinion and I'm to lazy now to check if disabling any other 8 cores will fix it too.

Indeed. And maybe it is caused, because I'm still using Win-10. Non computer nerds will have a hard time to use this workaround. But because we are playing DCS most of us are some kind of such nerds. Edit: checked it twice now and it worked both times.

I think I found a solution for this. The workaround that works for me: Start DCS Open the Task-Manager Go to the details tab and find the process DCS.exe Rightclick the process and select "Set Affinity" Uncheck all cores that relates to E-cores for your CPU Point 5 is the tricky one. I'm using a 13700K with 8 P-cores and 8 E-cores. The E-Cores are the last 8 in the list (as far as I know). After unchecking the last 8 cores, the loading times are even faster or at the same level than non multithreaded.

I tried a few things in the last hour and RAM is not the issue here. But I think I found the solution. because I'm using the 13700K there are 8 P-cores and 8 E-cores. So I tried to disable the E-cores for DCS by setting the core affinity in the task manager to the first 16 cores (8 real cores and 8 HT- cores of the real P-cores). And.... well... the issue is gone. So it seems that using the E-cores causes the problem.

For me it's the opposite. Load times are much slower (about 5 times slower) in the multithreaded version.