Hiob

Thanks, that's the quadrillions iteration though. Practice makes perfect....... There is still much more about FreeCAD that I can't do than I can do. E.g. I don't have the slightest idea how to assemble different object together. But I'm confident that everything that I need to do can be done 'somehow' in FreeCAD. And anything open source is a force for good in my opinion, so I give them some slack for an unpolished UI and some bugs....

These of the designs I made for the Crosswinds FFB mod:

good advice SSLR and SSAO are gfx options that in most scenarios are barely noticeable unless you specificaly look out for them. I would add, since you mentioned that you have generally 180 fps and then dips, try limiting the fps to 120 upfront. There is diminishing returns past that (it's not a fps shooter after all) and it may open up some breathing room for your GPU and make dips less noticeable. And if possible don't use the ingame limiter, but either Afterburner/Riva or the GPU driver.

Nur so am Rande. Die Cockpits in MSFS sind auch "clickable". Was hält Dich davon ab, beides zu fliegen? Wird hier nicht wenige geben, die beides regelmäßig benutzen (mich zum Beispiel). Ebenso gibt es viele, die in DCS nicht (nur) Waffen einsetzen, sondern zum Beispiel auch Search and Rescue-Missionen mit Hubschraubern (gibt nen ganzen Server dafür) oder einfach Navigation fliegen. Mache ich auch oft.

Correct. That's exactly what I've mentioned earlier as well. Either way the MoBo will only choose one XMP profile for all sticks and if they are as close together as the ones OP listed, I would be very surprised if any problems would show. Also, using XMP profiles is technically overclocking. It is not guarantieed to work. To be clear. I didn't mean to advocate for mixing wildly different Sticks. I would always stick at least to the same brand, dies, clock and main timings. Just in the specific match OP suggested I personally would confidently risk it.

"Any" is the wrong term here. As I said - memory timings and subtimings are controlled by the MoBo. You can always loosen the timings and the RAM stick will be happy. Problems start to rise when you try to force the sharper timings on the inferior RAM sticks. Regarding two vs. four sticks. Four sticks do put more stress on the memory controller. For regular use (and users) that isn't a concern. For all normal uses including XMP-Profiles it'll work just fine either way. It the edge cases (as always) like extreme (memory) overclocking where it will make a difference in stability. I don't think this is a concern of the OP though. If somebody really wnat to go down that particular rabbit hole, I recommend this youtube channel. This guy is a wizard: https://www.youtube.com/@ActuallyHardcoreOverclocking

If highest performance in DCS is the main goal, I would stay with the X3D series of AMD if possible.

I would think so. And personally I wouldn't hesitate, but of course I'm no authority and you take the risk yourself.

When you can't find the exact same specs you're playing it save when you dial in the less tight timings. Then it should work without issues. Depending on the sophistication of your MoBo it will probably choose the "correct* settings on it's own. Just to be clear, you can (generally) mix and mismatch RAM. It will work, most likely without noticeable issues, it just isn't ideal in regards to performance. A slight mismatch in XMP timings in negligible, since the actual timings are controlled by the Mobo anyways.

I'm pretty confident that you won't notice any difference between 3000 or 3200 or two vs. four sticks. Just go the easiest and probably cheapest way and buy another set of you're existing 2x16 to make it 4x16. Just make sure it is the same RAM with the same timings.

You're right. What I had in mind was the stability and potential for overclocking. But that doesn't apply here.

The easiest, least headacke option probably is to simply get another pair of the exact same sticks you already have. That's sure. If the small improvement of 2x32GB 3200 is worth it, pretty much depends on the price. But realisticly it won't be most likely.

200 Megatransfers are probably not noticable in practice. Double so since four sticks are slower than two (also barely noticable but still). What are your system specs in general? Intel, Ryzen, Chipset? The sweetspot for a Ryzen 5000 system for examble would be 2x32GB 3600 MTs

Do you no how to import lines from existing bodies into a sketch?

Which workbench are you using in the first place? Normally you just highlight (click) the face you want to sketch on and select new sketch. Usually the best starting point is the parametric part workbench.

Nope sorry, turned it off and never looked back.

Apart from that - how do you get along with the 1.0? I'm starting to get more fluid with it - learned some lessons on my part (don't be lazy on naming your steps) as well that improved my workflow. Thanks for the link!

Did you receive any updates? Windows, Chipset, GPU-Drivers? Especially with updates to the game and gpu-drivers, you need to purge the shader cache. It will then stutter some on first loads while the shaders are recompiled, but should be better grom the second time.......

That doesn't help with stuttering unfortunately. If you want to eliminate the power plans as possible culprit, you need to disable the service entirely. Just open the services in windows, find the power service (or the translation for your region) and disable it. Don't worry, you can't break anything and you can turn it back on as easily the same way. Personally I have very good experiences with doing that, so I strongly encourage it. (A little bit of background on why that helps. The power service makes A LOT of interrupt requests to examine the powerstate and load on various system components. Every IRQ is a possible stutter. Unfortunately it doesn't stop doing the interrupts even on high performance settings.)

Unfortunately it seems that there is no "one fits all" solution, when it comes to such issues. There are so many variables at play that all need to work seemlessly together to have a really smooth experience. There are some things that I have personally put in the "pretty sure it is bad for performance" drawer - but even those are sometimes opposed by others. I would say it is reasonable to just build a toolbox of possible culprtis and solutions and then just apply it in the personal preferred order when necessary. Hyperthreading/SMT on/off is just one of those tools. Not the most important (imho) but one of the easiest to test. Therefore it is pretty high up on my personal checklist.

I think that‘s merely one tool in the arsenal of thing to try if you experience problems like stuttering. It probably depends on the make and model of your specific CPU and other system parameters if it is helpful or not. There where times when I had SMT (the AMD term for HT) disabled and had noticeable improvement on my 12 core Ryzen 5900. It wasn’t a problem, because I didn‘t use it for other heavy tasks except DCS. Currently I have it enabled (I think) but still don’t have problems.

Hyperthreading and Multithreading are different things. You can’t drop one in exchange for another. Multithreading is the workload of an application split into different distinct tasks (threads) that can be executed in parallel. Hyperthreading means that the CPU can execute two threads (simultaneously) on one core by splitting it into one physical and one logical. For the scheduler it looks like two cores. In other words -multithreading requires changes on an application level, hyperthreading is enabled by the hardware and microcode and managed by the scheduler. In short, it won’t affect DCS at all. The application doesn’t care if its different threads (multithreading) are executed on physical cores (no HT) or physical and logical cores (HT). If anything a cpu with physical cores is more performant than a cpu that achieves the same amount of „cores“ by means of doubling them with HT

Only if you 100% positively need them for anything. Geforce Experience (Legacy), Aorus/Gigabyte, Logitech G-Hub, Corsair iCue..... all that crap that runs in the background can be a potential source for latency, frametime spikes and overall problems. Emphasis on "can". If possible, I would keep my system as light and free of bloat as possible.

I did. And In still think it's voodoo. Personally I get the best results with everything set to 120 Hz. But I don't want to start a fight over it. I was just interested in the reasoning.

A little OT, but why? I know it's repeated over and over again, but it makes no sense whatsoever. When your display can drive 120 Hz why would it be of any use (apart from lower GPU usage) to stay below that? Or better phrased - not to match that.