Ranma13

I've done an analysis of RAM usage before on Reddit: Copy and pasted here for convenience: It depends on your graphics settings and, if you play multiplayer, which server you join. Using Process Explorer, we can see how much RAM DCS is using. I tested Dynamic DCS, a server with very high script usage, and JUST DOGFIGHT, a server with very low script usage. For the following graphs, Private Bytes is the amount of RAM used by just the DCS.exe process. On the High graphics preset, joining Dynamic DCS will peak memory usage at 22 GB before dropping back down to around 17 GB: On the Low graphics preset, joining Dynamic DCS will peak memory usage at 18 GB before dropping back down to around 12 GB: On the High graphics preset, joining JUST DOGFIGHT will stay at around 12 GB memory usage: On the Low graphics preset, joining JUST DOGFIGHT will stay at around 8 GB memory usage: From this, we can see that there's about an additional 4-5 GB used for High graphics preset over Low graphics preset, and that joining servers with a high number of scripts will use a ton of RAM at first before settling at a lower level, and servers with a low number of scripts will use a relatively stable amount of RAM. 32 GB will definitely give you the "I can do anything" solution, but 16 GB is fine too as long as you play on lower graphics settings and/or lower resolutions.

Generally speaking, the more VRM phases a motherboard has, the more it allows your CPU to be overclocked. This is because the VRM's not only run cooler, but they also allow lower voltages to be used for your CPU, so you can bump the speed up more before the voltages get too high. Higher voltages also cause higher heat, so lower voltages will also lead to a cooler-running CPU. However, this really only applies as you get up there with the overclocking speed. At more modest overclocking speeds, nearly any Z390 motherboard will be able to hit 4.9 to 5.1 GHz as long as the CPU itself is able to. Better/more VRM phases might allow you to get to 5.2 to 5.4 GHz with the same CPU, but unless you're joining an overclock contest, the extra 100-200 MHz won't be noticeable in typical workloads outside of benchmarks. Where more VRM phases really start to make a difference is when you start going beyond 5.4 GHz, but at that point you're also no longer using just air or water coolers and starting to enter exotic cooling territory, such as using liquid nitrogen or peltier coolers. There are of course exceptions depending on the CPU you get (some people just get really lucky and get a good chip that can overclock really well, and we refer to this as "winning the silicon lottery"), but for the vast majority of people, this will be the case. It's for this reason that I feel that for the vast majority of people, the number of VRM phases a motherboard has is mostly just a number and a marketing tactic; most people won't overclock their chips high enough to the point where VRMs start to make a significant enough difference for it to matter. The whole controversy surrounding the Asus Z390 motherboards is that the marketing materials indicated it has 8 phases, but it actually only uses 4, albeit 4 "fat" phases that are more robust than typical VRMs. Der Hirte's whole argument centers around "4 is not 8, therefore it sucks", but benchmarks have shown that although the VRMs run hotter, the overclocking ability of the Asus boards is comparable to other boards that have true 8 phases. If this is all confusing to you, just know that the number of VRM phases a motherboard has will not affect you in any meaningful way. By the time you get to the point where it will make a difference, you will have enough knowledge about it to draw your own conclusions, but for now, don't worry about it. As for the RAM speeds, the CPU will be able to use the higher RAM speed. The only thing that needs to be done is for your system builder to enable XMP (make sure you tell them to do so, it's easy to forget). XMP is a mode that allows the RAM to run at a higher speed than the official specification; it's programmed into the RAM module itself, but needs to be enabled manually on the motherboard for it to work.

Der Hirte, since it seems like you can't be arsed to have a civil discussion without resorting to personal attacks and insults (which, based on your post history, seems like your modus operandi), then I can't be arsed to pay you any attention. Welcome to my ignore list, enjoy your indefinite stay.

OP doesn't even understand basic overclocking, talking about VRM phases is going to mean absolutely nothing to him. As long as the board can push an i9 series chip to 5.1 GHz without issues, it doesn't matter if it has 2 phases or 256.

What's the button number when you press the two rocker buttons? If it's above 32, then the USB game controllers panel won't show it, and some sims can only support up to 32 buttons per controller.

In my opinion, the i9 9900K is not worth it for $530, especially just for DCS. As I've stated already, DCS only uses 2 threads max, and one is barely used as it only handles audio. For $410 ($120 cheaper), you can get an i9 9700K that can overclock just as high. It only has 8 threads instead of 16, leading to a small drop in performance for certain workflows, but it won't matter in DCS because the main game loop is not multi-threaded. For the Kraken X72, you certainly don't need a 360mm radiator unless you really want one for aesthetic reasons. A 240 or 280mm radiator will have similar or better performance than the 360mm. See the images here, which show various 240mm radiators outperforming the Kraken X72: https://www.google.com/search?q=kraken+x72+benchmark&rlz=1C1CHBF_enUS824US824&source=lnms&tbm=isch&sa=X&ved=0ahUKEwjssYzhxNHgAhWLLXwKHdXkCKMQ_AUIECgD&biw=2752&bih=1010#imgrc=8gu8N_-SWp_qMM: For the motherboard, $270 for the MSI MEG is really just too much. Unless you know that you absolutely need the extra PCI slots (and based on what you said, I'm pretty sure you don't), save your money and pick up a micro-ATX board instead like this one: https://www.newegg.com/Product/Product.aspx?Item=N82E16813145096 It's $140 ($130 cheaper), has extra PCI slots for expandability in the future, and will perform just as well as the MSI MEG. Don't get too caught up with people talking about VRM phases; that only matters at the extreme ends of overclocking, which 5.0 GHz is not. You've needlessly picked RAM with a faster timing, but is $120 more expensive for almost no noticeable speed difference. Pick up these instead, which has the same frequency, and despite slightly slower timings, you'll never notice it: https://www.newegg.com/Product/Product.aspx?Item=N82E16820232748 The Asus 2080 Ti is at the top end of the 2080 Ti GPU pricing. Unless you have a specific reason for picking this card, I recommend that you get something $100 cheaper with the same performance, like this one: https://www.newegg.com/Product/Product.aspx?Item=N82E16814932064 A 750W power supply is more than enough. I doubt the people who are saying to go with something higher have ever plugged in a watt-meter to measure their actual system usage. Having some headroom is good; having too much is just a waste of money and space, like buying a 5-gallon tank because you want to drink a cup of water. With these changes, you can save at least another $470 for almost no change in performance in DCS. For the other replies, understand that this guy barely knows anything about computer hardware. Talking about VRM phases, XMP, QVL, and virtual machines without explaining what they are is not really helping.

Having just built a new PC myself last month, I can give some insight. For Threadripper, if you don't know if you need it or not, you don't. DCS only runs on two threads, one main thread and another for audio processing, so more cores won't give you any benefit; what you really need is the highest clock speed you're willing to get. The 2950X is way overkill for just playing DCS, and you would be much better served by an i9 series chip like the i9 9700K that can hit 5.0 GHz easily, which right now seems to be the sweet spot when it comes to price vs speed. For RAM, you're going to want 32GB for DCS World. The speed doesn't really matter, just pick up whatever is within your price range. If you're on the fence, get 3200 MHz RAM. It's not that much more expensive than 2400 MHz, and there's a large selection out there. The 970 Evo is a good choice for a SSD. The EVGA 1000W PSU is way too overkill. Modern day computers don't use that many watts, and 1000W PSU's are fairly long, which can cause compatibility issues depending on which case you get. Stick with something around 600W or so. Even with everything overclocked and running straight-up benchmarks on both the CPU and GPU simultaneously, you're going to have trouble getting past 500W. A 600W PSU will give you a 100W head room in the unrealistic case that both your CPU and GPU are taxed 100%. For the CPU fan, if you don't want to get an AIO water cooler, I recommend either the Noctua NH-D15 or the Thermalright Le Grand Macho RT. They're huge heatsinks, but they're also the best of the best air coolers, matching the AIO coolers in performance. And here's where I come to the part that may sound rude, but I feel that it needs to be said. It sounds like your experience with computer hardware is very limited, and whoever is telling you that you need to buy X and Y either does not know what they're talking about, or is trying to rip you off by selling hardware that you won't even get close to the limit of. I strongly, strongly recommend that you do some research into computer hardware and read posts either here or on reddit.com/r/hoggit and see what others have asked and have recommended. Things get a bit subjective once you start to zero in on exactly what you want, but right now your requirement is that you want to get to the grocery store by riding a space shuttle.

I've run into a small hiccup. FFB commands are supposed to be loaded into a 'slot' (called an effect block index), which allows you to layer multiple effects on top of each other. For example, you can have a damper effect, spring center effect, and a periodic force all active at the same time, and each effect will take up one slot each. The problem is that vJoy doesn't change the slot number and uses the same number for every effect. This can be potentially troublesome because the FFB command can be something like "stop the effect in slot 1", and I don't know which effect is the correct one to stop. Luckily, it looks like DCS doesn't ever send the "stop the effect" command, instead choosing to set the force amount to 0. This isn't really an issue because each force has fairly unique parameters, so I can figure out which command belongs to which force simply by looking at what parameters the command has. I don't know if this behavior is the same across all sims though, which could prove to be troublesome. In the meantime, I'm looking at some alternative libraries to see if there's one that supports changing the slot number. If not, I'll go with the "just figure out which command belongs to which effect" method, but it could cause issues depending on the sim. They all have support for FFB, just at varying levels. Some are fairly detailed and have effects like wind buffeting and stronger stick pressure at higher speeds (Su-25T, TF-51D), and some are barely better than a spring-based joystick, only supporting return to center and trim (Su-27, A-10C).

According to some reviews online: https://forums.frontier.co.uk/showthread.php/341636-Tobii-4C-eye-tracker-vs-TrackIR-a-personal-review The tracking isn't as good as TrackIR, and it only supports 3DOF.

Some things I'd like to point out: 1. At 1:04, it's true that the ABRIS is not linked to the PVI-800 except in one case: when creating a nav target. If you create a nav target through the PVI-800 then select it, it will display on the ABRIS as a square with the nav target number in it. This is a great way to see the coordinates on the moving map, for example if someone feeds you the coordinates of a ground target for attack. 2. At 2:21, in addition to pressing the push buttons, right clicking on the right ABRIS wheel when you have a menu item selected will also select it. This can save time if you're already using the wheel to change the menu item. 3. At 5:09, holding the mouse and dragging on the right ABRIS wheel will scroll it faster with better control than scrolling the mouse wheel. This is useful if you're trying to move the line a long distance. 4. At 10:40, the autopilot is not overridden at over 20%. It always has 20% control authority, and will always pull the helicopter towards the last trimmed position unless disabled. This is what people usually mean when they say they're "fighting the auto-pilot" in the Ka-50; unlike the fixed wing aircraft in DCS, the autopilot does not disable once you exceed a certain threshold. 5. At 12:17, I wouldn't sweat rounding off the number in most cases. The difference between 1 second in coordinates is only 100 feet or 31 meters, which is close enough for most purposes. 6. At 13:11, it might be important to emphasize that "always N and E" only applies to the Caucasus and Persian Gulf maps. For the Nevada map, it's always N and W. 7. At 14:38, the dim light is due to the broken cockpit lighting which hopefully will be fixed sometime this year. In 1.5.x, it was very visible when lit. 8. At 14:50, it's not specifically mentioned, but he's copying the coordinates from the ABRIS to the PVI-800. You need to do this because the ABRIS is a separate system from the aircraft and does not control anything on the helicopter, whereas the PVI-800 is the actual navigation system. You can think of it as a GPS device that you put on your dash; it's not linked to your car. 9. At 15:49, to clarify, there is no backspace key if you screw up the number input. If you input a wrong number, you have to hit Reset and start the input all over again. 10. At 17:39, to be more specific, the autopilot channels consists of two functions: a stability augmentation system that smooths the pilot's inputs, and the autopilot hold, which tries to hold the attitude of the aircraft at the last trimmed position. The Flight Director button disables the autopilot hold, but keeps the SAS on. If you ever feel like you're fighting the auto-pilot, turning on the Flight Director will usually make the helicopter feel like how you'd expect it to. 11. At 20:00, that's another instance of lighting issues with the Ka-50. This also occurs with some of the other aircraft, where the anti-collision beacon shines through the cockpit as if it's transparent. 12. At 20:43, you are correct. With the altitude hold channel engaged, the collective brake will set the desired altitude when it's released. In the real aircraft, this is not something you need to think about because in order to move the collective, you need to squeeze the brake. However, in the sim, you can move the collective without squeezing the brake, so what I usually do is just tap the collective brake to set the altitude once I get the collective where I want it. 13. At 25:15, an alternative to holding down the trim button the entire time is to turn on Flight Director, get the helicopter to the attitude you want it to be in, press the trim button, then turn off Flight Director. I find this a little easier because I don't have to constantly hold a button down. 14. At 25:28, you mention "right where I want her, rudder and everything". To clarify, the rudder position is only saved as part of the trim if you enable rudder trimmer in the special options for the Ka-50. Otherwise, you will have to hold the rudder even after trimming. Without the rudder trimmer option enabled, you'll either want to constantly hold the rudder position, or get the helicopter trimmed so that you don't have to apply any rudder once trimmed.

FFB in DCS (as well as other sims like IL:2) uses the DirectInput FFB API. The API consists of a number of force types with direction, magnitude, and frequency (whichever one applies to the force being applied). These settings are sent to the joystick and stored in 'banks' so that multiple forces can be layered on top of each other. The joystick must know how to interpret these commands and convert them to actual motor movements. If you want to check out what the commands look like, you can install vJoy, then download their Feeder SDK: http://vjoystick.sourceforge.net/site/index.php/download-a-install/download In the SDK, there is an app called FfbMon under the apps folder. If you build it, the .exe will output all the FFB commands that are sent to the vJoy device. If you don't want to build it yourself, I have a pre-built version here: https://drive.google.com/open?id=1IafRADoaqu7CqcpFOjALZ9mrKv_JYXlF As an example of what the FFB commands look like, here's a screenshot from an app I'm currently working on. It's basically the same output as FfbMon, but more condensed: An easy way to test FFB without having to start up DCS every time is to download ForceTest.exe from the bottom of this page: http://www.fs-force.com/support.php It will allow you to apply a couple of force effects and see what the commands sent to the joystick are. The important thing to keep in mind is that the FFB commands aren't 'raw' commands that tell the motor what position they should be in, but are rather effect types like periodic, ramp, spring, friction, etc. You can see a list of the commands here, under section 5.3.1: https://www.kernel.org/doc/html/v4.17/input/ff.html The joystick must be able to interpret these commands and convert them to the correct motor movements.

Not all of us are in an environment where we can constantly speak voice commands, and even if we can, we may not want to. Using VoiceAttack to activate switches is a workaround, not a solution.

Brunner themselves use telemetry data too so they're a more likely partner, but more realistically what they'll do is take my work and implement it directly into CLS2Sim. As far as I understand, they already have a vJoy feeder, they just need to update it to interpret FFB commands.

One thing to keep in mind is that not all the aircraft have good support for FFB. I'm currently writing a vJoy feeder that can read the FFB messages and output them to another app for processing, so I played around with a bunch of the aircraft to see how they handle FFB. For helicopters, it seems like the only thing implemented is spring offset, and force trim and damage modifies the spring offset. However, no other forces are implemented. For fixed wing, it depends. The A-10C, F/A-18C, and Su-27 only use spring offset, and is controlled by the trim. This is better than nothing, but is very rudimentary. Other aircraft like the Su-25T and TF-51D will actually modify the spring force as airspeed changes to simulate the controls getting stiffer as airflow over the wings increases, though I can barely feel the difference with my MS FFB2. The TF-51D will also periodically send a square wave to simulate wind buffeting, and the stick will shake as you're approaching a stall. Bottom line is, don't expect the same nuanced detail in the FFB as you would in X-Plane and Prepar3D. If I can get FFB working in DCS with the CLS-E, I'd say it's still worth it, but depending on what you fly, the experience can vary from a somewhat enhanced experience to "barely better than a spring-based joystick".

It's just a micro-switch, it doesn't involve any more work than wiring up the current 3-way switch. All you have to do is replace the 3-way switch with a 2-way, then connect the remaining wire to the micro-switch: Binding it would be done exactly the same way that you'd bind the 3-way switch. Micro-switches are among the cheapest input devices you can buy, about $1.25 per switch at the retail level. Here's a whole list snap action switches that would fit the bill: https://www.mouser.com/Electromechanical/Switches/Basic-Snap-Action-Switches/_/N-5g2v?No=25 And there's already enough space on the PCB to fit one in. As for closing the cover, I have to guide it down a bit so that it doesn't produce the loud snap sound and bother the other members of the household, but as this video shows, it will consistently get stuck trying to push the switch to the lower position: Every time I close the switch cover, I need to press it down a bit more in order to ensure that the switch is pushed down. It's also not the switch cover, I have 10 other switch covers and I tested 2 of them and they all have the same issue.

No, it's not a brilliant idea. They should have put a micro-switch that gets pressed/unpressed by the cover itself. Instead, we get this kludge where if you close the cover too slowly, it doesn't have enough pressure to push the switch into the lower position.

To me it is. I'm very peculiar about how the hats feel, and I bought the throttle in large part because it's supposed to have 4-way hats. I've already mentioned two problems I've had with the new hats that I noticed within seconds of using them. I also can't return the throttle because the shipping back to Belarus will cost me over $450. I'm currently looking for replacement hats. The one on the T-50 stick seems to be super-glued on so I can't remove it to see how it'd feel, so I might have to end up shaving down the stem on the new hat so that it sits closer to the base.

I just got my Virpil throttle in yesterday and I noticed that the 2 side hats and the one under the index finger changed from 4-way hats to 5-way hats, with a new hat design: I posted about it over at Reddit: According to one commenter, it looks like the older version was being sold up until the end of January, and the new change started with the February batch. This is actually an unwelcome change for me, as I vastly prefer 4-way hats over 5-way hats. The new hat is now raised about 4mm instead of being almost flush like how it is on the T-50 grip, which makes it feel less like a tactile 4-way switch and more like a really short joystick. I also find that the additional height makes it hard for me to use the pyramid hat comfortably because the base of my thumb will rest on the new hat, and I have to take care not to accidentally move it. The hat under the index finger also feels very mushy with no tactile feedback compared to the other 2 on the side, which feel like proper tactile buttons. Currently, the only place where this change is reflected is the Rest of World store, with the EU store still showing the old renders and description for 3x 4-way hats. I really wish Virpil communicated this change out to us before stealth implementing it, as I wouldn't have purchased the throttle otherwise.

Yes, it's a 3-way momentary switch where lifting the cap will release the switch to the center position and closing it will push the switch down to the lower position. There's also an upper position that you can momentarily move the switch to once the cover is open. I wish they made it a 2-way switch instead with a micro-switch to detect when the cover is open. The 3-way switch feels clunky when closing the cover, as it now has 2 'positions': closed but not fully pushing the switch into the lower position, and closed and fully pushing the switch down. I find myself either slamming the cover shut or having to close it and then push it one more time to ensure that it pushes the switch down.

You're going to have to provide a lot more information. What devices are you trying to use with DCS BIOS? What's your experience with electronics, and what exactly are you hoping to put together? Simply saying "I want to do this thing and I don't know how, please help" is no different than saying "I want to learn how to play DCS but I don't know how, please help".

Small correction, Brunner has never explicitly stated that they're working on DirectInput FFB support, only that they're interested in it but want to focus on their primary market for now (general aviation):

Reading comprehension seems to be a major issue in this thread. I've already stated that my video was a "get it started as quick as possible without ground power" version, not a tutorial on how to do it properly. I don't care that I powered the fire extinguishing system after I started the APU because the APU will never, ever catch fire from a cold start in DCS. Not even the official LWin+Home auto-start sequence will enable the fire extinguishing system before starting the APU; it's only turned on halfway through the left engine startup. For nearly every switch and lever in the startup sequence, the order doesn't matter. Using a macro, you can flip every single switch and lever to its correct position in the same instant with no adverse effects. My startup procedure takes advantage of the fact that many things aren't modeled or don't matter, to get airborne ASAP. We should teach the proper startup procedure so that people know what it is and can adapt their workflow accordingly once they're familiar with it. It's the same reason why we teach people to drive the speed limit despite everyone going 10 over, or keep your hands at 3 and 9 despite many people driving one-handed, or fully stopping at a stop sign despite many people doing a rolling stop, or giving a turn signal for at least 100 feet before initiating the turn despite many people doing it immediately. You establish a baseline first, and once that's understood, then you give leeway to break those baselines. This is why the Ka-50 Leading Edge startup video is so good; it goes over the proper startup procedure from beginning to end and explains not only what you're doing but also why you should do it. Once you understand what the proper procedure is, you can then start cutting out things you know don't matter, like the built-in tests and switch flipping order.

The video isn't for you, nor is it a training video showing the correct procedure for starting the aircraft. It's an unlisted video intended solely to dispute Rogue Trooper's claim that he can start up the aircraft quicker with ground power than without. It also happens to be the way I usually perform my startup because though I know the correct procedure, I'm not interested in performing it for the umpteenth time. Saying that I lost credibility because I didn't do it the 'proper' way is like me saying that you don't know how to drive because you did a rolling stop instead of stopping for the proper 2 seconds. There's the proper way and the expedient way, but when we teach, we always teach the proper way.

Suit yourself: Instant action mission Cold Start in Mozdok, airborne at 2:20, you're dead at 2:27.

I specifically mentioned live maps, tutorials, and airport charts, amongst other things. You can't put a live map on the kneeboard, nor view a tutorial website. You can't load up a 200-page PDF and do a search on it, and you can't take notes. You can't zoom into the small details on an airport chart, and you can't refer to a physical book. Reference materials come in all shapes and sizes; it's not limited to just Chuck's guide. The kneeboard is not a great solution, and it's worse in VR due to the lower resolution. The other points are my experience and I've stated them as such. If you think that it's still "flat out wrong", then I stand by my statement: you misunderstand what "viewpoint" means.