Voyager

It's been about a year since I last did it, and it was actually for a different game, but there is a way to set up Performance Monitor such that is shows the specific process threads in terms of CPU usage, rather than the CPU's in terms of thread usage. This is very illuminating since Windows tends to cycle heavy threads between the cores, to spread the thermal load out, but that makes it look like each core is sharing the processing load, when it could easily be one or two really heavy threads. I want to say this was what I was starting from, but it would take some fiddling to get it laid flat: https://www.windowscentral.com/how-use-performance-monitor-windows-10

I believe it's past rumor now. The official released images show two eight pin power connectors on it. Also Anandtech in their 30 series article pointed out that we should expect to see performance per watt stag mating in most electronics products: https://www.anandtech.com/show/16057/nvidia-announces-the-geforce-rtx-30-series-ampere-for-gaming-starting-with-rtx-3080-rtx-3090 Apparently the physics that made smaller parts run cooler doesn't continue at the node sizes we are operating at,so we can expect things to keep getting hotter over time. Note: RDNA2 has a 50% performance per watt over RDNA1 due to logic and design improvements, not a node shrink. I think that's largely because Vega was just such a power hungry architecture that it left a lot on the table, even after RDNA1 improved over it. Kind of the way Zen 1 being a huge leap over Bulldozer didn't necessarily make it a great processor.

Agreed on waiting. Flight sims seem to all be very mhz dependent. I've got a 3800x and it does well enough, and was a huge step up from my 4770K, but the Sky Lake chips at 5Ghz+ are just faster than the Zen 2 ones right now. I'm hearing good rumours about the Zen 3, and with FS2020 becoming a regular benchmark, I think we should get a solid idea of how it would compare to the Intel parts at launch. Thank you, Harry Voyager

Only when not CPU bound. Understand, most AAA games are built around the last generation of consoles: i.e. less CPU than an FX-4100. I imagine DCS would not be happy with one of those, even in the most optimized state.

So I found out what was going on with the 30 series CUDA Cores. According to the reddit thread here: https://www.nvidia.com/en-us/geforce/news/rtx-30-series-community-qa/ nVidia changed the architecture of the CUDA Cores. If I understand it, each core was previously able to do an INT32 and an FP32 operation, but they changed it so that, if the game calls for it, the core can optionally instead do two FP32 operations, effectively doubling the FP32 performance, at the expense of Integer performance. So this becomes a question for the dev team: Is the DCS graphics engine more dependent on the Floating Point operations, or does it still require more INT32 operations? If this is a FP32 demanding engine, and doesn't use much INT32 operations at all, we could see a true 2x lift, but, if DCS is much more dependent on the INT32 operations, we may see little or no lift at all.

Also, don't forget, the 980 TI had competition from the Radeon cards at the time, and the 1080 TI released when everyone was expecting Vega to be groundbreaking. nVidia was only able to charge what they were for the 2080 Ti because AMD didn't have anything that could seriously compete with the 1070, much less the 1080 for years. Even with the RDNA1 cards, AMD has not, yet, released anything that meets or beats the 1080 Ti.

That's sort of where I'm heading, though I'm on the back end of the 1080 TI cycle, so would like to get upgraded soon. On the one hand, the numbers nVidia is showing are absolutely impressive, on the other hand, I don't see how they could have doubled the cuda core count for only a 50% increase in transistor count. If the 3090 is really double the performance of the 2080 Ti, I'm ok spending $1500 on it launch day, but I want to see the 3080 benchmarks first.

When I did some comparisons of the 1080 TI and 2080 Ti I did find that most of the performance difference was in Dx12, with Dx11 having a much lower performance delta. Given that DCS and other flight sims are mostly Dx11 for now, I suspect we may see a similar reduced difference in performance. However, nVidia does seem to have done something interesting with its CUDA cores this time and seems to have set them up in a way that effectively doubles the core density, though with some limitations. Also, DCS is eventually moving to Vulkan, so the Dx11 limitation may cease to be one. I'd be extremely cautious about doing DLSS in a combat flight sim, given just how dependent we are on fine detail for spotting. The YouTube algorithm is fantastic as low loss compression, but it absolutely destroyes targets. I'd be very concerned that DLSS would do the same, and I'd need to see some form of formalized spotting impact testing before I'd even suggest the devs spend the time to incorporate it. Harry Voyager

That is clever. How did you do that? I'm probably going to steal that myself...

Ah, thank you very much. Now I just need to actually do some cable routing for my nest :)

Dumb question time, should one expect a verification email that their order was placed, or just that the invoice is ready? I think I sent it all out correctly, but just wanted to know if it actually got through intact, or if I should be looking for a responce that vanished into the aether... Thank you, Harry Voyager

Very cool :) Having been on the vendor side of the equation I know how it's always a nest of design trades :) And with Star Wars: Squadrons coming out in October (and it reportedly supporting VR with mouse) I'll have to make some trades of my own :)

Given that USB-C seems to be more robust than the micro USB form factor, is it possible to use the USB-C connector, or is it to late in the prototyping process for that to be economical? (I've dealt with supply chains before, so I know how long it can take to get one to change direction...)

I think it's a general bug. It didn't log any of my F-14 time either.

I was thinking about that one too, when I just remembered the pilot manuals specifically caution about raising the flaps to quickly after takeoff, and that, when flying with flaps deployed, there was a risk of one flap or the other suddenly retracting. When retracting the flaps in flight, at least one of them directed pilots to 'nurse' the flap lever up to keep them from retracting too quickly and spilling the air under the wings. I'll have to see if I can find it. Harry Voyager

And because it's cooling the mixture, for a given pressure it will have a higher fuel/aor density. So in theory, if we control for the inlet temperature, we should get the same power for a given MP, with or without water, but since that was a test of total power, one would expect that they used the same intercooler flow for both tests. So that begs the question, where is the Carb Air temp gauge measuring from? Before or after the WI nozzle? Given that the manuals allow up to an additional 17C in carb inlet temperature with WI activated, I'd have to guess that it is before the water injection, so we probably don't have a measure of the true charge temperature in the plane. Given that, I suspect we would need to do specific performance at altitude tests to determine if we are or are not getting the equivalent of WI performance at this time. Thank you, Voyager

Last I tested, we're able to get 64" at 2700RPM, so we should have full power, regardless of the implementation the WI system. WI allows the engine to be run at a higher pressure setting without ejecting large chunks of cylinders, but it does not, in and of itself, increase power. What you are more likely running into is that the P-51D and Fw-190D are still extremely capable up to 20,000-25,000ft. The advantage of a turbocharger is that it is a more efficient way to extract energy from the engine, than the gear driven supercharge is, but that difference in overall power really does not start to show up until well after 20-25,000 ft, so at 25,000ft, you are still fighting them on roughly equal terms. However, by the time of the P-47D-30, from 25,000 to 30,000ft, the turbo really is not big enough, anymore, to take advantage of the extra power available, so from 25,000ft to 30,000ft, so you're really performing at the level of a late 1943/early 1944 aircraft. It's more a measure of how exceptionally good its high altitude performance was in late 1943, that that level of performance is still competitive with a 1945 aircraft. That's the key piece of context to remember here. The P-47 developed its reputation as a high altitude fighter in 1943 and early 1944, when its primary competitors were things like the Spitfire Mk 9, which topped out at slightly above 415mph at 22,000ft, the Bf-109G-6 (without water injection) which didn't even break 400 mph at 22kft, the Fw-190A's, which generally peaked at ~400 at 20,000ft, or the Hawker Typhoon, which generally didn't even break 20,000ft. By contrast, the P-47C/D models were chugging along at 420-430 mph at 30,000ft. 20-30 mph faster, at 10,000ft higher than pretty much everyone else on the field. And this is where the B-17's are hanging out at the time too. If you looks at the history of the Merlin powered P-51, and the Fw-190D, they are both direct responses to the poor altitude performance of their predecessor designs, and were largely successful at it. By contrast, the P-47's high altitude performance was never and issue; it's challenge was always low altitude performance, rate of climb and range. And when you look at the key design changes from the C-2 to the D-30, that is what is reflected: an enlarge main fuel tank for extended range, new larger paddle-bladed propellers to increase the low speed thrust, higher boost limits for the engine allowing more power below the critical altitude. But we really don't see things that increase, or even really change, the high altitude performance. As near as I can tell, the turbo charger did not get more powerful until the P-47M model, and in all of the earlier models, once its past the systems' critical altitude, the power at altitude appears to be nearly the same as the 1943 D-5 version. So to recap, in the 20-25kft range, you're only just starting to see an advantage from having a turbo instead of a super charger, but its not a large or decisive gap. Further, the P-51D, Fw-190D, and Bf-109K all have significantly improved higher altitude performance over the earlier war aircraft, while the P-47, at high altitude, is still pretty much performing where it was in 1943, so while you should see better performance, it won't be to the margin that earned the plane its original reputation.

It would have been and it would have been one of the lighter D-5 razor back versions, but probably would have been before they'd implemented the paddle bladed propellers. I'll also note that Johnson was know for having a surprising amount of upper body strength (P-47 roll rate at speed becomes force limited rather than stops limited) and he said he was using the rudder heavily on the rolling scissors part of the play.

My understanding is that the connection was added to simplify power manipulation in combat. If you need to increase power you need to go to Auto rich, the pancreas prop to 2700rpm then add manifold pressure, then add WI in that order, lest bad things happen. The WI timing was also tricky because if you added it at too low a power setting, you risked drowning the engine. (You're dumping water into the air intake. You want it so the stuff doesn't go bang before it should, but if there's not enough heat that its all vapor by the time it gets to the part that goes bang, it might not go bang at all, which would all be bad). There is also the complex feedback loop going on between the turbo boots levels, and the engine power levels that adds a lot of turbo lag into things. (More boost meant more power which meant more boost) Ultimately pilots found if they needed to suddenly apply power, it was often easier to just grab both levers and run both forward at the same time rather than dealing with them separately. This sort of "quick and dirty" safety is also likely why later versions added in a catch to the prop that would drive the prop level forward if you tried to advance the throttle beyond it. If you'd linked the boost and throttle, and got jumped, you could just slam the throttle forward and hit the WI, and the catches would help keep you from breaking the engine. Experts could still run everything individually for maximum performance, but if you were fresh out of flight school and just deploying into combat for the first time, keeping things linked was good enough and could save your life.

Not sure which forums this really goes best in, but I'm wondering if there is a way to display the various icons, simple aircraft state read outs, etc to the secondary video output stream, but leave them off in the primary output? As in, make them visible on the 2D desktop mirror of the VR projection while not making then visible in the VR itself? I've been kicking about the idea of streaming and one of the things I've noticed in watching other's videos is, when they don't have icons turned on, the stream algorithms tend to make things vanish. Since, to best stream VR, it seems best to stream the desktop mirror rather than the game proper, if there's a way to add in the icons only to the desktop overlay, that may be a way to ameliorate that issue for the viewer, without having i have icons on for myself. Thoughts? Thank you, Harry Voyager

There appears to be a known issue with the oxygen system that's causing it to run out after about 30m of high altitude flight. Might be what you are running into: https://forums.eagle.ru/showthread.php?t=276620

Honestly, they need to get the thermal model nailed down before they start implementing engine damage due to overheating. Last flight I tested you basically need full open cowl flaps to keep the engine within limits at 200 mph IAS at 2550/41". Given that you need to close the cowl flaps above 225 IAS, that implies an issue with the cooling model. Basically if the thermodynamics cook the engine before you get off the ground, we're going to get a lot less out of this that we would with an invulnerable engine.

Did the propeller stop spinning? As long as the prop is turning it's still acting as an air pump, so I'd expect you'd still see pressure at the exhaust, albeit I'd expect not as much as when it's firing. The question's going to be how to convert the back-pressure into rotation resistance, and it would likely need a unique feedback loop for it. That said, when I last got my engine on the wrong tank, I wasn't paying attention to the turbine speed, so would be an interesting test to see how long the turbine spins, and how it reacts.

I seem to recall it was the European theater in particular that ended up with 130 octane fuel early,and I want to say it was mostly that it was something the British had developed(?) but I have not even begun to dig into the logistics of how all that happened and who had what when. I would be very surprised if the Soviet P-47's saw any 130 gas on the Eastern Front. All I do know is the planes that were naked at a 58" redline were marked before the aircraft was re-rated for the 130 octane gas. I also had the impression that the 150 octane fuel didn't buy the plane much over the 130 octane gas, until the 'M' and 'N' versions with the larger turbo charger were fielded. You got more speed at lower altitudes because you could run higher boost, but once it got past the critical altitude ended up being largely the same, just due to the available manifold pressure. Granted, since most online combat takes place at low level, I can see that being a significant factor in online play, though.

I'm not familiar with conversion charts in this context? Are we talking about weight to fuel, or percentage full to fuel load? I'll have to check the manuals for actual fuel in tank. The pilot handbooks mostly just note that the main tank indicator isn't that accurate, which is why you keep a 15 gallon reserve in the aux tank. There is an actual expectation that you'll run tanks dry in WWII aircraft. The indicators were not necessarily great and piston engines, as long as the crank is still turning, should not have any issue resuming power after losing fuel flow. Jets and turbines are a rather different kettle of fish. Right now the main tank appears to only hold 200 gallons, with the rest going into the aux tank. The 200 gallon main tank would be the right size for the razorback models, but the bubbletops, as far as I know, all had the 270 gallon main tank. All of them had the 100 gallon auxiliary tank. Yes, PSA means Public Service Announcement. I'd been seeing a lot of instability threads, and found if you load up 100% fuel and follow the takeoff procedure it is rather early to forget to turn over to the aux tank and end up flying around very tail heavy.