G MIDY

-30 seems great and good enough if you're stable. You could push it further but serious instability seems to really kick in at around this point. A good thing to do is to keep an eye on the Event Logs > Windows Logs > System for any WHEA-logger errors as these always indicate CPU instability. The same logic of undervolting can also be applied to your GPU. At stock settings Nvidia GPU's apply massive voltages to ensure that the GPU is always 100% stable. This gives far higher voltages and temperatures than is necessary. At these higher voltages the GPU will eventually hit a power limit where the GPU is limited by the power coming from the PSU. At stock settings you actually lose potential performance because the GPU can't remain at its clock speed because it needs to constantly reduce voltage to account for it hitting this power limit. Undervolting the GPU can really help performance although it does take quite a lot of testing. There is a sweetspot where you can even achieve a greatly increased overclocked GPU boost speed at a lower voltage providing stability is maintained. You'd be surprised how far away the stock settings are from this spot! This is worth a look if you have time and interest as many people will be running their GPU at higher temperatures and voltages for no reason.

The 7800X3D relies upon a curve which decides the current clock speed against the voltage needed to sustain that clock speed. If the CPU is worked hard then the CPU will try to boost to the maximum boost clock speed but this also demands the maximum voltage. Higher voltages mean higher temperatures and the CPU may eventually begin to downclock to control temperatures which means less performance. A negative PBO offset is basically an undervolt of the voltage element of the curve where the CPU will instead try to boost to the maximum clock speed using less voltage. Because less voltage means less temperature, some level of increased performance is possible because the CPU is less likely to downclock due to lower temperatures. So what is the limitation? Stability. As voltage is decreased CPU's eventually become unstable. This however really depends upon the silicon lottery as every CPU is different. I run around -25 negative PBO offset which seems to be my sweetspot, if I go beyond this then things start to go wrong and the level of instability begins to outweigh any gains. I've found that the performance gains are variable, you'll barely notice them in MSFS but they can be really noticeable in benchmarks. It depends upon your intended use of the CPU but for the most part you aren't going to notice any difference in gaming vs stock. However, this doesn't mean that it shouldn't be done. A negative PBO offset can really bring down overall temperatures and power usage so it's a good option even if the performance gains aren't really worth the effort. If anyone does intend to get a 7800X3D then the cooling solution is by far the most important factor for the performance of the CPU alone. This isn't necessarily because the CPU is a hot CPU, it's more because reaching that maximum boost clock often is almost completely dependent on temperature. A negative PBO offset can really help but it won't beat a quality cooling solution. To put it mildly on the 7800X3D, I won't be going back to Intel for a very long time. Whatever you do don't compare this CPU to what an Intel can achieve in synthetic benchmarks, the Intel will always win but with gaming and MSFS specifically the 7800X3D is king. 🙃

I hope they have fixed the horrible bug where the assistance options wouldn't save... I downgraded out of the beta because of that.

One thing to remember is that to produce an aircraft accurately you need source data such as technical publications and technical data. The A320 is a digital platform developed in an analogue age where much of the information over time has been made readily available both physically and digitally. The A320 is also an extremely common aircraft and so the data is far more widely available, you can potentially source it from any number of willing (ex)operators. Any aircraft built prior to 1990 will be difficult to produce because the publications are always hard-copy based in which case you need to buy it, that is extremely expensive. Out of service aircraft are even more difficult because the real aircraft often do not exist anymore and the remaining publications were probably sold or destroyed. A good example is sounds, if you wanted to produce a B727 today then how would you replicate the sound of the switches, clicks, engine sounds? You can't really do that without access to the real aircraft however there are barely any B727's flying now compared to 20 years ago. It may be impossible to do it accurately and that's where developers may have to blur the lines. I can understand why developers are likely to focus on the likes of the A320 or B737, there is plentiful cheap information and the real aircraft are still flying. The newer aircraft such as the A350 are a little more difficult because that information is still new and nowadays it's located in proprietary electronic tools. I can buy an A320 flight manual or even a complete maintenance manual on ebay, however try that for the A350. There are many people and companies who produce this information, for example as part of a training package offered to an airline but there's zero chance that they'll offer it to a sim developer for free.

I love aircraft like this. Dragging 5-7t on the tail, less rudder authority, limited automation. Certainly not dangerous, but definitely unforgiving. It will certainly test those who rely on the aircraft stopping itself from killing you!

Aamir, I think it's sometimes pointless engaging on these sort of discussions because ultimately anger in the sim community is soon forgotten and the community generally is fickle. Nobody will not buy the next Fenix product no matter how long it takes. Until then though it's impossible for you to fight against the energy and vitriol that the community can inflict upon developers. When I read on this forum I realise that there are a large amount of people who are deeply interested in aviation but actually very few understand or have some knowledge about it. That's however because most are not engineers or real pilots. That isn't their fault but the consequence is their expectations are varied and founded in that misunderstanding. At the end of the day you're trying to replicate a living and breathing thing that has been designed by teams of engineers within a multi-billion $ company, and even that environment is fluid as aircraft are always changing. Hardware and software wise, the first A320's built in the 90's are vastly different to an A320 built yesterday, most people really don't understand that. It's ridiculously complex when you throw in engine variants or the sub-configs used by various airlines. As you say, you could trick the ECAM to show EPR values, roll in the IAE model and you're done. However even with the model alone, will everyone recognise that the flap extension at config FULL is around 5 degrees different for IAE vs CFM, what about the aerodynamics impact? The majority won't recognise these intricacies but this is why even a variant of the same aircraft takes so much time. I guess what I'm trying to say is I don't think you need to answer the door in terms of criticism. It is impossible that you could ever truly appease everyone.

Rather stupidly opted out of the SU13 beta and got the dreaded reinstall message (130GB update!). I let it run for 2hrs where it only installed 30% before I realised that it was installing to the default location and not my usual drive location, hence the claimed need for a reinstall. So I've started again with the installer pointing to the correct location, 25GB update now required which is downloading at 2Mb/s on a 900Mbps connection... I certainly don't think that I'll be flying this weekend at this rate.

The funniest thing I find about the obsession with EFB's is the fact that they're so boring. I literally can't think of anything worse than a screenshot of an EFB but kudos to PMDG if they manage to put one out there and successfully dress it up as exciting. Maybe once a screenshot is released we will see arguments on here about the shade of grey on the plastic surrounds... EFB's in the sim are also ridiculously awkward to use. A ground chart tracking the aircraft for example is useless in MSFS because you cannot see where you're going whilst looking at it. I prefer to do this on my iPad or iPhone driven by Navigraph which totally defeats the point of the sim based one. You cannot look at charts on a sim EFB and the FMC at the same time which also renders it almost pointless because it's just really awkward. The only benefit I see from any EFB is for performance calculations but meh, a quick Google nets you something fairly usable and similar. TopCat did this 15 years ago and even excel can do it! Personally, I think the only critical importance for any MSFS EFB is the ability to drive it on an iPad. Now that would be crazy good, Fenix and FSLabs are the only examples that come to mind. Until then it's best to get an iPad and use that as your EFB. For the rest of time, don't worry about it, fly the aircraft and look where you're going! 🙃

There is a difference between an automatic approach and automatic landing. In an automatic approach the aircraft will follow a form of guidance and then the pilot must decide to manually land the aircraft at the decision height or go-around. Beyond this point the autopilot system isn't certified nor is capable of landing the plane safely. GA aircraft generally aren't equipped for autoland because they just don't need it and it would be extremely expensive to implement for little if any mission gain. You would never takeoff let alone try to land a Cessna 172 in 100m of visibility. A Cessna 172 will also generally not fly into a major airport, it'll fly into small airports where instrument landing systems are not available or not certified to a high standard. Larger turboprops could have them but turboprobs also generally don't operate into major internationals with a certified ILS. So even with an autoland system equipped, it still may may not be possible to land at some destination airports which defeats the point of the high cost of having the system. It perhaps remains an emergency system except for the large airliners who often fly into the large and well equipped airports.

I think there is a misconception that developers are pricing greedily to turn large profits but this isn't really true. A second misconception is the lack of knowledge in the community around the development curve and its associated costs. All of the costs in aircraft development lie in the complex programming and aircraft systems knowledge (sorry cliche I know but time*skill=money). FS development is niche, there are very few developers with the expertise required. It is very technically demanding and for study level aircraft it goes even further because you need the involvement of technical and industry experts. It's also true that it's much harder to develop some aspects of the F28 because it's a very old aircraft and flying examples are non-existent. The technical documentation probably doesn't exist and you can't ask experts because unfortunately many aren't alive anymore. Unfortunately MSFS sales do not work solely on a sort of demand scale like most think, the economics are actually quite risky even for what would be the most popular aircraft releases. A study level 787 for example would sell extremely well however it would take years to develop and so there's a lot of risk in deciding whether those long development times and costs can be met or are justifiable. It's not viable for example to run a 5 year development cycle which ends with the requirement to charge customers $500 per copy to justify your costs. All of the above reflects what happens behind the curtain and the difficulties developing in a niche yet complex market. In terms of the F28, it's very likely that the F28 will be a lower volume product so it has to be sold at a higher price to meet costs, that's pretty much it.

It's still possible, on the CF6-80 there are two sensors located at around the 10 and 2 clock position whilst looking through the air intake (see the below pic). These are not visible in the screenshot but it may just be the angle (and perhaps purposely? 😁). Thinking of it, the A300-600 uses a CF6-80C2 engine which should also have these sensors visible in a similar location, so I don't think they're referencing the A300 unless they're producing the older variant which is the A300B4-200, they didn't do this variant for XPlane. So the A300 seems more unlikely and I wouldn't really call it 'ultimate'. I am betting on the DC10 or B742... The B742 however had several engine options whereas the most popular variant of the DC10 (-30) only had one...😁 Unfortunately I just cannot find any good photographs of a DC-10 fitted CF6-50 inlet, they're very old engines now!

The A330 is ruled out as it used CF6-80E engines which has 34 turbine blades, you can see the blade count in the image is 38. I can't see a P2T2 or vibration sensor and the CF6-80 definitely has them so for me that rules out the 767/747-400 too..... I don't know anything about the CF6-50 so my bet is it's a DC-10-10/30, 747-200 or more than likely their A300. 😛

Business travel was the primary driver for the continued existence of Concorde. In the end it operated almost like a private jet with relatively low passenger volumes and high ticket prices. It was set this way because offering another level of business travel was the only way to justify high ticket prices to meet the running costs. Had it not been for the British/French governments subsidising loss-making production then it wouldn't of existed in the first place. You can do the math relatively simply. Concorde typically used around 93t of fuel for London-New York which is ~115,000 litres. At today's prices that would cost you £130,000 for fuel. Let's also say that Concorde is running that flight with 120 passengers. That's £1,000+ per ticket just to break even on the fuel whilst ignoring all other costs. To account for all other costs that £1,000 would become perhaps more like £3,000-£5,000 which is excessive and you don't get a lie-flat bed either. You can absolutely forget economy fares. It's these economics that killed Concorde and those same economics haven't changed unfortunately and they will kill any future supersonic/hypersonic aircraft too unless the technology is radically different. That being said it was an incredible aircraft however I view it in the same way of the Space Shuttle, great machines but really bad ideas.

For those not seeing the update there is an MS Store update for MSFS first, do that and it'll appear on the next launch.

It's a little bizarre and disheartening that someone will actually talk down their own product to manage expectations. I'd rather put my energy into delaying release to work and expand the feature set so I can talk it up and not down. At least why not promise to release it with initial features and updates to improve it? When MSFS was first released the freeware stuff was pretty average at first but they're becoming excellent, the likes of the FBW320 are getting close and then you have the Inibuilds A310 which is excellent. If a payware developer can't compete with freeware developers on quality then they're doomed. The message I get is our market for this product is the unsuspecting naive who are there to be taken advantage of because what do they know? It's really sad to be honest.

I bought a Dominos Pizza takeout one night last week. I'm annoyed that I decided to buy a medium sized pizza in the knowledge that they offer a large sized pizza. Next time however I will buy the large sized pizza. The problem is it's the same as the medium sized pizza and the only difference is it's just larger and now that annoys me too. (simmer logic)

Usually a UK winter looks like one of two things, it's sometimes like today where it's very cold, the air is dry and there are few clouds (so little chance of precipitation or snow). Or second, the Atlantic fronts drive in moisture, precipitation and increased temperatures (which again makes snow more unlikely). To show how influential the Atlantic fronts can be here - I live around 10 miles from the coast and it's currently sunny with zero clouds but it's a relatively very cold -2°c. Next Monday however it's forecast to be 11°c but with clouds and rain.

I have bought the -600, -700 and now the -800, they're absolutely great because you always get quality from PMDG. By the logic used by some then it seems that PMDG should have produced the -700 and left it at that... However then you'd find another chunk of people moaning that there isn't an -800 or -900 or yes even the -600. I've even seen people moaning about the lack of a wingletless model... It's painful reading and you just can't satisfy everyone in this community. Personally I'm just thankful that these aircraft exist in the first place. I've seen a lot of comparisons made to the Fenix A320 but any simmer with a bit of age under their belt will remember that worthwhile Airbus aircraft didn't exist at all until the FSLabs came along only 6 years ago. These comparisons are therefore lost on me because we've never had this luxury of choice. For me I'm like a kid in a candy shop and so I'll always look to buy it all whatever the price, it's unfortunate that others don't feel the same way. A good example of my feeling is to try this - go and fly a 767-300ER... You can't because it doesn't exist. That's what flight simulation has been like for over a decade. To be able to compare the PMDG 737-800 at this quality in this sim to the quality of the Fenix A320... It's a dream and honestly guys you don't know how good you have it. I bought their 777 twice for FSX and P3D, I'll buy it a third time when it comes out.

I don't really get people's obsession with them, a simple google search or an Excel spreadsheet can be used to achieve the same thing. There are countless human factors studies available on the subject of EFB dependence. I guess nothing rings more true on this threat than in flight sim as it seems many people are forceably grounding themselves on safety grounds due to the lack of one... 😁

I'm not surprised P3D aircraft sales have stagnated because unlike MSFS there haven't been any new aircraft for P3D for a very long time. The last really big release was the NGXU but that was several years ago and it came off the back of the NGX. Since then we've just had variants but variants tend to only appeal as niche products, far fewer will buy the cargo or BBJ 737. The 777-200ER was welcome but sales were probably poorer because it wasn't updated to NGXU standards so it appeals far less. It's hard to forget that many people owned the 777 for FSX and PMDG's release to P3D meant we had to effectively pay twice (I was one of them). The full 777 product line still today costs over $200 which is an insane amount of money for a very old product. This is why I find any 'P3D is dead' argument to just be lazy to be honest. The vast majority of the sim community outside of Xbox will have come from P3D. Many of us follow the aircraft on offer so if something new came out for P3D then we'd all buy it because we've done it for many years. I just added up my spending with PMDG over the last 5 years and it's $605!

Try removing/relocating the .agn files away from the MK_EFHK/texture folder so they aren't loaded and see if it improves. It may just be that there's a lot of trees placed in the satellite photo area. Lowering the draw distance will reduce the total number of trees visible and so the CPU has to do less. The tree autogen library for P3D was at some point converted to PBR by LM, this might cause some performance discrepancy between users as many run ORBX Tree replacement products which aren't PBR converted (ORBX users will probably see better performance).

I don't even think that the dynamic lighting is bugged or anything like that as the lights themselves behave no differently to standard effects. The problem is a complex airport scenery can have hundreds of these lights placed in locations designed to direct light onto PBR surfaces and this is the problem. There are two settings in the P3D configuration file which control the maximum number of these lights in a scene (normally it's 250). If you set these maximums lower then performance will improve but lights can be suppressed and even the lights in your aircraft VC can be suppressed if there are too many in that particular airport. Rendering performance has increased massively over the last 5 years (mostly thanks to the move to DX12) which has meant the sim generally works well despite these changes. However, those on older hardware will of course suffer and scenery developers want their sceneries to look the best which means more lights and more PBR textured objects etc so less performance. Volumetric clouds are the latest thing from LM which are no longer just fake textures but are GPU calculated/generated and so will further hit your GPU. We're already on DX12 so unless LM work on more friendly calculations and shaders then the only way around all of this is to reduce the impact on the GPU which means a lower resolution or to reduce the amount of post-processing by lowering AA. Everyone has said for a long time that P3D is a CPU bound platform but on modern hardware I really don't think that this is true anymore.

I do have EFHK so can try but does this only happen at night or in the day (ensure the season is set to summer to test)? If it's only happening at night then it sounds like it must be the dynamic lighting? It seems that this will happen a lot with EFHK specifically because it's more or less always night in Finland at the moment. Dynamic lighting performance is very affected by resolution as the lighting PBR calculation result must be rendered at this resolution by the GPU. Add in aircraft lights too and things get very heavy on the GPU.

Trying reducing your resolution down to 1080p and see if it improves. If it does then your GPU probably can't do what's it's been asked to do. I never had FPS issues with Prepar3D until I got a 2k monitor and since then my FPS has noticeably suffered. The worst stutters I get are at night, especially with dynamic lighting on where stutters even happen at smaller airports. The biggest killer airport I've seen is Orbx ESNQ but again only at night and it's a tiny airport so it cannot be the scenery itself so I believe it's the dynamic lighting that causes it. I once tried reducing resolution when I was getting stutters and this completely fixed performance. Obviously running at a lower resolution or with dynamic lighting off isn't an option. The only solution for me has been to reduce AA and to set EA to low which gives something acceptable. Those running much better GPU's compared to my RTX 2070 will likely see far better results. In my mind, it's the general increase in monitor resolutions over the years that is primarily responsible for poorer performance as opposed to something LM or NVidia have introduced (dynamic lighting excluded). I also have display scaling (125%) enabled because my vision is bad and I haven't really tested if that affects things too.

No-one has really mentioned this but this is an excellent update from T2G, easily their best ever, looks great with True Earth.