Flying Penguin

And that's exactly my point. The rest of the conversation around this is merely best guesses with varying levels of optimism that we will not have a firm answer to until implementation of those new functions.

So, remind me again why a statement that increased complexity will tend to decrease performance from current (impressive) levels, and that we can't yet quantify the exact impact is a controversial statement? As I said, I am not in any way criticising Aerofly, it looks like a well engineered solution in as far as the implementation to date goes, however we are talking about non-frame-limiter performance, therefore something physical is the limiting factor.

But that was exactly my point, so I'm not sure what you are objecting to? Whilst add-ons will have some overheads that inbuilt functionality will not (mainly around usage of the relevant API), exactly the same point about increased complexity (whether through increased polygons or working mixture levers/FMS etc) tending to decrease performance in the absence of something else giving way (through changing of settings or accepting a lower frame rate) applies to inbuilt functionality as add-ons, unless Aerofly has somehow worked out how to break physical performance limitations.

It's relatively straightforward and easily demonstrable. Start with any sim you like (probably XP or P3D as they have bigger add-on ecosystems). Turn autogen off, set clear skies, disable any atc addons, load a simple plane and fly above New York. You will get a framerate, doesn't matter what it is for our purposes, let's call it X. Start turning on autogen, ATC addons, load a PMDG aircraft, set the weather to cloudy or stormy. At each stage you will see a framerate drop of some magnitude as those cycles dedicated to chucking frames out the door at X speed are spent on something else. At no point is the framerate ever likely to exceed X. X is 100 for Aerofly on HiFlyer's system at the moment. Whilst it probably uses resources more efficiently, Aerofly is subject to the same physical resource limitations as any other programme, therefore if Aerofly achieves the same level of complexity and completeness that a "fully loaded" XP or P3D setup achieves, framrates will drop (we can discuss whether that should be their goal, but let's assume for the moment that is the case). Does that mean a 10% or a 50% drop? Not a clue as to the specific figure. But given the delta in complexity is large it's probably non-trivial. That is not a criticism of Aerofly, merely a recognition that on any given machine they have a fixed budget of CPU and GPU cycles to spend.

Thanks for the hard work getting things back running, looking much better. As a follow up to my earlier point about the mobile site, it seems that the menu appears in the right place now (on the visible section of the screen!), however the menu text is now dark grey on slightly darker grey (except for the arrows, which are white). Whilst it works, it would save everyone's eyesight if you could add tweaking that text colour to the list... Cheers, J

Well if you don't want to click, Google: "Cloudflare parser bug". TL:DR version: If you've signed in to any major site since Sep 2016, the chances are you've used one based on CloudFlare and your password could have been leaked. Any passwords older than 18th February 2017 are potentially affected. If you use the same password on more than one site, the risk of misuse is higher. And I assume by "don't click on any links", I assume your homepage link in your signature is on the list of links we shouldn't click on?

Hate to be negative, but the mobile view is pretty bad. Unlike most sites where I click the three lines and the menu just appears on screen, here the view just dimmed and it was only by accident I realised the page had been expanded and I had to scroll off the page to find it.

As another mug with a professional interest in this, I have to start from the basic truth that the only way to be safe without AV is to have the network cable unplugged. Routers get compromised (particularly those supplied in bulk by ISPs), trusted sites get hacked and start distributing malware and sometimes people just click on the wrong thing... Whilst AV (in the narrowest sense) is not a complete solution, when combined with the web filtering, on-host firewall and ransom-ware protection features that come with all but the most basic suites, it represents a no-brainer investment. The top names have already been mentioned (BitDefender & Kaspersky) though others such as Windows Defender are adequate (albeit they generally achieve lower detection rates), but the key is to have something running continually. There's no point relying entirely on running a periodic malware scan when the first thing any decent malware will do is disable common AV/AM programs. I'm running BitDefender Internet Security (1st line of defence, on access) & Hitman Pro (2nd line of defence, periodic scans only, fulfils similar role to MalwareBytes Free) and they've never had a noticeable impact on performance, especially if you exclude known safe processes from the on-access scanning. There is another option that I haven't seen mentioned, and that is moving the scanning to the network level, i.e. aiming to stop nasties between modem and your PC. This used to be the preserve of large organisations, with racks of servers and highly trained staff or expensive support contracts, often under the umbrella term of Unified Threat Management*, but is now starting to appear for the home user, the advantage being that it protects every device on a home network without having to put AV on each individual machine (although it wouldn't be a bad idea to do it anyway). The BitDefender Box (http://www.tomsguide.com/us/bitdefender-box,review-3766.html) is a good example of a (nearly) plug and play solution, although if you are technically minded, willing to get your hands dirty and have a spare old machine to dedicate to it there are a number of free (as in beer) licences available for full-on UTM solutions (e.g. Sophos UTM 9 Home, pfSense) that will give proper enterprise-grade security if properly configured. Cheers, J *Basically a fancy way of describing the combination of AV, signature based intrusion detection/prevention, firewall, web filtering and a whole other bunch of security capabilities into one box.

Microsoft can't maintain an infinite number of OS' so it's a question of life cycle. W7 is now in what Microsoft refers to as "Extended Support", i.e. it'll get security updates but no new functionality (including adding the ability to take advantage of new tech, new chip architectures often require OS support for new features). It was in "Mainstream Support", i.e. it gets improved where necessary to use new technology, for 6 years (running out in 2015), and even W8 will leave Mainstream next year ,6 years after release, leaving it in the same situation. And in under 3 years, at the ripe old age of 11 and 4 years after the last Windows 7 PC was allowed to be sold, W7 will go the way of XP and stop receiving any support at all. Anyone still on 7 now should be eyeing the move to 10 (or 11...?) in the next 18-24 months or so unless being on an OS that doesn't get security patches (and therefore becomes increasingly vulnerable) sounds like an appealing prospect. The exact dates are freely available: https://support.microsoft.com/en-gb/help/13853/windows-lifecycle-fact-sheet

Definitely agree. Birthday present to myself and very happy so far If only more other developers had release versions as reliable as this!

Nice vid Regarding cutting the corner, the only reference on CATH1 to MEA is once you've done the full swing back via the 104 radial, and for that specific SID it is common to get specific instructions whilst airborne to allow you to cut the corner, but given the controller was talking to another aircraft it wasn't the case here. The giveaway is that the whole 104 radial and then back round via the 124 radial section is below the gap in the text instructions, it is therefore the only transition section, and you should always expect to fly one takeoff section (with runway numbers) and one transition section (with transition names, or just the bottom half if only one transition exists). There is no such provision in the US that allows you to leave the horizontal path of a SID once above MEA unless you get specific instructions from the controller, reach the termination of the SID or if it explicitly refers to an altitude above which you may proceed on course. It's likely that the reason the controller didn't say anything is that by the time he looked back to you, you were where you would have ended up anyway and they may not have noticed the shortcut (or decided life's too short). There is such an option for diverse departures, but it's not relevant here.

Buried inside the 747 you say?

It's probably not fair to call it "old school" yet, a mature technology with a clear successor on the way, sure, but until VR matures a bit more in terms of performance on non-cutting edge machines, quality machines and interfaces for cockpits and things like chart reading, I'm not sure you can call TrackIR "old school". There is lots of life in TrackIR yet :smile:

If the vis is good, hand fly from as far away as you feel like, 1-2 miles is a good comfortable ballpark, more if you feel sporty. For instrument approaches, you can reasonably leave it in until your decision height. If you are flying to CAT I ILS minima, you can reasonably leave it in until the CAT I minimum decision height of 200ft above the ground, at which point it's AP off then pretty much straight into the flare. For the LNAV approach the decision height would be higher and approach specific (specified on the charts), but 6-800ft would be a good ballpark figure. You can bring LNmap up to date by purchasing an up to date AIRAC from Navigraph and subscribing to fsAeroData. The Navigraph cycle will update your Dash, and is the input to fsAeroData to update the sim and LNmap. You only need one cycle as once they are in synch you needn't update them unless you fly online. I don't believe there is a way to get them in synch otherwise as they can't read each other's formats. Here's a document written for the Majestic forums by a RW Dash captain that is 27 slides of just pure landing advice: http://majesticsoftware.com/mjc8q400/resources/LandingTheQ400.pdf Bit of light reading

I know that this wasn't the question, but if you are using it as a way to get back in the saddle, perhaps it's worth looking at PilotEdge? It's a paid service, and the coverage area is either ZLA (all towered airports + approach/centre), or ZLA + key western US hub airports . Whilst you lose out on coverage area and it not being free, what it does give you, over VATSIM is defined service hours (15 hours a day), guaranteed coverage of every frequency from clearance to centre in the coverage zone and a consistently high standard of controller training. The coverage is achieved by having controllers cover multiple frequencies, so you might (worst case) talk to the same person from clearance to centre and back down, but they will be handing you off at appropriate points and respond based on the frequency you call them up on, even if you've just been talking to them via a different frequency. It positions itself more as a training aid for RW pilots than being a community of enthusiasts, so if training value is what you are looking for, it might be worth a look. Either way, there's a 2 week free trial available, nothing to lose really :smile: