JaneRachel

thanks for that info Stefan, appreciated!

Has anyone noticed the aircraft wobble (almost like asymmetric thrust) when reducing from FLEX to CLB power on takeoff? Started happening with me after SU15. I installed the new version today and it is still there. I have checked my throttle calibration (TCA Airbus throttle hardware) and reconfigured in the MCDU and checked. The problem persists though in the new build for me. I am also seeing the cabin crew voiceover cutting out again in this build with the engine starter cutout overriding the safety announcement and losing the safety audio. Jane

Morning, To be clear, I don't have or ever had access to PMDG code in any way, so I cannot speak for their internal code. You are, of course, quite right about the CPU in avionics kit being less powerful than consumer hardware, but the difference is that those CPU's are dedicated to avionics, whereas our CPU has to render the aircraft, the world and do everything else also. I also agree on replicating maths formula. That formula though can get expensive and I would argue that a level of accuracy is required, in interpretation of ADIRS data etc. I defer to your knowledge though in the sim (as I have never written this code for an aircraft, only in "mainstream" games as a games AI and pathing developer).

Morning Gang! With my game developer hat on (my day job), I wanted to quickly comment on the pathfinding algorithm. There is always going to be a trade-off on a home simulation as to put it simply, perfect pathfinding is going to eat all your processor time. The more processor budget that is spent on this, the less time for anything else getting done. Perfection costs! So you have to balance CPU cycles across all the competing needs. The pathfinding at PMDG has come an awful long way over the years, both with the rise in computing power freeing up more cycles and Vangelis and the team really honing their craft (I started my relationship with PMDG in 2003 on the beta for their first 737 and beyond). We have come a long long way. There is still more to be done but much of that will need greater computer power as we iterate through new generations of PC's. I have spent a large part of my life creating pathfinding algorithms for commercial products and I look back at my previous work on this (such as Alien Vs Predator - 30 years ago this month) and cringe, but its all the processor I had to play with. Bottom line is that we always rob Peter to pay Paul when it comes to resources. Processor has to be balanced across so many competing needs in a sim. The guys have done amazingly well as have other developers with the level of fidelity we see today. I have seen the "jiggling" over the waypoints, which is much less pronounced on the 777 than the 737. I am speculating, but this may be linked to waypoint precision. Code tries very hard to hit a waypoint exactly. Quite often (and we see this a lot in videogame pathfinding, be it Call of Duty or a racing game etc), the position slightly overshoots. Lots of reasons for this, which would eat up a whole post on the technicalities... The code then tries to bring the position back. So you get the little jiggling with the overcompensation. One trick to help alleviate this is to add a little damping as you approach close to a waypoint to avoid oscillation and over-correction trying to hit the waypoint on the nail. I am sure the guys will continue to tweak it. It is a very common issue in pathfinding algorithms when you are aiming to hit a waypoint in 3D space very exactly. Again, the smarter you get with this, the more processor it will cost. Bottom line, until we are all running supercomputers (and I used to make a living writing games for computers with 1k of memory!) we are going to have to live with little trade-offs to manage performance. Have a great day everyone, Jane

I spent some time with a guy from the MSFS dev team early this morning. It turns out not to be a parking brake issue. Its a weird calibration thing on the pedals themselves. We had to recalibrate a number of times and then it works. I can confirm for anyone seeing this, don't mess with your parking brakes, it is a red herring. Resetting the parking brake on and off just temporarily resets the toe brake calibration. The poster above was correct in that in the game device calibration, you have to ensure that after calibrating your pedals the cursor in the calibration window sits and the bottom right hand corner of the window. It took us a few attempts to convince it today! Once that calibration is finally successful the problem goes away. (We tried this with a number of pedal manufacturers). It seems to be an issue in the windows game device calibration. Again, it is not a parking brake issue, that just temporarily unsticks the brakes, so avoid going down that rabbit hole if this happens to you. It is a pedal calibration issue confirmed now. Hope that helps others who may see this issue! Happy to help out anyone who also sees this happening. all the best (and thank you Fielder for your help!) Jane

this is now happening to me since SU15... any advice would be most welcome, many thanks Jane

A quick update, as it may be useful for people. It transpires that GSX can cause seriously bumpy taxiways! A solution is posted here... Troubleshoot - GSX Causing Bumpy Terrain, Double Parking Spots & Broke – Macco Simulations

thank you, I will give it a try!

Hey All, Has anyone else seen an issue with this scenery? All of a sudden I am now facing sudden hills and craters on the taxiways (so steep the aircraft can get stuck!). I have reinstalled but to no avail. Also turned up the mesh resolution as far as it will go... but still huge issues. Any advice would be most appreciated, many thanks Jane

Typically these days we have a primary product team, then a much smaller support team to do the followups. The support and tidy group is usually entirely separate from the teams leading flagship titles. We are lucky that the sim has proven popular in sales, which makes a support team cost effective to create all the service packs. Sometimes in the lifecycle of a product we have to draw a line on service packs simply because the cost to develop the pack outweighs the current strength of sales revenue, based on a calculated "per head revenue" generation of everyone in the team. Keeping service teams smaller really keeps the costs down and drives that equation of salary vs revenue contribution for each dev to something manageable.

Its very much the nature of the games/sim industry. In 44 years in the games industry (and sims too), its not the spec that kills you, but the bugs that suddenly defy squashing. I had a major title held up at EA with a crash bug that took my team weeks to find. It was a simple thing, in a couple of lines of code, but brought the entire project to its knees. We had pretty much every coder and myself hunting through code to track down. Conversely, I have seen bugs that look like they are going to be mega difficult, that I have ended up fixing over a cup of coffee in minutes. You are right about hundreds of people, although in terms of many issues, that boils down to just a few talented programmers to fix, with the majority of teams being art, design etc etc.. The specialist programmers are a minority on most games and sims (for lots of reasons, but I wont write an essay 🙂 ) Back when people paid me to be a programmer, rather than to run studios and publishing companies, we had all sorts of issues, for example, with Alien Vs Predator for Atari. It was coded by just two of us, myself and Mike (a very talented programmer). Some days it was bug city which we had to splat like the Aliens. Other days life was smooth. The moral is you can never tell and projects have a life of their own when it comes to shipping dates, despite so many processes we now have in place in the big studios. I routinely factor in a 20% time contingency on every project with the studios that I lead (which indeed can be hundreds of people now). Its why sometimes that many of us wish that we made business software which is an order of magnitude easier than 3D software and games, that come with animation, physics et al 🙂 Jane

thank you Aamir, Learn something about this bird everyday. Appreciate the response! all the best Jane

Loving the update.. A couple of things I have noticed. The aircraft seems to often spawn at the gate with the gear up and promptly slams in to the ground, requiring me to go back to the main menu. I have tried Heathrow and Manchester, same result. It is not every flight so looking for an obvious cause and will report back. A slight bump on the taxiway and the aircraft sometimes goes a few feet in to the air, even at very low speed (just had this again at the payware Manchester a few seconds ago). I keep getting repeatedly asked to enter DEST data in the MCDU, even when already entered (I saw this on the previous version also) Pedantic, wishlist item - button repeat on the MCDU pls so I can scroll through the flight plan etc by keeping the button pressed as per the real unit :) Fantastic work guys. Such a lot of hard work here that is appreciated by many, all the best Jane

A head up for anyone having issues with tablet functionality. The 787 mod (787-XH) needs to be removed from the community folder and you should be good to go. This particular mod creates conflicts, which are currently under investigation. all the best Jane

and Jon's advice is really always good! He is one of the most experienced "real-world" pilots here for a major airline I will not name 🙂