1 minute ago, NextGenSimmer said:

My apologize Ray. I honestly didn't knew this post will set such fire and hate. Believe it or not, I found this website on Google just yesterday. 

I did try to make it a discussion on not some random collection of rants. 

Feel free to lock it if you think is necessary.

No need to apologise. As you've discovered feelings can run high with MSFS discussion. Everyone is entitled to an opinion. The site is well moderated so if you feel any post in inappropriate for any reason report it and we'll take a look.

2 hours ago, NextGenSimmer said:

You really don't think there should be some reasonable time between updates ? For example, I will come back in another 3 months or even 6 months, and ask again "what is going on with the Fenix update ?", will you still reply in the same manner ? Since Fenix themselves did not vouch or gave any clue how long its going to take it could really be true.

I would start worrying if the devs would stop responding but that didn't happen. And other than that... I seriously enjoy the plane as it is right now. I pointed out one little bug to them and they quickly fixed it but... they also said the fix would come with the next update. Now I do wonder every now and then why they don't release a small update fixing small things (my bug was a little texture bug) and why everything has to be taken care of with ONE BIG update but in all honesty... the plane is perfectly fine for me as it is. And even of the devs would decide to call it a day and stop working on the A320 I wouldn't mind. But that's me. I don't care for IAE engines or sharklets or even the current fuel usage and VNAV implementation.

I think it's important to note in a discussion like this that not everyone has the same expectations or desires when it comes to addons. Quite often people THINK they are talking about the same subject but in fact they are not because the views out of which they speak are often totally different or even opposite. (This last sentence sounded better in my native language but this will have to do... which btw is another reason why some topics here go bad... 😉 )

Anyway, if you come back here in 6 months and ask "what is going on with the Fenix update?" I would also say "Yes, it's taking a bit too long now... WAZZUP?!?" But I would still be enjoying the plane. 😉

This is an interesting topic - I think it may be worthwhile for me to put up a little context from the other side of the river, so to speak. 

A lot of folks are comparing the process of updating the A320 to other aircraft. While that may seem like a fairly reasonable equivalency, it's not really quite so true. A good "marker" for this would be another A320, so for example, the guys and gals at FSL - they do some really good work with their A320. They brought that airplane out in 2016. They released the A319 in 2018, so 2 years in. And then the A321 in 2019, so a year after that (3 years in at this point). The Sharklets were a 5 year wait, from 2016 to 2021.

Fly-by-wire, the fantastic team of freeware community developers are 2 years in and also still working out the kinks in their VNAV implementation - and they have some seriously smart people on their team.

What's the point of sharing this? Well, the A320 is a hideously complicated aircraft. That's more or less what I'm getting at. Changes to one system knock on and cascade into.. well pretty much everything else. 

To give you all an example from this update cycle, we changed the elevator's physical properties to adjust the rotation behaviour of the aircraft to more closely match the real "feel" - a lot of users were complaining that the nose "drop" after releasing the controls during rotation is too strong.

So we went back to a Level-D sim, repeated the behaviour, recorded data at a number of different weights, came back - and then the work started - first a comparison - what does the Level-D simulator do? What do we do? What's the difference, qualitative and quantitative? Once we've looked at that, and know what the behavioural difference is in the data and the "feel", we begin assessing the Fenix. What is the fly-by-wire system doing? Is it was doing it properly? Is the speed of trim input by the aircraft during the post-rotation blending phase correct? Put the aircraft in Direct Law and assess the feel and behaviour there. Repeat, repeat, repeat. A hundred takeoffs, at different weights, and CGs, observing for any inconsistencies in the rotation. A pattern is found, the answer is the elevator. It's not behaving quite the way we want it to. Why? Over to the FM team. They study, assess. They start with a theory about the fly-by-wire system and the underlying "direct law" simulation that feeds into it. The system team steps in, makes some adjustments, return it - out for testing it goes to our type-rated tech team. The result is inconclusive, the aircraft no longer demonstrates this behaviour, but now the rotation is suddenly super fast midway through rotation, and that is very unpleasant. So the bad behaviour is gone, but rotation is no longer smooth. Now what? Artificially smooth out the rotation? No, that's not right - we aren't adding any "fake effects", so back to FM team it goes - they bite the bullet. We have to change the elevator's properties to work at this from 2 angles. Ok, sure. They get to work. It takes a week of iterating before they come back with something that's usable. We throw it in, start testing. Results are much nicer, but now the aircraft is auto-rotating off the runway by itself, in certain flap configurations, at certain trim settings, at certain CGs. Back to FM guys it goes, fettled, and returned. Cool, now it's working pretty well. The tech team are happy, and I am happy because it matches the data from the Level-D - and we've not accidentally simulated auto-takeoff. The rotation feels nice and smooth too. So what's next? Start doing some proper flying in it. A-to-B. We get that going, and lo and behold - everything is wonderful up until cruise - where the aircraft starts gently oscillating. Why? Well, that'll be because we changed the elevator. So, the PID controllers that "feed" the fly-by-wire in normal law now need a retune to accommodate the changes to the elevator, as the airplane thinks it's working with something it doesn't have, or vice-versa. Back to systems it goes - the specific weight, CG, FL, weather and wind, etc passed on, it's recreated, and then fixed. Back to beta. They test it. "Hey, feels good now, really happy with it!".... until an oscillation is found in some other flight phase, or the PID controllers for the auto-pilot need a tune also.

That's one off the list. One. All to resolve: "The airplane dips it's nose a little too much if you drop the stick, post takeoff. Unrealistic."

That's why things take time. There's an interplay with a half dozen control computers somewhere up and down the line for pretty much every change you want to make on this airplane, be it flight or systems related. A 737 is a little less finicky, I would say. Doing just the Sharklets involve changing the belly fairing of the aircraft, the ELAC (Elevator Aileron Computer) standard, different fuel pumps and associated characteristics, technically updated ECAM standard, etc etc etc. This is why FSL charged for it (and it didn't just show up the next day!). 
 

15 hours ago, NextGenSimmer said:

This actually makes me think, what happened during the initial beta testing before release ? How come a new released product needs an overhaul of the whole system ? I mean i can understand small bugs went overlooked thats normal, but replacing a whole system cause the current one is not good ?  

Well, VNAV is math. Math is either right, or wrong. If the math is wrong, the math is replaced. This is a lot of math. We are not rewriting the laws of VNAV. We're rewriting the math behind it, and making it more granular. The test cycle for this was something like 9 internal revisions and 4 beta revisions, with one more to come. We've also fixed some bugs with the VNAV rules, however, such as when the aircraft doesn't respect alt constraints in certain situations.

The same for the flaps. We changed all of that too. Tweaking one value will knock on into a dozen other behaviours. Flaps create both drag and lift. You don't affect one without affecting the other. It was a couple of weeks of work just to put them into the right window and then the subsequent fly-by-wire PID tuning (both AP and FBW.... again) for this change also. 

15 hours ago, Zimmerbz said:

I am curious, what is incorrect about the vertical navigation?  I find it to be quite good.  SIDS and STARS are followed very well for me.  Is it more of a technical thing?

The airplane would drop to the lower "gate" of allowed speed during its descent - a real A320 tends to run toward the upper gate. It can and should be anywhere inside those gates, but we've revised it to be more representative of the character of the airplane, so the math now has it "fast" most of the time.

16 hours ago, guenseli said:

Simply calculate what he has earned after release and think what you would do after the first rush is over: best holidays of your life! 😋

I wish :)

But, anyway. In the meanwhile we've also redone the entire engine soundscape and some other bits on the soundscape also, following feedback. Once again, that's... a significant undertaking. Not really something you do in a month. But we've gone after it because we're committed to building the best aircraft we can, not just in our eyes, but in yours - and this is something the community wanted. Clickspots, also painful to deal with, but has been completely revamped following feedback.

tl;dr The aircraft has been out for 4 months now, only. The first set of updates will always be big, bold, sweeping changes until we mature the product over time in market. We also need to adjust to our userbase, what they want, and what they don't - even from an infrastructure perspective. I'll talk more about this stuff once we get this next update out there. 

6 minutes ago, Aamir said:

 

Aamir, thank you very much for taking the time addressing it as you just did ! 

I honestly say now, with this post you reassured me we in good hands for the next update ! 

 

Keep it up ! 

14 minutes ago, Aamir said:

This is an interesting topic - I think it may be worthwhile for me to put up a little context from the other side of the river, so to speak. 

Thanks for providing the update Aamir. I also want to point out, to the people that wanted a response from Aamir, you got your response, but it probably took Aamir 45 minutes to type that up and also check what he wrote for errors.

That's 45 more minutes that Aamir could have spent fixing the Fenix A320 today 🤣.  This is why when some people complain that they can't interact directly with the devs at Asobo, and have to go through the Community Managers instead (like Jayne, Seedy, etc), that's actually a good thing that they have to go through the Community Managers, IMO. 

Anyways, it was very gracious of Aamir to type this up, so kudos to you 👍. You probably have to work an extra 45 minutes today to make up for this post though 🤣.

Edited August 18, 20223 yr by abrams_tank

Such a super insightful post as to just how the timescales of modern sim development and community expectations still seem to rarely align.

I always point this out as an example, but there are 10K+ hours into the NXi at this point. And that's just an avionics system without any FM or aircraft systems work. I'm completely serious when I say that building something like that is exactly the same amount of time, effort, and code as most mid-budget indie games, _on its own_.

Every new study level aircraft or avionics addon you can think of as buying another whole game's worth of development, easily.

-Matt

5 minutes ago, abrams_tank said:

Thanks for providing the update Aamir. I also want to point out, to the people that wanted a response from Aamir, you got your response, but it probably took Aamir 45 minutes to type that up and also check what he wrote for errors.

That's 45 more minutes that Aamir could have spent fixing the Fenix A320 today 🤣.  This is why when some people complain that they can't interact directly with the devs at Asobo, and have to go through the Community Managers instead (like Jayne, Seedy, etc), that's actually a good thing that they have to go through the Community Managers, IMO. 

Anyways, it was very gracious of Aamir to type this up, so kudos to you 👍

Exactly!

26 minutes ago, Aamir said:

The first set of updates will always be big, bold, sweeping changes

Understood! Great post. However. 😉 There also are a few little small bugs that could easily be solved with a simple hotfix (no installer, just some new files you could drop somewhere). Now obviously creating hotfixes also takes precious time but still... it COULD be a way to keep some people happy (and silent LOL) and it shouldn't take TOO much time. 😉 Like the little texture bug I posted about on Discord quite some time ago (which you probably won't remember but that's not the point here 😉 ). Within half an hour, if not quicker, someone from the team told me he figured out what was wrong and had it fixed already. (In fact he said already knew what the problem probably was once he read my post.) Such a little change should be perfect for a hotfix...!

But having said that: as I posted above I am perfectly fine with the plane as it is so I won't get hysteric of you decide to not release hotfix(es). 😉 It's just a thought you might (re)consider.

Edited August 18, 20223 yr by tup61

3 hours ago, NextGenSimmer said:

You really don't think there should be some reasonable time between updates ? For example, I will come back in another 3 months or even 6 months, and ask again "what is going on with the Fenix update ?", will you still reply in the same manner ? Since Fenix themselves did not vouch or gave any clue how long its going to take it could really be true.

There should be reasonable limits for show stopping, or severe problems. The airplane and it's systems work now. Are some of the numbers off? Sure they are. Amir and the team know that. They have communicated their awareness, as well as an action plan to correct it. That being said, I have a fully functional airplane. I can plan and successfully fly any flight that I want, and enjoy it. Fenix is enhancing their product. There is no requirement that they do that. If they stopped working on it altogether, I would continue to use it as is. I got what I paid for. If they chose not to update it, then maybe some other developer will come along and do another A320 without those issues. If I like that one, I might buy it, too.  

3 hours ago, Christopher Low said:

I am also curious about this. Does the developer simply think that the original code is not good enough, or has some aspect of the Microsoft Flight Simulator update process messed everything up? If it was the former, then surely it could have been seen to be not good enough before the product was released?

I think that they put together version 1 of a working product, and sold it to get their company off of the ground. Prior to release date, all they did was spend money. Now, they have a revenue stream, additional staffing, and the luxury of some time to work on both new and old projects. My company sells software. We fulfill the customer's base requirements and give them the product. From that point on, things always change. Either their requirements will change, or we find ways to enhance the product. If the SW doesn't work, we stay right on it until it does. Devs skip lunch, or work weekends and holidays if necessary. But if the changes are cosmetic, or are simply new feature adds, then the pace of development is much more relaxed. You only listed two possible scenarios. I think that there is a third. I suspect it was more along the lines of 'We can use the existing engine model, which works but has some incorrect parameters, or we could hold up the entire project (for months) while we develop a new one. Let's use what we have available, and put the new engine in version 2.". Compromises sometimes need to be made in order to keep projects moving. I am certain that there were some deficiencies in MSFS as well that limited what they could do, but now they have time to address those as well.

I keep coming back to a few basic points. #1 = What they sold me works. #2 - It is what I expected when I bought it. #3 - I do not feel slighted in any way over what they delivered. #4 - We're talking about pretend airplanes that we fly for entertainment. Waiting a couple of months (or even a full year) for a developer to finish doing whatever enhancements they feel are necessary is okay with me as long as I can use and enjoy what I have at the moment. #5 - I had ZERO Airbusses in my hanger throughout my FSX and P3D days because I felt there was no acceptable choice available to me. Now I have 2 great ones (and both are in a cycle of continuous improvements).

Edit: Amir was typing his response while I was typing this. There was a fourth option neither of us mentioned. They are just flat our passionate about their product. I am glad to be their customer.

Edited August 18, 20223 yr by MDFlier

1 hour ago, Aamir said:

This is an interesting topic - I think it may be worthwhile for me to put up a little context from the other side of the river, so to speak. 

...

Now that's probably the most informative post a dev has ever done on Avsim. Thanks!
Just to be clear: I think - due to translation errors - the word "update" is ambiguosly used by some people over here (me too sometimes). Some mean a patch, others might mean an information update AKA news. I think most people actually only just want some news. So while you did a great job explaining what takes the patch so long, this still doesn't explain why you can't do some short-medium announcements every let's say three weeks. To keep everyone in the loop and keep them excited about what to come.

4 minutes ago, Fiorentoni said:

Now that's probably the most informative post a dev has ever done on Avsim. Thanks!
Just to be clear: I think - due to translation errors - the word "update" is ambiguosly used by some people over here (me too sometimes). Some mean a patch, others might mean an information update AKA news. I think most people actually only just want some news. So while you did a great job explaining what takes the patch so long, this still doesn't explain why you can't do some short-medium announcements every let's say three weeks. To keep everyone in the loop and keep them excited about what to come.

I will be honest I meant a 'patch' in my OP. 

With that said, if Aamir above post would have come as news on their announcement section on Discord, this whole post of mine here probably wouldn't exist. 

1 hour ago, Aamir said:

I wish :)

Just want to make clear, that it was meant as a joke!

And I really, really hope that your Fenix Airbus is a real financial success, because in the end, that is it what feeds your family and I really hope you can make great vacations! You deserved it!

The generation of I want it now...

Even Carenado has delayed their PC12 release (in part to SU10).  

Have patience.

@Aamir

Devs 10 Flight Simmers 0

Edited August 18, 20223 yr by Ray Proudfoot
Don't quote whole post for no good reason!

3 hours ago, Aamir said:

So we went back to a Level-D sim, repeated the behaviour, recorded data at a number of different weights, came back - and then the work started - first a comparison - what does the Level-D simulator do? What do we do? What's the difference, qualitative and quantitative?

You've made your A320 be a copy of a copy? Why not use a real A320? The Fenix A320 is literally unflyable now!
[/sarcasm]

The level of refining described is wonderful - repeated tests at varied weights is the sign of a team dedicated to getting it right.

I'm excitably waiting for the next update of the Fenix, but with today being a school day when it comes to Aamir's and Matt's comments about sim aeroplane development, I'm going to have to be even more patient for further improvements in my favourite aircraft [plural].