Jump to content

Comment.... It is a complex world.


Stephen

Recommended Posts

post-2-0-96906200-1389837604.jpg

 

It is a complex World

 

The overwhelming thing I have noted on X-Plane Reviews over the last six months has been "you will need to spend time on this aircraft". It is an interesting point to consider. Being a reviewer you have certainly have a set of skills in debating an aircraft, but you don't today have the luxury of time of a Monthly or Bi-Monthly publication. The Web is immediate and constant.

 

The scale of simulation development over the last few years is also worth considering. In Xplane even a few years ago most aircraft systems and functions were still based around the fundamentals of Plane-Maker and if you were really brave then rudimentary add-on plugins. Ben Supnik noted in a recent FlightSim interview that:

 

"At a certain level of systems modeling detail, you need a plugin to implement the exact systems model of the real airplane. This is particularly true for airliners, where the systems are incredibly complex."

 

and that in the future projection of X-Plane in five years he noted that:

 

"I think the strength of Laminar Research is not that we make good guesses as to what will happen in five years - rather it's our ability to respond rapidly and try new ideas as soon as they seem even remotely possible. So I don't know where we'll be in 3 years"...

 

Both comments are very interesting to weigh up, one commentator noted on his site that Laminar Research had run out of ideas and has lost it's way... Thus the comment. I agree but only to a point on that in that as I have noted that in some areas Laminar Research do need to as the American's note "To step up to the plate". The basics require attention and (weather and ATC) and they needed to be addressed far more quickly than 2 years or so after the X-Plane 10 version was released. As for regional art Ben Supnik notes this:

 

"We definitely want to build a complete set of basic autogen before we start regionalization. Unfortunately we're just not going to have European-style autogen that soon; we are still working on the first set of autogen and it's a very labor-intensive process.

I think there is a big third party opportunity here; the simulator can handle regional autogen now if someone were to make such an add-on. We definitely hear the strong demand for it - it's one of the most requested features we get. (Unfortunately we are just limited by development time.)"

 

The problem is and as good as Alex (Gifford) is, and we certainly admire his excellent work. There is still only just one Alex?...  Laminar Research can spend hundreds of thousands of dollars on Charity work (and I support that), but won't spend a few dollars to employ a team or a studio of artists to create this art for the very product we are paying for. I found a few years ago a full set of Arab buildings (that would look really good in Dubai!) and I understand they are not in the correct format needed, but this stuff is still out there and available and cheap by the standards of the simulator....  Ben Supnik notes that we can have that now?

 

But I do agree with Mr Supnik on his main comments and why the commentator is missing the point on Laminar Research in why it does things in the way it does. You can't see the future in what will be released in technology even in one year and never mind in what it will look like in five years, all software companies that survive in this data age is that in the best they are reactive more than proactive. In the years past a company will spend years before releasing a product in testing and quality control to meet very high standards. That model has gone out of the window with most of the huge software companies now releasing software in a rudimentary form and using us "crowd forming" to work out their own problems. The trick is how quickly the company responds to the issues and in that environment Laminar Research's track record is very good, but it puts the basics aside to do so (because of limited resources) and that is not good in the eyes of the simulation community. However Laminar Research is also very good in giving us the latest developments now and that is why we have a very good progressive simulator that is becoming world class.

 

But it is Ben's first comment that is the most interesting. When I started in simulation if you knew the basics of Xplane's default systems then it was pretty easy to covert quickly to another aircraft. Take the default FMC for example.

 

The basic default FMC (Flight Management Computer) is good, but it still was and still is basic. If you moved to another aircraft in programming in a flightplan which was was quite easy, and then you could also save a full folder of routes that could be used in almost every aircraft in Xplane. If you today have purchased the high end aircraft with the authentic FMC, then your flightplanning becomes a whole new different level of systems management. Not only are they highly complex, then most also have various different needs on programming, to a point I accept that the actual flightplanning flow is the same in the way that you use and programme the units, they are still however very highly complex. If we demand this sort of complexity then there is a very large learning curve to understand the systems. You can't do this with just a few flights...

 

In the real world of aviation your head is usually stuck in books as you wait for the results of your latest medical. The real guys are working just as hard as we are to understand the same systems. The difference is they will sit on one type of aircraft for years, and that depth of experience will mean they will know that type inside out. We want it all now and can fly almost anything that our heart desires. A person noted on this site recently "SSG, please make a video tutorial on how to use the FMC." on the Boeing 748i.

 

Well kid here is the bad news "there are no short cuts" if you want to fly the Boeing 748i correctly and well then there is a huge amount of homework to be done. You can easily copy the button pushing but to really enjoy the aircraft it will mean hitting the books and flying the B748i for months. Then add in the release of the Boeing 757 and your workload has just tripled as the manual alone is 793 pages long.

 

Mr Frooglesim who has suddenly found the delights of our simulator notes that on "how complex" the Boeing 757 really is. And he has a point. It won't take you a few days here to understand this aircraft, but months...  even years. To simulator junkies these aircraft are to be slain like a dragon and to be conquered and mastered. But how far are you really be willing to go to really be a real professional on that aircraft. Once a week? twice a month?...   every day?  We don't have the luxury of time to spend that long in the cockpit in the same way real pilot do so day in and day out. But a commitment is certainly needed. I spent a whole year flying the FF Boeing 777 last year and do I know it down to its last rivet, no chance. In fact my learning curve has barely started in knowing every system and every pipe in the aircraft. It will take years to fully understand the complexity of flying such a huge machine though out all its different weights, distances and the various weather conditions to really be competent on the type. 

 

But the real experience is getting to the point of the "feel" of the aircraft. We try to find that level quickly in doing reviews. We are lucky as well because we run simulation after simulation (day and night) to find out the most information we can to explain the aircraft for you to make the best judgement on the aircraft. But usually at the end of a review you still need to fly the aircraft over and over again to get to the point of getting the right feel and usually then it is very hard to move on as you are now really liking (and wanting to stay) on the aircraft. But it can take weeks sometimes to get to this point. The problem is if you have a lot of complex releases in a short time there is a lot of research to do.

 

The SSG Boeing 748i is a case in point...  When it was released (in beta) form it was a mess. My computer could simply not run the aircraft. There is and are many bugs in beta aircraft. Many users want to be beta testers to get their hands on an early product (usually for free) but I don't recommend it, beta aircraft are nasty and very far removed from their final form. If you are really good then you can give the correct feed back and that is what beta testing is all about. As I tend to go very deep and fly early aircraft a lot, then I usually hit the problems earlier than most.  Most times it is usually just a gut instinct, in that "it just does not feel right". But you have to first divide the point of, is the problem the programming or is the problem you?

 

It is very easy to disguise a problem by having a powerful machine. If your computer is powerful with huge amounts of VRAM, then computer will override a lot of the faults, as it will easily absorb bad heavy textures and poor overweight modeling counts. But the B748i was a very different machine in the fact it was physically very big in Xplane. If had flown Peter Hager's huge A380 you would understand this effect, and you have to adjust your render settings to offset this. But with the B748i I had a double whammy. Laminar Research had just updated the ground textures in beta 10.25. And they have a strange effect on certain aircraft. In flying the huge B748i I found the aircraft reacted badly to these textures. With a large group of these textures on the surface the frame-rate would simply plummet to new lows. (as the computer worked overtime to process not only the huge model up in the sky, but the complex heavy textures on the ground)

 

What I found was that you had to set the aircraft up very tightly with the render settings to eliminate the effects. There were a few bugs (in the FMC) as well that diverted the problems, but once it was all sorted and most importantly with weeks on the aircraft it became very useable and flew very well. I spent many flights in the Christmas - New Year break flying the aircraft over and over and the more time on the machine then overall better it became. In this case It was my adjustment to the aircraft. But the case in point is that there are so many variables in the way we use aircraft now in X-plane is the fact that our sweet spot is a constantly moving target.

 

And this is the complexity that we are faced with running simulation in this era. The simulator is highly complex in the way it interacts with your computer. Gone now are the days of Xplane9 and default aircraft running around on a standard set of profiles. Xplane10 is highly reactive, but the plugin monsters that now prowl our screens are highly demanding of resources and more importantly of our skills to understand and learn them to a high degree. You have to really understand of how high the level has moved in only a few years...  and it is going to get worse.

 

Choosing and spending time on one type of aircraft for a long period can help your skill level in many areas. One is that you will come to really understand the aircraft and enjoy it more to its core (That is why the x737 is so popular), Two you can give yourself a bench mark to the effects of the changes set out around you. If the simulator changes then you will know because the aircraft you fly well is different in certain aspects. Resetting the simulator to return to same position after any Xplane changes will mean you will have the correct level to benchmark over other aircraft you are flying or notice how they are also being affected by the changes. If the aircraft is upgraded by the developer then you can see if it is a forward or backward release. Overall these new generation of plugin driven aircraft demand time and commitment to deliver the sort of simulation that is our Holy Grail.

 

So Ben Supnik is right in that Laminar Research can only react to the changes and not forge a certain path. He certainly doesn't know where the next change will come from, and only that to adjust the simulator to suit the demands is the best policy to cover, as this is the really the new way that the world is right now, highly complex and always constantly changing.

 

In flying we now have amazing but hugely complex aircraft to fly. We want simulation to reflect the real world, but the real world is highly complex. But mostly to fly these machines you going to have to work hard, they demand a higher level of skill not only in operations, but in systems management and knowing the aircraft. The results are simply overwhelming, we want the realism and authentication of the machine, but that does mean they also demand the same level of commitment as they require in the real world.

 

Stephen Dutton

 

16th January 2014

 

©copyright 2014: X-Plane Reviews

Link to comment
Share on other sites

Thanks Stephen for this nice write up about complexity ... which is definitely one of the main aspects of our life these days (not only in flight simulation!).

There is only one point I wanted to "correct", without being a nitpicker: Albert (Laubi) from Switzerland is "only" - at least until now - working on all the ground textures (and usually "only" in the natural, non-urban world). It is Alex (Gifford) who is responsible for all urban stuff (textures, and ina very large aprt all the 3D modelling / creation of the autogen world). And I am, Andras (Fabian), usually am working on the data structures, data processing, scenery generation etc. (think of it like the scaffolding that holds all the artwork in - the right - place :) ) ....

... hmm, I just realize, that it is indeed a bit irritating, that we all have names with A's to begin with :) (triple A).

Link to comment
Share on other sites

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...