I'm just curious as to the relevance that one gets with a default airplane if one doesn't use the default planes. I would have thought it would make sense to test using a plane that one would likely be using with the scenery.
It's just wrong to "test" a scenery using the airplane that one would likely using with a scenery because, if this airplane has a problem, one would end up not using any scenery anymore because he would get the wrong perfomance impression from each and every scenery out there, you are not testing a scenery anymore, you are testing the airplane. Note that, I'm not referring to the QW757 airplane in particular (there are *far* worse airplanes, from the performance point of view, out there), it's a generic concept: to correctly assess performances, one needs to test items separately.
An example:
- Suppose one has 40 fps with JFK default, using a default airplane and NO addons installed whatsover.
- Suppose installing FSDT JFK the fps goes down to 33 fps. Which means, a 17% hit over default
- Suppose installing an addon airplane, the fps goes down to 17 fps. Which means, 50% hit after installing the scenery
- Suppose installing an AI traffic package with a busy real world schedule at 100% fps goes down to 10 fps, another 45%
So, we have a scenery that impacts for 17%, an airplane that impacts for 50% and traffic that impacts 45%. Since the scenery weights for 17%, even if one chose not to install the scenery in the first place, but still using his favorite airplane and the traffic package, the fps would have been 13, which is as disappointing as 10...
That's why it's important to check items one by one. The first route to get better performances, is to KNOW how much each addon impacts and working on THAT one first.
In any case, my reply didn't had anythign to do with the airplane. The original poster said he has better frame rate with "all other big airports sceneries", like Aerosoft EGLL. To that, I replied with a post that clearly shows JFK is FASTER than EGLL, under the same conditions so, I'm suspecting the poster haven't tried JFK under the same condition of the other airports. Perhaps, he had a different airplane then, or he changed something else because, as demonstrated by the posted screenshots, JFK is clearly faster than other similarly sized airports. But of course, I'm testing in the only correct way: default everything with BOTH sceneries.
Because, if one has installed something else in the JFK area (eg. Aerosoft Manhattan, Imaginesim KLGA and/or KEWR) that is affecting the fps, but doesn't have anything of the sort in the London area that is affecting EGLL, it's clear the test was flawed.
JFK is an incredibly fast scenery, considering its complexity and its size, but it can't do miracles: if the fps in the same area was already bad to begin with, surely JFK can't improve it.
And, in any case, that machine the poster was using, is underused in FS9, and JFK is particularly suffering from this, because it has plenty of optimizations (a big part of its code runs outside FSX by our Python scripting engine that is a separate .EXE, which means it takes advantadge of multi-cores) which are only valid for FSX. As I've said, I have better fps in FSX, but I also don't get THAT much better fps in FS9.
