Here's a quick A-B of FSX and FSX-SE with a pretty close vanilla setup everything maxed. No FSX Ai traffic, same .FLT file same .PLN, same weather, and same Ai injected. Getting around 15% improvement on fps here. One anomaly with FSX-SE to note is that some airport objects can disappear at the extreme zoom out, yet are all fully drawn in FSX. Those objects were fully drawn in by the time I zoomed in from 0.3 to 0.5 with FSX-SE.

 

Thanks for this comparison, SteveW. But also on your picture I can instantly see that the cloud draw distance is reduced in FSX:SE...

 

I know, it is nutpicking I do here and honestly, if you all get those miraculous 15% more FPS, that is fine. The only thing I am insisting here is, that

 

A) it is almost impossible to reproduce two identical situations in FSX (or FSX:SE), therefore benchmarking is difficult

 

B ) for the correct statement that FSX:SE is really performing better, both sims must use exactly the same settings

 

As long as this is not the case, I continue to doubt that your 15% performance gain is from anything else than reduced settings hidden somewhere. Just think about the hardcoded HIGHMEMFIX, it IS possible that FSX:SE also has some other adaptations for the sake of more performance.

 

And to that number one reply: I can also post FSX screenshots with high FPS numbers. So what? It is all about the COMPARISON of the two. And again, unless you show me two pictures of FSX and FSX:SE where I do not instantly see a difference in the settings, well, then I can start to accept that the performance gain is not due to little tricks in the hardcoded settings.

 

Example: http://www.orbxsystems.com/forum/topic/89728-ms-fsx-vs-dt-fsx-se-the-conclusive-performance-benchmark/

Those are so far THE ONLY screenshots I have seen where the user actually took uttermost care that the settings are as close as possible, and guess what? The FPS do not differ at all, they are even sometimes better in FSX:MS. Yes, it still might be "smoother" and of course, that you can not judge by looking at FPS and pictures and the VAS issue seems to be addressed better with FSX:SE, but performance? I still doubt...

 

BTW: no, I am not going to install FSX:SE myself, simply because it would take a to big effort to switch all my stuff from FSX to the steam edition and because not all of my addons are yet compatible. This has absolutely nothing to do with this thread, as I am AWARE that a brand new install of FSX:SE on my machine would in ANY CASE perform better than my actual FSX setup. But some people do not even understand this basic fact here...

AviatorMoser, Screenshots from outside views is always going to give higher framerates. I want to see the PMDG 777 on the ground at EGLL in the virtual cockpit with AI traffic and cloudy weather.

 

Anyone can tweak FSX and FSX:SE to get good framerates on the outside view. And I noticed you didn't show a framerate count with the picture of the 777 in clouds

I can instantly see that the cloud draw distance is reduced in FSX:SE

Turning off weather altogether, and choosing situations where all objects were drawn yielded the same results. There are anomalies with clouds too, but draw distance is not reduced.

You see, that is exactly what I am talking about. I am SURE that you did not reduce the draw distance yourself and that you use the same settings, but your screenshot clearly shows otherwise, no? You also did not add the "HIGHMEMFIX=1" line in your FSX:SE.cfg, but still it is used...

Besides the fact that when I set my frames to unlimited, they jump around 50 - 80, so it would be easy for me to take a screenshot in FSX once showin 62.3 FPS and then showing 74.3 FPS. You rather use the advanced Shift-Z display where you have also the min, max and average FPS numbers and you give the average FPS numbers you get after 20-30 seconds. 

As it happens I had the logging on during the tests:

 

 

 

 

HIGHMEMFIX made no difference.

I give up. HIGHMEMFIX is hardcoded into FSX:SE by Dovetail, that's why you do not have to put it into FSX:SE.cfg as you had to for FSX all over the years. THIS is the reason why I think that Dovetail might have also changed some other parameters outside the config without letting us knowing it. And again, when you log your FPS with the difference in clouds shown in your screenshots, I am not surprised that FSX:SE has higher FPS counts.

As it happens I had the logging on during the tests:

 

 

Can you tell me what program you are using for frame rate logging, please?

 

Pete

I give up.

That result was with no weather at all. In general when you mess around with it, you get the idea SE runs slicker.

 

Can you tell me what program you are using for frame rate logging, please?

Yes Pete, it's IF10 >augmented. Cheers!

Ok, that is indeed a valid point. I still do not get why, when you finally turn on clouds, there is such a obvious difference in draw distance although you use the same settings. The screenshots in the thread over at OrbX clearly show that IF the settings are the same and also show EQUAL on the screen, FPS differences are gone. Same here: http://forum.aerosoft.com/index.php?/topic/90621-fsx-steam-better-then-fsx-i-dont-think-so/ There is NO visible difference and again FPS are the same. Why Mathijs states that FSX:SE even uses MORE VAS, is somehow strange.

 

For the moment, it seems that some people get more FPS under certain conditions. Others do not see any FPS difference at all, but they have less OOM. Again others see no difference in FPS and even find that FSX:SE uses up more VAS compared to FSX:MS.

 

Now, tell me who is right. Pete for example clearly states that the increased performance is remarkable mainly when using the sim under heavy load conditions. You show a graph clearly depicting that FSX:SE results in a overall increased performance at a clear day, which is for sure not a "heavy load" situation. Basically we are back to the start. No questions solved so far.

 

Just one thing: http://forum.aerosoft.com/index.php?/topic/89774-fsx-on-steam/?p=643165 This guy is honest, at least. He also observed 10FPS more above Schiphol. Nice. Then he confesses that he did a brand new install of Windows 7 along with a brand new install of FSX:SE. And then he compares FPS...

 

 

BTW: no, I am not going to install FSX:SE myself, simply because it would take a to big effort to switch all my stuff from FSX to the steam edition and because not all of my addons are yet compatible.

 

With all due respect then, what's your beef in this thread if you aren't even interested in switching over???  I still fail to see what you're trying to prove other than trying to dis-prove the effort others have put up.  I never saw someone so vehemently defend an old dinosaur of a program.

 

If you look objectively at your own posts, it comes of as you're highly insulted that a company (dovetail) is trying to breathe new life in a program by making it perform better.  If, by your own assumption, they turned down a few things, then maybe it's a GOOD thing as people naturally try to make FSX do things today's hardware can't do because it just becomes a support nightmare wit those who have never used FSX.  I don't agree with your assumption of reduced settings...I actually USE the program and don't see any real difference that I can really notice other than I can do a flight without having an OOM....

 

 

 

Thanks for this comparison, SteveW. But also on your picture I can instantly see that the cloud draw distance is reduced in FSX:SE...

 

I disagree...if you look real closely, there are less clouds in the FSX shot.  The SE shot has more cirrus clouds and definitely some cumulus on the horizon...FSX does not have anything on the horizon.  If anything, FSX has killed the distance.

With all due respect then, what's your beef in this thread if you aren't even interested in switching over???  I still fail to see what you're trying to prove other than trying to dis-prove the effort others have put up.  I never saw someone so vehemently defend an old dinosaur of a program.

 

 

Well, as mentioned before, the topic title asks if FSX:SE shows really increased performance. And people started off to discuss, as some people think it is placebo and some think it is not. THEN, people started to show so called evidence and by closely looking at it, it is clearly not (besides the logged FPS). This is a personal thing of mine that I do jump on non-scientific comparison trying to prove something. That's all.

 

If you look objectively at your own posts, it comes of as you're highly insulted that a company (dovetail) is trying to breathe new life in a program by making it perform better.  If, by your own assumption, they turned down a few things, then maybe it's a GOOD thing as people naturally try to make FSX do things today's hardware can't do because it just becomes a support nightmare wit those who have never used FSX.  I don't agree with your assumption of reduced settings...I actually USE the program and don't see any real difference that I can really notice other than I can do a flight without having an OOM....

.

 

Regarding this point, it is more diverse: as I am still not absolutely happy with FSX performance, I desperately hope that FSX:SE performs better and that we would sooner or later know the trick. But I am also very aware that regarding FPS, most Users exaggerate greatly or simply ignore obvious differences or do not take it so exact as you should if you actually contributing FSX:SE better performance. As a matter of fact, besides the performance issue, I never had an OOM in FSX, so this alone would (if I had) already be enough for me to switch, basically. And without offending you: yes, you USE it and you do not SEE any difference, but I have seen so many users in my life not capable of telling the difference while playing between MSAA and FXAA that I really would like to see hard facts and not Users telling me: yes, it performs better. I do not trust anyone, that's also a personal thing...

 

I disagree...if you look real closely, there are less clouds in the FSX shot.  The SE shot has more cirrus clouds and definitely some cumulus on the horizon...FSX does not have anything on the horizon.  If anything, FSX has killed the distance.

 

This somehow verifies my statement above about people seeing things or not... Yes, FSX:SE shows cirrus (not relevant to performance...) and some single cumulus clouds at the image borders. But the cloud draw distance is reduced in FSX:SE. Do I really have to play with MS Paint again.. OK, here we go. I put a red line where the drawing in FSX:SE stops, now tell me, which sim renders more clouds please...

 

 

BTW: this difference is what you see if you reduce cloud draw distance. Again, I believe if SteveW says he used the same settings and the same FSX.cfg, but what ends up on the screen is different. Explanation?

There are some anomalies in the view, like the jetways popping but there's no change in performance. Loading up with unnaturally bad thunderstorms across sim, brought them much closer together, around 5%.

Yes Pete, it's IF10 >augmented. Cheers!

 

Sorry, what is IF10 >augmented?

 

Pete

Sorry, what is IF10 >augmented?

 

Pete

Ideal Flight 10 Augmented Pack. http://www.codelegend.com/idealflight/default.htm