About the naming of the aircraft being tested... I'd recommend you be more specific about "F1 Mustang". They offer both a P-51 Mustang, and a Cessna Citation Mustang - two very different aircraft, and I'd be surprised if they didn't offer very different performance as well.

Ed,When this was done before, specific instructions were given for FSX configuration. The testing was done at a specfic airport (I believe it was Edwards), clear weather theme was used, all AI traffic was to zero, no weather utilites used, there were specific terrain settings to be used etc. It's not as if everyone did whatever they wanted and then results compared. The atttempt was made, as best it could be, that all involved were testing in the same FSX environment.Regards,Bob

Nevertheless... different systems will handle 3D terrain and such better than others. To truly gauge the aircraft performance, you must eliminate all other items that affect overall system performance.

About the naming of the aircraft being tested... I'd recommend you be more specific about "F1 Mustang". They offer both a P-51 Mustang, and a Cessna Citation Mustang - two very different aircraft, and I'd be surprised if they didn't offer very different performance as well.

You're right, I completely forgot about their P51The one in this chart is the C510 Mustang...

Nevertheless... different systems will handle 3D terrain and such better than others. To truly gauge the aircraft performance, you must eliminate all other items that affect overall system performance.

Oh I agree, for a more error-free gathering of data, but I think what the original designers of this theory wanted was a generalized performance ratio. We all know FSX is fickle with regards to hardware, there's so many different variables.

Which Citation X? v1.0 or v2.0?Also... system specifications?Posted in his signature!You're leaving out important information. :unsure:Things I know affect frame rates:1 - Scenery complexity. For testing all scenery settings should be at absolute minimum.I have to disagree here. Static Scenery Complexity, should be set to what he normally uses. Those settings should be disclosed though. The results won't get skewed, if the settings are the same for each aircraft tested and would reflect his normal use. 2 - AI. For testing all traffic settings should be at absolute minimum.Absolutely all non static options should be off, including AI and weather.3 - Drive access. For testing accuracy non FPS measurements should be made until FSX has finished loading data.If you have the capability, by all means. 4 - CPU/Video card. These two items can have significant impact in performance results.Specified in his specs in his sig.5 - Background tasks. These can also impact frame rates significantly.Nothing to add here!

Ok... I just did a quick run of the default 737 and the Citation X 2.0... not because I'm concerned with representation of it's performance... but because I happen to have it. :wink:I have an AMD Athlon Dual Core 4050e at 2.10ghz.For the 737 I got 22fps. For the Citation X I got 17fps. That means the Citation X is 77% of baseline on my system... which is significantly different than the baseline percentage being shown in the OP. If the test process is correct, the variance between his results and mine shouldn't be so widely divergent.As I've tried to state through this topic... this 'index' only shows how well a given setup can run a specific set of aircraft. It's leaving too many external factors in the data to show how well a specific aircraft actually performs.If you think I'm wrong... backup your FSX.cfg... turn off ALL scenery, weather, traffic and do the test with both aircraft that way.

Ok... I just did a quick run of the default 737 and the Citation X 2.0... not because I'm concerned with representation of it's performance... but because I happen to have it. :wink:I have an AMD Athlon Dual Core 4050e at 2.10ghz.For the 737 I got 22fps. For the Citation X I got 17fps. That means the Citation X is 77% of baseline on my system... which is significantly different than the baseline percentage being shown in the OP. If the test process is correct, the variance between his results and mine shouldn't be so widely divergent.As I've tried to state through this topic... this 'index' only shows how well a given setup can run a specific set of aircraft. It's leaving too many external factors in the data to show how well a specific aircraft actually performs.If you think I'm wrong... backup your FSX.cfg... turn off ALL scenery, weather, traffic and do the test with both aircraft that way.

If there is that much of a variance between no scenery options and scenery, then that invalidates the whole premise of the test in either case. If anything it makes it more important to test with current settings. Who cares if you get less or more of a variance without scenery, that is not how you use the sim normally. I do believe though in this case the test can only be used as an individual benchmark, not as a compilation of user results.

For the 737 I got 22fps. For the Citation X I got 17fps. That means the Citation X is 77% of baseline on my system... which is significantly different than the baseline percentage being shown in the OP. If the test process is correct, the variance between his results and mine shouldn't be so widely divergent.

Thanks for testing! Yes there will be a variety of responses, such as the guy above who said his C510 Mustang is much worse on fps than his Citation X. Be sure all the avionics are loaded because without the INS aligned and the FMS's up and running the Citation gets really great performance on my system.

Thanks for testing! Yes there will be a variety of responses, such as the guy above who said his C510 Mustang is much worse on fps than his Citation X. Be sure all the avionics are loaded because without the INS aligned and the FMS's up and running the Citation gets really great performance on my system.

I get a much greater variance with the Mustang, (With the avionics fired up)Boeing 737-800 = 77 fps - Note: My default 737-800 aircraft now uses ISG gauges, so that could influence the result here.F1 Mustang = 30 FPS.For 39% of baseline.Edit: Just tried it with the default 747-400 (no ISG gauges) and the variance is even greter with a 90fps baseline for 33% of baseline for the Mustang. That's at Edwards Runway 4m with an E6850 @3.0Ghz, 4GB mem, and a 9800GTX card. Settings are Extremely dense scenery, dense Autogen Water 2Xlow, Texture 1m, mesh 10m, clear weather, and no traffic of any kind.

Ok I ran this test with the settings I posted above with my complex aircraft, and this is what I got, using the 747-400 default as baseline

Default 747-400                     90 (Baseline)Wilco 737 PIC                       42Wilco Airbus Vol 1                  45Ejets 1                             39Ejets 2                             38CS 727                              37CS 757                              32CS 757 Block B with LVL-D merge     38CS 767                              34CS C-130                            50Ariane 737-800                      33PMDG 747-400                        25PMDG 747-8I                         30PMDG MD-11                          36Level-D 767                         33Flight1 Super 80 Pro                52Flight1 Cessna Mustang              30Realair SF-260                      75A2A P-51D Mustang                   106

Now to throw another wrench into it, how are you all measuring the framerate? If your using the FSX frame rate counter you'll definelty have a wide variance in FPS. Again going back to the orgianl index, FSX FPS set to unlimited, FRAPS was used, FSX ran for a specified time period, min, max, and avg FPS were captured with FRAPS, and then all the results that were submitted were added to the overall averages of min, max and average framerate. There were meterics put in place with the original Petraeus Index. Everyone attempting to monitor performance however they see fit is pointless.Ed,I'm clearly seeing your stance now.

Now to throw another wrench into it, how are you all measuring the framerate? If your using the FSX frame rate counter you'll definelty have a wide variance in FPS. Again going back to the orgianl index, FSX FPS set to unlimited, FRAPS was used, FSX ran for a specified time period, min, max, and avg FPS were captured with FRAPS, and then all the results that were submitted were added to the overall averages of min, max and average framerate. There were meterics put in place with the original Petraeus Index. Everyone attempting to monitor performance however they see fit is pointless.Ed,I'm clearly seeing your stance now.

What I did was use FRAPS, and let it settle a couple of minutes before recording the result for each aircraft. The best way is to use the Fraps benchmark option for a set period of time, and record Low, Max and Average FPS. There was another benchmark method posted a while back that I thought was even better. It was a saved flight over Seattle flying a predefined flight path. You'd set FRAPS to benchmark for 5 minutes and record the result. It was only used with the CRJ though.

Now to throw another wrench into it, how are you all measuring the framerate?

You're right, I completely forgot about FRAPS. I was using the built in fps display, vsync set to OFF via nHancer, and I took the mean after waiting about 3 min for the variance to settle down.I'll try to add in my other planes that are sitting idle in folders...I still say this is much better than developers saying "I get 50 fps over rural farmland at 30,000 ft." There's nothing to be made of this, because you don't know exactly what their settings are. If you ask them to report to you their fps VS the default B737 (in the exact same situation) then at least you have some sort of guess as to what your performance may be. Clearly, as shown in responses here we're not going to get the identical responses, Tom gets 33% fps of the baseline in his mustang. I get a little better. But, the fact is, we both take quite a hit from it. The earlier RealAir stuff has been fairly FPS friendly, and Carenado, without a doubt is almost better on FPS than stock aircraft. These are the generalizations I'm trying to see through the Index.I must give kudos to Aerosoft - they recently displayed a series of screenshots with their new Airbus compared to default in various situations, FPS displayed at the top. That, imho, is a good indicator of performance because they are comparing fps against another plane that every user can relate to.

There was another benchmark method posted a while back that I thought was even better. It was a saved flight over Seattle flying a predefined flight path. You'd set FRAPS to benchmark for 5 minutes and record the result. It was only used with the CRJ though.

Yep, Gary's FSMark07 (RESET MCP ALT username), that was a fairly good benchmark. Everything was pretty constant (a saved flight the user loaded), unless you had addons. I liked how he did that.