Here is version 3 of the Petraeus Index. Thanks to everyone who contributed - PontiusPilotus, Angelique van Campen, and, particularly this time, Sargeski.For the most part the changes are refinements of readings for aircraft that were previously in the index. However two of the new additions are particularly interesting. The Leonardo Maddog shows that there is an FPS penalty for all that functionality; and the Ariane 737-800 X2, although it beats the Wilco 737-400, is certainly no great performer.The figures are broken down for convenience into three sets - airline, GA and other (bizjet, military etc.)Full details of what it's all about can be found in the AVSIM Wiki entry . It's also the place where you can add in your own figures - there are full instructions there. You can either supply your own numbers for aircraft already shown, to give us a larger statistical sample, or add new numbers for aircraft we don't have details of yet. So please pitch in and help this to become the definitive source of relative aircraft performance data. - the better the sample data, the better the overall reliability of the Index. The numbers are relative to the stock Microsoft CRJ700, which is given the value of 100. So something with a score of 33 will typically only give a third of the framerate of the CRJ on your system; something with a score of 150 will typically give 50% more frames per second.AirlinerGAOther

Hi Petraeus,I'm back to FS after being away several years. Thanks for the new Petraeus Index, I understand it is a FPS impact measure for aircrafts addons. I'm trying to get more infos about your index but your link to the AVSIM Wiki entry seems dead.

I'm trying to get more infos about your index but your link to the AVSIM Wiki entry seems dead.

Yes it was, sorry about that, now fixed. :(

Thank you it works now :( At first sight I am suprised that there is so much difference between the RealAir Decathlon (127) and Scout (89) because they are very close in terms of VC and 3D model but I will read more about the construction oh the Petraeus Index.

Petraeus,Thanks for the comparisons. Very helpful.Why is the FSD Saratoga glass listed twice in the GA graph? Should the first entry read "FSD Saratoga analog"?

Thank you it works now :( At first sight I am suprised that there is so much difference between the RealAir Decathlon (127) and Scout (89) because they are very close in terms of VC and 3D model but I will read more about the construction oh the Petraeus Index.

So am I! Maybe the graphs refer to different versions for FS9 or FSX, but the panel detail, textures and polygon counts in the Scout and Decathlon are almost identical on both aircraft and empirically they show precisely the same performance. Also I don't understand why the Marchetti is lower. Ask our customers and they will all tell you the Marchetti somewhat out-performs the Scout package consistently in frame rates. I think this guide is very useful...but more so if it reflects what everyone sees and experiences in practice. Nevertheless it is a useful broad measure.Cheers,Rob Young

I have all the realair stuff except the Spitfire (not my kind of flying)I'd say the Scout and SF260 are in the same neighborhood, whereas the Duke is a little more hefty on the FPS load, but of course it's more complex, and I'm sure it has more polys.They are my favorite GA's :)

What NASA systems are they running?

So am I! Maybe the graphs refer to different versions for FS9 or FSX, but the panel detail, textures and polygon counts in the Scout and Decathlon are almost identical on both aircraft and empirically they show precisely the same performance. Also I don't understand why the Marchetti is lower. Ask our customers and they will all tell you the Marchetti somewhat out-performs the Scout package consistently in frame rates. I think this guide is very useful...but more so if it reflects what everyone sees and experiences in practice.

The Marchetti SF260 was the first plane I purchased for FSX several years ago! It's always been top performer... smooth, fluid fps and beautifully rendered gauges.Btw love the way you implemented click and drag technique to rotate knobs in the VC. Makes it so easily to manipulate gauges when the VC is bumping around. I wish more vendors would follow your lead.

Nevertheless it is a useful broad measure.

Yup, I'd agree with that, rather than imagining it is a unequivocal guide to what will happen on a specific computer. I think it is likely that different RAM configurations and graphics cards would tend to throw up different performance benchmarks on different machines, primarily because not all aircraft would use the same texture optimisations, and the sophistication of the avionics on some add-on aircraft might make them struggle on a computer that has limited graphics or RAM and which has to dump data to load everything; something less likely to occur with a default FS aircraft on machine with limited specs.As a broad guide it has its uses and is certainly better than nothing, but it's a stretch to imagine that it would always indicate the relative performance of something on every machine.Al

I have to agree with Al. The "index" is useful, but also very misleading. And from a statistical and reporting perspective, contains significant errors.The Wiki says all the aircraft are "compared" to the default CRJ, then given a percentage rating on FPS performance in FSX. Well, if that is the case:Why is the Flight1 Cessna Mustang shown as "45%" on the GA chart, and "73%" on the Other chart? There is no explanation in the Wiki for this. As shown in the charts, the Mustang is being compared to the "relative performance" of ALL aircraft contained in a "given category" (GA or Other categories), and not solely to the default CRJ. If it was being compared ONLY to the default CRJ, it would NOT be listed in two different categories, with TWO different percentages that are 28 percentage points apart.A difference of 45% vs 73% is quite significant, and an indication of an error in the process used to determine the results of the index.FalconAF

If it was being compared ONLY to the default CRJ, it would NOT be listed in two different categories, with TWO different percentages that are 28 percentage points apart.

Simple explanation for that - the Mustang was erroneously shown in two categories, rather than just the one, with a subset of results in each case. It has now been consolidated into the one entry, in 'Other', with all the bizjets, where it properly belongs. Thanks for pointing that out.Also thanks, UlfB, for pointing out the duplicate FSD Saratoga entry. Simple finger trouble with the MS Excel Sort function, hopefully there are no more gloops. :( It's getting to be quite a big database these days.I'll show the corrections in the next version.The differences between Decathlon and Scout are surprising, but they come up on more than one system. If people can send me more measurements of these two than we can get a more precise fix.Overall, I would be the last to claim that the values in the index are absolutely definitive. Until I get 1000 readings for each plane, to iron out the vast range of system variabilities (and Heaven forbid, I have a life as well!) it will be just a "general indication" and nothing more. Having said that, all the results tend to come into the same ball-park - the good performers are consistently measured as good performers, the bad performers and the middle-of-the-roaders similarly.

I have to agree with Al. The "index" is useful, but also very misleading. And from a statistical and reporting perspective, contains significant errors.The Wiki says all the aircraft are "compared" to the default CRJ, then given a percentage rating on FPS performance in FSX. Well, if that is the case:Why is the Flight1 Cessna Mustang shown as "45%" on the GA chart, and "73%" on the Other chart? There is no explanation in the Wiki for this. As shown in the charts, the Mustang is being compared to the "relative performance" of ALL aircraft contained in a "given category" (GA or Other categories), and not solely to the default CRJ. If it was being compared ONLY to the default CRJ, it would NOT be listed in two different categories, with TWO different percentages that are 28 percentage points apart.A difference of 45% vs 73% is quite significant, and an indication of an error in the process used to determine the results of the index.FalconAF

The F1 Mustang 1.02a patch cut my frame rate in half. Weird!

Could you perhaps correct the "Aerosoft Long-EZ" to "AlphaSim Long-EZ"?Would be appreciated... :(

Hi Petraeus,It is not easy to get remarks as well as it is not easy to carry out a sensible project like this.However I believe your idea is relevant and I am willing to feed the Index with measurements for a few payware aircrafts I own.But to avoid multiple/useless measurements, I have a few questions:

To get the Petraeus Index measurement for plane xyz on your system:- go to Edwards Air Force Base (KEDW) and line up on runway 4R at noon.- set AI and autogen to zero, set weather to 'Clear Skies' with no changes over time.- set FPS to 'Unlimited'.- set VSYNC to Off (under graphics card settings)- get all the engines and avionics running.- look out of the front of the VC, and after about 5 minutes, note the average FPS figure*- load the stock CRJ700 into the exact same situation, with engines and avionics running.- again, after 5 minutes note the average FPS.- Go to 'discussion' for this Wiki entry, and add in your name, full details of xyz's aircraft model and developer, the FPS for plane xyz, and the FPS for the CRJ700. We'll incorporate your figures in the next version.It's as simple as that!

1) Edwards Air Force Base is in the desert then I assume that scenery settings (complexity, mesh resolution etc.) are neglectable? If so for clarity and to give people confidence in the Index, maybe you can state it in the wiki.2) Once the (monthly?) Index is published, the 'discussion' for the Wiki entry http://linux.myalbemarle.org/wiki/index.ph..._Petraeus_Index is wiped out? No tracability?3) To avoid 'spam' and let the audience judge the realiabilty and trustworthiness of the measurement for a given aircraft, IMHO you should indicate not only the result of the measurement but also the number of people who sent their reports. And the method of calculation (eg: a score for an aircraft is the mean of all measurements?) should be stated clearly in the wiki.Please forgive me if this has been discussed before but unless I mistake I see no such reference on the wiki.