Hey gang...Ok, it seems like there's an aweful lot of confusion and conflicting information about what sort of performance people are getting with FSX. This is compounded by the wide variation in system components, FSX config settings, flight location, etc.What I propose is the following (and I'm looking for suggestions on how we could do this....)1) generate some sort of "benchmark" (for those FPS players out there, think "timedemo") flight that users can download and have FSX run. Basically make it a straight and level flight... or maybe a 15 minute run that starts out away from a city and then does a full fly to and approach and landing at an airport. (Is there a way to capture a flight like this to a file and have the sim refly it automatically?)2) a series of standard CFG files at various levels of detail settings. That the users could also download and use on their machines.3) Allow the user to use the downloaded CFG file, and the benchmark and we somehow capture the FPS they are getting (think something like the way FRAPS does it... Is there a way in the SimConnect API to capture and log FPS???)4) After a benchmark run, the user posts into a web form his FPS log file, and the form also captures which CFG file he was using, what his CPU, MB model, RAM amount + speed, video card(s) w video memory, hard drives, etc. were in a standard format that everyone uploads in. Items #1 and #2 locks down the variables so we can directly relate the FPS performance to the settings used and the hardware configuration. I think doing something like this would greatly help the community track down what is causing the slow downs, what hardware seems to help it the most and where to look for fixes in the system.What do you guys think?--2002cbr600f4i

I've suggested a similar idea on one of these threads. Great idea but it appears a majority would rather spend their time whining instead of doing something constructive. Craig

Yep you have to face facts some are he-LL bent on the downfall of FSX.All I have is the Demo 2. I have seen the FPS go as high as 161 down to 6. But with the tips and tweaks I have seen I can now keep the demo in the high 30s with maxed air traffic and high terrian settings. No bloom. Cars and boats at 50%My system is average nothing fancy3800 [email protected] ramATI x1800xl 256mb video card. 80gb sata hdI think that the AMD cpu single core maybe getting better results then the dual cores.

I think that about the dual cores too. I've had good results with the beta and have a sys similar to yours. C

>I think that the AMD cpu single core maybe getting better>results then the dual cores.I think your absolutly correct. I've been reading the reports of performance today and the issues some are having with dual core CPU's, so I went to Newegg and decided to build a nice, inexpensive single core AMD FX-55 CPU that comes with a Seagate 250GB 7200RPM SATA drive for 200 bucks. Throw in an ASUS MB and 7900GT and I'm out $520 bucks. Combined with the the hardward I already use I suspect I'll get decent performance for not too much money. Other than BF2 I don't use the machine for other programs or games, so I don't really need a dual core anyway....(Sorry to take the thread off topic. I'll help with data when I can :-) )

With all due respect, forget the benchmark. I have FSX Deluxe and have had it for almost a week. FSX will fly well dependent on the situation more than ever. For example: when bush flying in the Beaver I was able to max out the water AND autogen both mesh and texture resolution full right with decent frame rates. If I tried those settings on the tarmac at KSEA (water and autogen) it would crush my frame rates.My point is that you will find more than ever a wide variation of user settings dependent on the context. It would be nice if FSX had a preset save file for settings which could be loaded up with the given scenario.One thing I can tell you as a certainty as far as my system is concerned and I dare say on most other systems: Shut that Light Bloom off. It's a HUGE frame sucker.

>With all due respect, forget the benchmark. I have FSX Deluxe>and have had it for almost a week. FSX will fly well dependent>on the situation more than ever. >>For example: when bush flying in the Beaver I was able to max>out the water AND autogen both mesh and texture resolution>full right with decent frame rates. If I tried those settings>on the tarmac at KSEA (water and autogen) it would crush my>frame rates.>>My point is that you will find more than ever a wide variation>of user settings dependent on the context. It would be nice if>FSX had a preset save file for settings which could be loaded>up with the given scenario.>>One thing I can tell you as a certainty as far as my system is>concerned and I dare say on most other systems: Shut that>Light Bloom off. It's a HUGE frame sucker.>It's the same here. I like to do a lot of mountain flight, and water & auto-gen can be full right, but has to be cranked down for other situations. A "save" setting loaded automatically to saved flights would be great. Maybe it's there already, and I just don't know?

You can save different configurations in FSX.

You mean something like this? http://forums.avsim.net/dcboard.php?az=sho...ing_type=searchThis was in the Demo 1 era, which raised little interest because of the pre-release nature of the first demo. Now that the release version is out there (or at least coming soon), maybe it is time to dust it off and come up with a few extra recordings to cover the variety of areas the full release covers.Please note, I propose this benchmark with the aim of using a common configuration baseline to find out what hardware does and doesn't make FSX tick in its release configuration. The does make it tick list will be useful for those contemplating new system components for the current FSX iteration. The does not make it tick will be useful information for MS and the user community alike to collaboratively help make things right. This is NOT an MS bashing exercise.Let me know if ya'll want me to update this for comparative purposes with the FSX release version.Gary

Great idea except most of the folks in the forum would rather complain than actually take the time to approach what they perceive to be a problem using any kind of objectiviy. Makes sense to a few of us but from what I've seen in the forums so far we are a big minority. Can't prove it but I think most of the complainers are from the "instant gratification" generation. Craig PS. I dont have anything but the beta to benchmark yet or I would jump in there.

Well here will be their chance to put up or shut up!Gary

I don't know how you can save configurations. I tried saving a flight with certain settings. I then saved another flight different settings. When I went back to the main menu and loaded up the first flight the settings which were used came from the last settings used. Therefore, although it's saving the plane, location, time, etc. the settings are not saved.

If you go the the "Settings" menu your will see "Save" and "Load" buttons at the bottom of the menu. It's that simple. Your not saving a flight; your saving a "Settings" Config file. Craig

Len -Actually, you select the Save configuration option.Specify a name (like test.cfg) in the window, and the .cfgfile will be saved.We can then post that and anyone can use it.JerryG

I've been going through all the major forums and reading what people are saying about performance. It's perplexing to see people with mid range systems saying they're getting decent performance with "most sliders to the right", while others with bleeding edge systems are reporting abyssmal performance with minimal details.What's going on here? I'm sure for the veterans it's a no brainer to have one's computer optimized, with the harddrives defragged regularly and background processes kept at bay, but I wonder if these *are* contributing factors in the differences we're seeing.I propose that those of you who are getting good performance share your practices and philosophy regarding behind the scenes stuff. Are you using a dedicated harddrive with nothing but FSX on it? Do you have your system configured to not use unnecessary services (or use FS Autostart to create that kind of environment on launch of the sim)? What software do you run in the background, if any, such as antivirus, temperature monitoring, etc? Are you overclocking? I agree completely with standardizing some benchmarking scenerios, such as location, time of day, and weather. KSEA would be a good place for detailed city/airport, as it's not a monster like New York, but does feature various details including mesh, water, buildings (scenery specific, not just autogen) and so on. As for a non-detailed area, perhaps somewhere in the midwest. Oklahoma City or similar. The point, which I'm sure the original poster is trying to get across, is that everyone's reporting is so varied that it's hard to make sense out of it all. If everyone who is willing to participate in this kind of testing would agree to some universal conditions, it would go a long way towards providing something useful to the community and ACES from which to go on.Once I get my copy on the 17th, count me in on anything decided.