Eric,I'm glad I deleted my post--you essentially said what I was going to say. However, I think you received such a strong response by use of the word "Crap". Seems we can't use expletives in these forums, so we go with second best? The meaning is implied. Anyway, language has been going in the sewer lately in this forum. The word I quoted is one that should be left out of our posts, IMHO. It has almost as much offense as its counterpart beginning with "S".I think the thread has served its purpose--several suggestions have been made, some pitfalls pointed out of each, but I think most who responded feel the thread's author can improve his performance.-John

Hey back at you Scott!It's not the layers that bother me, it's the individual clouds. For example, I get great fps when I'm flying along, but if I look directly at a large cloud (over 4000 feet tall) in vc or spot plane mode, the fps crash. When I look away, all is fine. So, the problem with me isn't the layers, it's the individual clouds when looked at that destroy the frame rate. (But I'm sure the layers would get me if I got a lot of them.) I've got all the hifps clouds that Chris has done, but they still give me that problem. That's why I was asking the original question. I wanted to know if looking at clouds when you were right up on them hurt everyone's fps like mine. And it sounds like it does. Now that I've clarified the question, do any of your answers change?By the way, thanks for all the participation. You guys have some really good technical knowledge.:>

>Now that I've clarified the question, >do any of your answers change?No. :-smoochThere's an option for cloud density - I think that controls how many individual pieces FS uses to represent a cloud. If you crank that back, leaving all else the same, any improvement?

Something's amiss then. As long as I don't go heavy on the coverage (if I keep things 3/8 or less), I can set up heavy t-storm cells topping out between FL400 and FL500, and still I don't get a bad fps hit. My linked screenshot doesn't show all--really it's before/after impact that makes a difference. Even without clouds, fps in those shots would have topped out at 30-32--I don't see the 50 percent falloff you're seeing.I do have one thought, however. Early on in FS9's release suggestions were made to turn off AA in MSFS, and "force it" via your card's settings or through a tool like Rivatuner. I tried that, but in the end I prefer to use default AA. I don't really have any issues with it, other than aircraft which contain mips AA better than those which don't.That's just a thought--don't know if it may help or not. -John

again, they do not effect my system like it does yours, I do not get a fps drop like what you are experiencing.

Which set of Chris' clouds are you using? The dxt3 or 32 bit version? Whichever one you use, try the other one. Some people get better performance with the different sets.

Okay, this is my last post on it...Eric, your right, all those people that have stated they get better performance are lying, they dont know what they are talking about. You know everything, and you are 100% right in everything you say, please, open a church so we may all study your wise ways.You sound like a FSSW supporter who is determined to discredit FE as much as possible, I was told there were a lot of you in the AVSIM forum.believe it or not, FE is better, smoother, more realistic, thats it, people can try it for themselves, some will see an improvement in performance ALL will see an improvement in enviroment over what they are using now.Take care oh Eric the exaulted one.:-roll

rightseat, AA can degrade performance alot depend on your system spect and config. Be sure you have tested all size and format from 3rd party, Some have less performance with DXT, others have better fps with 32 bits clouds, in the 3rd party there is dxt and 32 bits clouds.BestEric

Right guys, FE2004 does use mip / map levels so to say that FSW has better performance by using them is not true.... saying it does worse is also not true.FPS. This is a very subjective thing and depends heavily on your hardware en system setup.The ONLY thing done by making the smaller or DXT3 format clouds, is making the phisical file size smaller thus reducing the amount of memory been used up by the vidcard mem or your phisical ram.What does improve the fps (somewhat) is the use of Alpha channels, which FE2004 does very well!!!So in all - you could say that FE2004 does improve on fps in some situations and in some areas - AND - with an increased amount of stunning visuals.!Cheers guys,Anthony Voshttp://members.home.nl/vos69/logofe.jpg

Hi, "You sound like a FSSW supporter who is determined to discredit FE as much as possible"I use fe only now, but see no difference on performance from before.Eric

Bottom line rightseat (without more FE commercials) is that MS dropped the ball, the weather does kill performance. I have the same setup as Spooky only with a radeon 9800 pro and 2 gigs high end DDR and even with FSW clouds and draw distance as limited as is acceptable to me visually, heavy overcast days will kill your performance no doubt about it (add AI and detailed scenery, makes for a nice slideshow!). While I could use a DXT texture, limit the draw distance more, limit layers with AS or even try FE, you really shouldn't have to in the first place since most of our hardware wasn't even out when FS9 hit the market. And to think the guy at MS that designed the weather system got an award, guess they didn't try flying in it. :-rollThe GMAX SDK has a cloud tool that allows for the creation of clouds, wonder if a GMAX guru could make better ones that still look good!There's always FS9 on the latest rig next year! LOLRegards, MichaelKDFWhttp://www.calvirair.com/mcpics/mcdcvabanner.jpgCalVirAir International

>Bottom line is that>MS dropped the ball, the weather does kill performance. I have>the same setup as Spooky only with a radeon 9800 pro and 2>gigs high end DDR and even with FSW clouds and draw distance>as limited as is acceptable to me visually, heavy overcast>days will kill your performance no doubt about it (add AI and>detailed scenery, makes for a nice slideshow!). While I could>use a DXT texture, limit the draw distance more, limit layers>with AS or even try FE, you really shouldn't have to in the>first place since most of our hardware wasn't even out when>FS9 hit the market. And to think the guy at MS that designed>the weather system got an award, guess they didn't try flying>in it. :-roll>Regards, MichaelI completely agree here, I think FS dropped the ball on a lot of areas in FS, but when no one else is making a good all world flight sim, you can pretty much do what ya want and people will shout for the next one that still has old code in it from the last version. I think FS needs a complete revamp and maybe bring in some of these talented programers that know how to render the world with out a massive fps hit. I only take a fps hit when the clouds are really overcast, really thick, but other then that, it's pretty constant. I lock my fps at 25 and get almost solid 24.9fps. Of course I get dips depending on the area, but it will go right back up after I pass it or look away from it.

Anthony, While your comments are interesting, use of alpha channels is not used to speed up fps in FS2004 (or FS2002). Alpha channels either control transparency or level of reflection, depending on how their use is defined in the aircraft or object which uses them. In the case of clouds, alpha channels help control transparency. My "zero vis" fix here in Avsim's library used them to that effect, although they hurt other areas of the sim due to the cloud bmps I used serving double duty.Also I don't think anyone's suggested FSW's use of mips makes their performance better or worse, unless I missed a post. I do think the participants in this thread all know enough of what they're talking about to know that fps dropping by half when clouds are on is not normal, regardless of the cloud set used.-John

Scott Gridley had an interesting suggestion, "dropping the cloud density settings." I always thought that this affected the ratio of 3D vs. 2D clouds. So if you crank it back, you got 2D clouds all over, if you put it all the way to the right, you got all 3D clouds. Have I been wrong this whole time? If so, then reducing the density would help, thanks Scott. And thanks to everyone else who is contributing. Maybe I should consider buying a 256mb video card to help with the matter?:>

My card in the screenshot example is only 128 megs. Cloud density does not control the ratio of 2-d/3-d clouds. What it does do is reduce the coverage. Try cloud density at 100 pct. with 7/8ths coverage, then lower it and you'll see what I mean. If the problem you're seeing is with minimal coverage (i.e. 3/8ths or less), cloud density won't help much.I still suspect your card may be trying to apply AA to the clouds--but that's just a hunch.-John