So, I did some testing of static weather vs dynamic weather because I’ve heard there’s a performance difference. I would get these occasional stutters that were driving me nuts. Just flying along and boom STUTTER with fps drop by half sometimes. Turns out it’s static weather. Dynamic weather runs beautifully even with heavy clouds.
Here’s a light mission with only 45 units flying in the F/A-18 at 25000 ft over SOH. Heavy cloud cover with static weather settings. Each one of the dips in the graph represents in-game stuttering.
DCS is glorious running at a locked 60 fps with no stuttering. Dynamic weather is a pain to setup but it works much better with current weather engine than static weather. Hopefully ED will fix this at some point but I’ve heard nothing about weather in over a year so I wouldn’t expect anything soon. A dynamic weather profile generator would be a nice 3rd party app project for DCS. Right now you have to manually edit the dynamic weather .lua files and it’s not pleasant.
I did some more testing with this to narrow it down a bit. The problem is with cloud level 5 - 8 when set with static weather. The same clouds set with dynamic weather run perfectly smooth.
That is really odd. As in, the same clouds when static cause a glitch? Might be worth reporting the findings at the ED forum or as a ticket, as that sounds like bug territory?
I posted about this the ED Forum under DCS World 2.5 > Bugs and Problems > Game Performance. However, I don’t see a lot of responses from the ED team under that sub forum so I’m going to open a ticket on it.
It definitely appears to be a bug that is easy to repro.
I too like to fly with a few clouds and noticed similar sudden drops of performance despite the rest of the sim running smoothly.
It even happens in a completely empty mission.
I almost never used dynamic weather, but I will definitely try that and see!
Do you have an example or good how-to of that, as kinda curious? I think there were static METAR weather generator utilities for DCS before, but wonder what the dynamic weather input looks like?
Notepad++ works nicely for editing the .lua files. Basically you create a new Dynamic weather profile in the ME and save it. Then you go back and forth between ME and Notepad++ editing to get the size and positioning of the weather systems where you want them. It’s a pain for sure, but the results are pretty nice.
The dynamic weather part is only about 30 minutes so it’s not that bad.
Also, I did some more testing and can reproduce this in the “Stable Release” version 2.5.3.24984 so I suspect this issue has existed for some time. I have a ticket open with ED so hopefully they take a look at it.
Here’s my test flight on Stable Release version 2.5.3.24984. Same 7 minute flight I’ve been using to test with static clouds set to level 8.
The combatflite tool has a user interface for setting up dynamic weather. Just choose dynamic weather under settings, click and drag to set up the cyclone system (under objects) and then export to .miz for flying (or more mission editing). No hassling with Lua’s.
That is awesome, I’m going to play around with it later. Does anyone know if there’s a way to set the base elevation of the clouds with dynamic weather? There’s a base = ### line in the Lua file but it doesn’t appear to do anything when setting it.
OK, I watched the video (the first 30 min) and I see how things work…where L’s run into H’s you have a weather front, stronger winds & clouds. I also get the idea of varying atmospheric pressure and its effects on A2G weapons (particularly the Viggen for some reason).
By why does it work better than static weather?
I would think that with the static weather options, you can fine tune the WX for the mission without any outside app.
…and, quite frankly, the WX systems modeled in the video were really small, I assume for purposes of the video.
Realistically, wouldn’t you be working with two to three much larger pressure areas?
I’m off now to play with if/how inputting Static WX data effects the Dynamic WX.
Static weather in DCS is currently bugged. You may notice with static weather while flying a mission that clouds will randomly disappear. This normally happens after flipping through the F2 view of other aircraft.
The main issue with static weather is it appears to be causing stuttering with cloud level set to 5 - 8. Why this is, we don’t know yet.
The nice thing about dynamic weather is you can experience a variety of weather conditions while flying your mission instead of just one that has randomly disappearing clouds. Yes in the real world systems would be larger, but to get good results in DCS I recommend using smaller systems and experimenting to get the results you want.
One thing I’m trying to tweak is the visibility of clouds when using dynamic weather. Right now when flying over a strong low pressure system you get a large square of heavy clouds that visibility doesn’t extend to edge of the horizon with so it looks funny. The cloud base level is also set to a static level of around 10,000 ft. I bet this can be tweaked through the config files but I’ haven’t figured out how yet.
I did figure out how to adjust the cirrus cloud level. It’s default is 42000 ft which seems a bit high to me. They look better at lower elevations. You can also adjust how often they fade in and out which the default is also unrealistically set to every few minutes. It rotates through 4 different low-res cirrus cloud textures.
Great info. I hadn’t noticed stuttering when the clouds are 5-8…but then I usually go for fair WX or completely overcast.
So in the level of the heavy clouds, I assume that visibility shouldn’t extend to the horizon…so I assume you are looking at this from an altitude not in the clouds…I am used to the FSX weather generator where visibility and cloud level are independent…which really doesn’t make much sense. The trick is to setup the visibility distances so that they correspond to the type and degree of cloudiness and make sure the altitudes (top & bottom) for the diminished visibility are “inside” of the cloud layer. If you don’t do that, or if you use real WX, you get a hard line at the top of the diminished visibility which looks silly…sometimes you see hills sticking through like islands.
Yeah…that wouldn’t work for me. I’d want a storm front to be just that–stormy–with reduced vis and low clouds…if you can figure it out let us know!
I have disregarded the dynamic weather for a long time because it used to have a fundamental logic flaw. Wind speed was directly proportional to distance from the center of the pressure system (constant angular velocity of the complete system). This always lead to insane wind speeds and is not at all how wind works. Is that still the case?
From the video, that sort of appeared to be the case - distance fro center and where the fronts collide.
So setting static WX as a start evidently does nothing. I was just looking at clouds… My thought process was that Dynamic WX would take the cloud setting from Static WX…maybe where theWX was the worst, collision of H & L fronts would be the clouds from Static WX…yeah, I’m over thinking this.
I set clouds at 7 and an altitude of 400m (I’m doing Mi-8 stuff…call it 3 ft to the meter; 400 x 3 = about 1200 ft) . Then I went back to dynamic, saved, and took a look. Where the fronts met was a few clouds.
If this is the case, and there is no in ME tool to set H & L areas size/location…I’m not enamored with Dynamic WX.
My gripe with the dynamic weather is not being able to set the cloud base level. I’ve been through every config and .lua file I can find and haven’t found a solution yet. There’s got to be a default level set somewhere but I can’t seem to find it.
That is a non-insignificant gripe. DCS features several aircraft that are considered “All Weather”…so a cloud deck of 500ft with low visibility below it, makes the mission that much more challenging.
Sure you can do it in Static WX… but then you are taking off and landing in the goo. You might want to be launching from and recovering to a carrier way out at sea where the WX is fair, to hit a target that is socked in. Right now we cannot do that.
People say many “a discouraging word” about FSX. However, its WX system (for creating the WX, not the depiction of clouds and stuff) is pretty good. You can do essentially a DCS Static WX but with several cloud and visibility layers. You can also make/edit a Weather Theme to simulate a weather front, again with clouds/vis at several levels and in different areas. Essentially, it is also a static WX system. However, you can go from Good WX to Bad Wx and back…which is really the whole idea.
I miss the real-world weather, which iirc updated every 15 minutes from Jeppesen. The change was abrupt, so that could be improved but it would be nice to have missions where the weather is different each time you play it too.
So in the level of the heavy clouds, I assume that visibility shouldn’t extend to the horizon…so I assume you are looking at this from an altitude not in the clouds…I am used to the FSX weather generator where visibility and cloud level are independent…which really doesn’t make much sense. The trick is to setup the visibility distances so that they correspond to the type and degree of cloudiness and make sure the altitudes (top & bottom) for the diminished visibility are “inside” of the cloud layer. If you don’t do that, or if you use real WX, you get a hard line at the top of the diminished visibility which looks silly…sometimes you see hills sticking through like islands.