Corrigan

Yes, I know. I'm saying that the 980 doesn't "nudge out" the 290X---it beats it comfortably, to the degree where the card you should be comparing the 290X to is the 970 rather than the 980.

Sure, but I dare say Novak could add to this discussion. He probably knows if there is such a tendency in DCS, and he obviously knows much more than we about how he's tuned the MiG high-alt performance, if it behaves as he's set it, etc etc. I'm not sure how we can proceed by our selves; we've found that there are problems with the cruising ability and service ceiling, but I don't know how we can investigate further.

That's due to the excessive LoDs they've used for the external model. It's known and probably a good way towards being fixed. :)

I think this is tied in with their plans for EDGE. The sound engine will be revamped then, and I suspect that file infrastructure that does not seem to work now is in place for when that time comes. In that case, I feel I can live with the generic placeholder sounds for a few months.

Honestly, I think the most useful thing right now would be for the MiG devs to chime in. Some inside info would make much of this speculation obsolete and we could concentrate on more constructive things.

Only that's not true. The 290X is actually closer to (and above) the 970 in the bechmarks/reviews I've seen.

I think he means that DCS has a systematic mis-modelling of high altitude performance? If that was the case, I can imagine how it would be hard to get low- and high-alt. characteristics simultaneously correct, regardless of aircraft.

It can't possibly be aerodynamics, that would be utterly unphysical. I was also surprised at what you're talking about. I assume that giving rudder input actually differentially brakes the undercarriage, but I'm not sure that it's working as intended. Was, like you, expecting having to brake with the lever and simultaneously give rudder to get ANY turning at all.

Also bear in mind that nvidia control panel settings can have a big impact!

Wow, TSSAA? Try turning that off.

I'd expect more from your system, but all that info is kinda incomplete without mentioning resolution. Also, is your CPU overclocked?

Saw that pic on Reddit. There's something iffy about it. Look at the top of the HUD frame, looks like a cut-and-paste job.

Even the DCS manual speaks of "altitudes above 18000 m" on p. 33. So I tried getting there. My setup: clean aircraft, starting with as little fuel I dared (25%), no gun ammo. Default mission editor atmospherics. I used reheat literally all the time, trying to zoom climb. I got to 15 000 m, but here I couldn't get past 450 km/h IAS in full afterburner in even flight, so I had nothing with which to zoom. Now I had only 350 L or so left, so I made a valiant attempt to convert the last of my airspeed and fuel to altitude. Flying on fumes, I finally stalled at about 16 600 m. Something is clearly very wrong with the service ceiling as well. Track is attached. EDIT: I also tried a quick test with the "Simplified engine management" option checked: no difference. ceiling.trk

Are you saying the MiG shouldn't be able to fly there? Table 3, p. 12 in this manual has descent data from 18 km. Or am I misunderstanding you? :)

It was a shadow passing them. You can't see the backlighting in direct light.

If I tell Nvidia you sent me, what will they do for me?

Well, you can perhaps show us a track where you don't ever move the stick to it's full extent. Try flying for a minute only using half of the available travel in every direction. If you don't have any trouble doing this, I think we can conclude that your controls are set up properly.

Here's a track which I find problematic: I have 60% internal fuel, the small centreline tank and 2 R3Rs. I start at about 10000 m, immediately go into Recovery mode, and use reheat up to about 500 km/s IAS. I then come out of reheat and just wait. After like 3-4 mins I'm down to 410 IAS, 12 deg AoA. Shortly after I start to lose altitude. The track ends about 7 mins after I started with me doing 370 km/h IAS, at 9500 m and falling, still in recovery mode. Clearly, it's not possible to cruise in this configuration at this altitude. The question then is, should it be? I'd that everything I've read indicates yes. 10kcruising.trk

Well, like I said before, I think we all appreciate the time. Getting back to your 10500 m cruising track: again, I worry about your fuel load. You're at 800 L even before you drop to 10000 m. The aircraft isn't tactically useful. I worry that it's an unrealistic scenario.

So, Viper, can I ask your impression? Do you think there's something wrong with the flight model? Or are you not convinced? Or do you think there's a slight problem but it's being exaggerated?

And as I mentioned, just after you start the climb, Viper, you're doing 930 TAS. Again, not to be a dick, but I think you would have done much worse if you had stuck to the manual's prescription.

Yeah, but it should take you just under that to 11000, which instead takes you 13 minutes.

I agree. Although if you look at the last part of my post which I just edited in, it's clear that the climb the manual is talking about is much much faster than the ones we can do (including your track).

Quick question: how does one access the flight recorder data? Watching your track now, Viper. Really not trying to be a dick here, but you start climbing at much more than 870 km/h TAS (as per the manual). I imagine that's what makes the difference. Also, if you look at the track, you can see that you weren't actually able to hold a steady climb at that airspeed. Looks to me (as a layman) like a series of zoom climbs. But maybe the manual takes some of that into account? I'm not very experienced with technical literature in this field. EDIT: Also, you have a much lighter aircraft than the other tests, looks like 1800 L of fuel. Not 100% sure what the manual refers to, but it doesn't really make sense to take off in a short-range interceptor with half fuel. Regardless of fuel state, the manual says a climb like the one you did, or tried to perform, should take 8:50, and I think yours took 14-15 mins.

I think that's intended. I have vivid memories of hearing PULL UPP, PULL UP in the A-10 while watching something else in external view, at least. EDIT: then again, the low altitude warning in the MiG-21 isn't audible in F2 view, so it's a bit inconsistent.