[CORRECT AS IS] Odd rocking in G-onset

[Hummingbird]

@BIGNEWY Do you know wether there's been any progress on this internally ?

 

Just curious.

[Hummingbird]

[Hummingbird]

So it's been 8 months since this was first reported, and over 3 months since @BIGNEWY said they would look into it, but since then we've neither seen nor heard about any change.

 

The FM really is the most important aspect of a flight sim, hence this ought to be top priority, but considering the time it has taken so far, without a word in between, it obviously hasn't been.

 

Communication on the part of the developers regarding the matter is long overdue.

[BIGNEWY]

Hi all,

 

Sorry for the delay we were checking the numbers and waiting for a F-16 pilots feedback.

 

The G-onset is correct as is.

 

As AoA increases, the G onset will slow down around 7.5 to 8.0. This is normal.

 

-3 G should be possible and that we'll address this; however, -3 G would never be used by a real world pilot.

 

Thanks for being patient.

Edited July 24, 2020 by BIGNEWY

[Hummingbird]

I slows down way before that though BIGNEWY, around 6 to 6.5 G's, and abruptly so, not smoothly. Furthermore it gets esp. bad at high speed in the DCS F-16, where G onset rate is incredibly slow and reaching 9 G's is a struggle (usually you only get 8.8-8.9)

 

The progression should be like in that other famous F-16 sim verified by F-16 pilots, where onset rate is very fast and reaching 9+ G's doesn't take ages.

 

So I'm very worried to hear that this is considered "Correct as is"

Edited July 24, 2020 by Hummingbird

[BIGNEWY]

We have checked the numbers and had a F-16 pilot review it.

 

We will be looking at the negative G issue however.

 

thanks

[Hummingbird]

We have checked the numbers and had a F-16 pilot review it.

 

We will be looking at the negative G issue however.

 

thanks

 

 

So the real F-16 struggles to hit 9 G's, and takes a long time to do so? That news to me, and not what I've now had several 16 jocks tell me.

 

A lot of people are going to be very sad to hear this.

 

To be clear: Atm for the DCS F-16 G onset rate is fast below the speeds where 9 G's can't be attained, but as soon as a speed where 9 G's can be attained is reached (high speeds) the G onset rate slows significantly, and you get odd stalling in onset rates at around 6.5 G's after which it then starts picking up again only to slow to a crawl around 8.5 G's. I've never seen a single HUD footage show such slow & inconsistent G onset rate for the F-16, and it doesn't occur in that other F-16 sim either where G onset rate is fast & smooth irrespective of speed.

 

Something aint right.

 

PS: That onset rate slows around 7.5 - 8.0 G's is normal, but according to the pilots it happens smoothly, and it doesn't take even a second to reach 9 G from there. That's NOT what happens in DCS.

 

Here's the Gripen, also a statically unstable FBW 9 G limited design, hitting 9 G's quickly and smoothly at 5:34 min (what you'd expect the F-16 be capable of):

eXaPfUs6sQw?t=332

Edited July 24, 2020 by Hummingbird

[NineLine]

Thanks for all the feedback guys, this issue is now closed. Wags and T-Day (4,200 hr F-16 pilot) spent yesterday testing this and reviewing available flight performance charts. We are happy with how it is performing by matching the available data and SME feedback.

 

Also, please note that the Viper is not a Gripen, and makes no sense to reference one for the other given the different FCS.

[NineLine]

Dear all, just to add to this, as it never hurts to be triple checking these things, I did my own tests on this last night and this morning.

 

Full disclaimer, I did find one issue in that the tolerance between our dev build and Open Beta is slightly different, I am assuming this is the on going tuning of the global issue of when blackout and G-LOC occur (the effect, not the FM). That said, beyond being able to pull a slightly higher G in our dev build (9.9 - 10 in OB vs 10.1-10.3 dev build), it didn't impact the results here at all.

 

For this test I only used the most current OB. I properly warmed up before performing the charted G's below. Warming up, especially in OB is very important, it staves off G-LOC quite a bit vs not.

 

The chart below is edited only in my data to best line up where it makes sense, again the supplied chart is poor to compare with as it doesn't give the points to really line up as well.

 

As well, I showed what would happen if the 'plateau' shown in the users chart was removed (circled in gray), this isn't 100% scientific, but I believe this was pilot error as I could recreate it by maxing AoA. even with this edit, if you look at the curve, its very sharp, and not smooth at all, not seeing the original flight for the chart, I would have to assume the pilot was fighting the AoA and losing. The Viper is a beast, but still needs smooth tender loving pilot hands, least in the virtual world.

 

Last disclaimer, I have been spending a ton of time on the P-47, and not as much as I would like on the Viper, so I did take some time to get myself comfortable with the Viper again, and learned where the controls liked to be and where they didn't.

 

Thanks for giving me the excuse to do this on my Saturday morning, I love this stuff :)

 

PS if it wasn't clear, the blue line was my flight vs the charted results.

 

Edited July 25, 2020 by NineLine

[bkthunder]

Dear all, just to add to this, as it never hurts to be triple checking these things, I did my own tests on this last night and this morning.

 

Full disclaimer, I did find one issue in that the tolerance between our dev build and Open Beta is slightly different, I am assuming this is the on going tuning of the global issue of when blackout and G-LOC occur (the effect, not the FM). That said, beyond being able to pull a slightly higher G in our dev build (9.9 - 10 in OB vs 10.1-10.3 dev build), it didn't impact the results here at all.

 

For this test I only used the most current OB. I properly warmed up before performing the charted G's below. Warming up, especially in OB is very important, it staves off G-LOC quite a bit vs not.

 

The chart below is edited only in my data to best line up where it makes sense, again the supplied chart is poor to compare with as it doesn't give the points to really line up as well.

 

As well, I showed what would happen if the 'plateau' shown in the users chart was removed (circled in gray), this isn't 100% scientific, but I believe this was pilot error as I could recreate it by maxing AoA. even with this edit, if you look at the curve, its very sharp, and not smooth at all, not seeing the original flight for the chart, I would have to assume the pilot was fighting the AoA and losing. The Viper is a beast, but still needs smooth tender loving pilot hands, least in the virtual world.

 

Last disclaimer, I have been spending a ton of time on the P-47, and not as much as I would like on the Viper, so I did take some time to get myself comfortable with the Viper again, and learned where the controls liked to be and where they didn't.

 

Thanks for giving me the excuse to do this on my Saturday morning, I love this stuff :)

 

PS if it wasn't clear, the blue line was my flight vs the charted results.

 

 

Not sure I get this right, but from your chart (blue line vs green line) I read the following:

 

RL F-16 hits 9g in about 6.8 s

DCS F-16 hits 9g in about 7.8s

 

1 second difference = 13% difference

 

RL F-16 hits 9.3g in 7.8 seconds

DCS F-16 hits 9.3 g in 9.6 seconds

1.8 seconds difference = 19% difference

[NineLine]

Not sure I get this right, but from your chart (blue line vs green line) I read the following:

 

RL F-16 hits 9g in about 6.8 s

DCS F-16 hits 9g in about 7.8s

 

1 second difference = 13% difference

 

RL F-16 hits 9.3g in 7.8 seconds

DCS F-16 hits 9.3 g in 9.6 seconds

1.8 seconds difference = 19% difference

 

Well and that is what I mean, I really don't know the times and such on the original chart, so it's not a great comparison, but compare the differences from my flight vs the red line that is supposed to be DCS, and I bet the difference is much greater, I am sure I can get better with more time as well, again, I am very green in the Viper right now, I am have been heavy on WWII stuff right now.

 

To put it another way, I don't know who the pilot was that did the real-world testing, but I am not as good of a F-16 pilot as him, I am pretty safe in assuming that.

 

As well, the DCS data shown on that chart, the red line, I cannot reproduce unless I fly in a way that I know is not correct. I think you can agree that if you saw only my blue line, and the green line of the real data, its no where near as startling as the red line.

Edited July 25, 2020 by NineLine

[Contact409]

So Nineline didn`t reach 9.3 at the moment it should be reaching 9.3. So the pitch damping is too great or what ?

 

https://forums.eagle.ru/showthread.php?t=281757

[Hummingbird]

Real life F-16 hitting 9+ G's a in a break level turn:

https://www.instagram.com/p/BojNMR7FSps/

 

The onset rate is visibly much faster here than in DCS.

 

Note the initial turn is paused at ~8 G's to keep within the showline, however later he pulls to 1 to 9+ G's much faster than what is possible in DCS.

Edited July 25, 2020 by Hummingbird

[Hummingbird]

Dear all, just to add to this, as it never hurts to be triple checking these things, I did my own tests on this last night and this morning.

 

Full disclaimer, I did find one issue in that the tolerance between our dev build and Open Beta is slightly different, I am assuming this is the on going tuning of the global issue of when blackout and G-LOC occur (the effect, not the FM). That said, beyond being able to pull a slightly higher G in our dev build (9.9 - 10 in OB vs 10.1-10.3 dev build), it didn't impact the results here at all.

 

For this test I only used the most current OB. I properly warmed up before performing the charted G's below. Warming up, especially in OB is very important, it staves off G-LOC quite a bit vs not.

 

The chart below is edited only in my data to best line up where it makes sense, again the supplied chart is poor to compare with as it doesn't give the points to really line up as well.

 

As well, I showed what would happen if the 'plateau' shown in the users chart was removed (circled in gray), this isn't 100% scientific, but I believe this was pilot error as I could recreate it by maxing AoA. even with this edit, if you look at the curve, its very sharp, and not smooth at all, not seeing the original flight for the chart, I would have to assume the pilot was fighting the AoA and losing. The Viper is a beast, but still needs smooth tender loving pilot hands, least in the virtual world.

 

Last disclaimer, I have been spending a ton of time on the P-47, and not as much as I would like on the Viper, so I did take some time to get myself comfortable with the Viper again, and learned where the controls liked to be and where they didn't.

 

Thanks for giving me the excuse to do this on my Saturday morning, I love this stuff :)

 

PS if it wasn't clear, the blue line was my flight vs the charted results.

 

 

I'm curious as to why are you running the tests with G-Loc enabled at all? All it does is make testing more difficult.

 

IMO it's important to always turn of G effects when testing aircraft performance.

 

PS: Keep in mind that graph is for a level (horizontal turn), not a pull out. In other words if your results are from pull outs they are not valid.

[NineLine]

I'm curious as to why are you running the tests with G-Loc enabled at all? All it does is make testing more difficult.

 

IMO it's important to always turn of G effects when testing aircraft performance.

 

Do they turn the G effects off in real life when testing? ANd obviously if that was your curve shown on the graph, I am not doing something wrong.

 

Real life F-16 hitting 9+ G's a in a break level turn:

https://www.instagram.com/p/BojNMR7FSps/

 

The onset rate is visibly much faster here than in DCS.

 

Note the initial turn is paused at ~8 G's to keep within the showline, however later he pulls to 1 to 9+ G's much faster than what is possible in DCS.

 

I disagree with your findings there. I have shown above, even someone with low hours in the Viper (DCS) that being me can pull 9 Gs within a reasonable time. It's funny that you send that today against DCS, and that was sent to me yesterday in defence of DCS. Interesting.

 

I think you need to look at how you are flying the sim before you are critical of it. I can see in your curve you are not smooth on the stick, and most likely pulling more AoA than you should be.

 

I'm curious as to why are you running the tests with G-Loc enabled at all? All it does is make testing more difficult.

 

IMO it's important to always turn of G effects when testing aircraft performance.

 

PS: Keep in mind that graph is for a level (horizontal turn), not a pull out. In other words if your results are from pull outs they are not valid.

 

Luckily mine was a level turn, because your graph has very little information, its very poor graph to judge anything on. Makes it hard to know what you re doing wrong. But yes, level turn. Anything else you need to know?

[Hummingbird]

I think you need to look at how you are flying the sim before you are critical of it. I can see in your curve you are not smooth on the stick' date=' and most likely pulling more AoA than you should be.[/quote']

 

Not sure what you mean by watching the AoA NineLine, we're talking max performance turns here, i.e. instantaneous turn rate. In the F-16 you achieve this by simply pulling back the stick and riding the max AoA the FLCS allows for at that specific speed.

 

Hence the F-16's ITR is actually limited below what the airframe is truly capable of (to prevent deep stall), which is why the ITR and STR curves are so close on the F-16 EM diagram, because the FLCS is constantly limiting the AoA you are able to pull:

 

So I'm sorry the excuse that I am pulling too much AoA doesn't make sense.

Edited July 25, 2020 by Hummingbird

[NineLine]

But your controls are not equal to the F-16 right? I mean you understand that your stick sitting on your desk isn't going to act the same as the one sitting in a real F-16 right?

 

I mean if you are expecting to just slam the stick where you want it and the sim to figure it out, you are not correct.

 

The data you showed for the curve in that graph was poor, and I showed that its not correct, and it backs up what Wags said as far as what T-Day tested. Please show me your track, tacview, maybe a video with your control indicator up, and I will submit that to the team for review if you like. But it seems pointless as I showed that my findings are much closer to the real life data you shared.

[TLTeo]

Do they turn the G effects off in real life when testing?

I mean, in a way they do turn off DCS-level G effects, because the tolerance of pilots in DCS is noticably too low, and that really gets in the way of testing a maximum performance test in the Viper.

 

 

Also, I don't understand your point about the controls. By its nature of having FBW controls, it's much more straightforward to compare joystick and real inputs - when you command stick full aft with a joystick, and translate that into stick full aft in the sim, you're commanding maximum G available. That's it, there's no technique to reach that maximum G efficiently or whatever, that should be handled by the FLCS because it limits what AoA you can pull. It's not a Tomcat where if you pull stick full aft, you're just going to put too much AoA on the jet and bleed speed without changing direction of motion.

[Hummingbird]

Testing from April where I deliberately eased my way to full back stick and max ITR, so no abrupt pull:

6s7XfzAVrrk

 

Quick testing just now with abrupt pull, i.e. fast full back stick (note the time it takes to hit 9 G first at 2:15 min):

4ORuLOpyAVs

 

Result = No change, still ~8.6 G @ 0.64 Mach where it should be 9 G, and 9.0 G not being hit until 0.76-0.77 Mach in DCS. So there's the proof that the ITR performance is too low, and by speed margin of 0.10+ mach.

Edited July 25, 2020 by Hummingbird

[NineLine]

I mean, in a way they do turn off DCS-level G effects, because the tolerance of pilots in DCS is noticably too low, and that really gets in the way of testing a maximum performance test in the Viper.

 

As I said in my post above, internally we have a slightly higher tolerance I hope to see moved to release soon. That said, at no time in any of my testings did the G-effects hurt or impede my testing. I started to see the black come in at about 8.5, higher if I properly warmed up before hand.

 

Also, I don't understand your point about the controls. By its nature of having FBW controls, it's much more straightforward to compare joystick and real inputs - when you command stick full aft with a joystick, and translate that into stick full aft in the sim, you're commanding maximum G available. That's it, there's no technique to reach that maximum G efficiently or whatever, that should be handled by the FLCS because it limits what AoA you can pull. It's not a Tomcat where if you pull stick full aft, you're just going to put too much AoA on the jet and bleed speed without changing direction of motion.

 

The controls of our simulator, as in the ones sitting on the desk and floor are not equal to those used in the real deal, we see the same thing with anything in DCS, that controls, even with control freezing modelled, or limits set, can still be negated by slapping your controls around willy nilly, as I said, he shows a plateau in his testing that I was easily able to recreate by trying to overpower the sim pushing the stick beyond where it might be.

 

It's a common problem with SMEs, or people who are pilots or have had the experience, and I talked to Nick about this in-depth as well, we don't get the same feedback, we don't have the same controls, we don't have the adrenaline, the fear and all the other senses that a real pilot would have. So if someone said I have a buddy who is a Viper pilot and he tried the DCS version for 10 minutes and said it doesn't feel anything like the real thing I would accept that, because it is, at the end of the day, different.

 

I am sure Nick, the owner of ED, will be the first one to tell you that no matter how good a sim gets, there are still many missing factors to make it feel exactly the same. Controls are one of these, even the best stick on the market is not going to simulate an F-16, P-47 or whatever 100%. You as a sim pilot have (and a great many already do) have to adapt to it.

 

You can see it in Movers vids, the sim doesn't respond like he is used to, but as he gets a feel for it, he excels at it, and he gets there much faster than those of us with no real-world experience, and in respect to Mover, this is my assessment based on enjoying his videos. I have not discussed this with him at all. But only base it on what I see him describe in his vids of trying out the Viper and different modules.

[NineLine]

 

Result = No change, still ~8.6 G @ 0.64 Mach where it should be 9 G, and 9.0 G not being hit until 0.76-0.77 Mach in DCS. So there's the proof that the ITR performance is too low, and by speed margin of 0.10+ mach.

 

You don't appear to warm up at all, and just jump straight into trying to pull max G's, please ask your friends that are fighter jocks if they do the same thing? I mean don't feel bad, even Wags had to straighten me out on that as well. And the real data you posted, that is the altitude they created that curve at? Dragging your wing in the water?

 

The curve that you stated was real world, please show the source, and all the specs on how it was tested. I just want to get some idea what you are basing this off of. Your bar is very foggy.

Edited July 25, 2020 by NineLine

[Hummingbird]

Mover also says it's easy to hit 10 G's by accident in the real Viper when transitioning from a heavily laden two bagged one you've flown for a long time and to a clean one, i.e. he got used to pulling 'this' hard to get 'this' amount of G in the heavy one, now all of a sudden pulling 'this' hard gives you a ton more G. So there's that. Again just proving that the G onset rate in the Viper is so fast that if you just jank the stick (as in apply too much force) then it will throw 9+ G's at you within a second.

 

That aside the control aren't the issue with regards to the F-16, because as TLTeo and I point out, the FLCS limits the amount of AoA the aircraft is able to pull. Once you reach a speed where more than 9.3 G's would result from 15 deg AoA, the FLCS switches to G control instead of AoA control.

[Hummingbird]

You don't appear to warm up at all, and just jump straight into trying to pull max G's, please ask your friends that are fighter jocks if they do the same thing? I mean don't feel bad, even Wags had to straighten me out on that as well. And the real data you posted, that is the altitude they created that curve at? Dragging your wing in the water?

 

The curve that you stated was real world, please show the source, and all the specs on how it was tested. I just want to get some idea what you are basing this off of. Your bar is very foggy.

 

No need to warm up when G effects are off. If you are refering to myself warming up, I did the test over and over again before recording, and then loaded out and in again to show temp etc.

 

The EM diagram is for SL, hence I try to stay as close to SL as possible.

 

What curve? If you're talking about the RL ITR, it's on page B8-65 in the HAF manual (hope it's ok to post this screenie considering all references already made to this manual on the board):

 

 

Edited July 25, 2020 by Hummingbird

[NineLine]

That aside the control aren't the issue with regards to the F-16, because as TLTeo and I point out, the FLCS limits the amount of AoA the aircraft is able to pull. Once you reach a speed where more than 9.3 G's would result from 15 deg AoA, the FLCS switches to G control instead of AoA control.

 

Again, FLCS isn't designed to compensate for every type of control on the market, its designed real-world F-16 controls. How could a sim be programmed to compensate for every variation of every stick that can possibly be used in DCS? I mean you know you can make unrealistic maneuvers in any aircraft in DCS because the control principals between something sitting on a desk and something actually connected to control surfaces are different.

 

I can put two Warthogs (controllers) next to each other and they may not be exactly the same.

Edited July 25, 2020 by NineLine

[NineLine]

No need to warm up when G effects are off. If you are refering to myself warming up, I did the test over and over again before recording, and then loaded out and in again to show temp etc.

 

The EM diagram is for SL, hence I try to stay as close to SL as possible.

 

What curve? If you're talking about the RL ITR, it's on page B8-65 in the HAF manual (hope it's ok to post this screenie):

 

We are talking about the one you showed off at Hoggit, the real world onset vs your onset you created. The same one I flew and tried to compare to, please make sure you are reading my posts and not just ignoring them.

 

And the visual effects vs effect of G on the FM are two different things, again, you are not comparing 1 to 1, so your tests are not going to be valid. Sorry. Turing off G effects is essentially like turning on other cheats to help fly, if you want to compare real world you should try to replicate that as much as possible, and also understand, this is a sim/game not the real world, there are more factors that can change vs a real F-16.

Edited July 25, 2020 by NineLine