[CORRECT AS IS] Odd rocking in G-onset

[Hummingbird]

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.

 

The one at Hoggit is based on G meter footage, which I already showed. There is no graph for this in the HAF manual, or any other F-16 document I know of.

 

 

And the 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.

 

How are my tests not valid?

 

Keep in mind I did the ITR test in two ways, 1) where I eased the stick back to full deflection and 2) where I just janked it back.

 

So there's no way it could've gone wrong, at all.

[NineLine]

The one at Hoggit is based on G meter footage, which I already showed. There is no graph for this in the HAF manual, or any other F-16 document I know of.

 

So you based your flight exactly the same as what you based the real-world footage on? Can I see that? I mean you should be able to do side by side. All things equal, load out, fuel load, etc.

 

You used that graph on Hoggit so that is what I am trying to match against when I saw how close my flight was to the real-world data that is why I opened up the thread again.

 

So show me all the work for that specific graph you used to say the FM is wrong. I mean that's what I am trying to get at here. Why could I fly so close to it, but yours shows so far off?

 

I know you guys think I am being mean when I ask for this, but its not the case, its everything Wags, Yo-Yo, and everyone else will ask me for if I bring them an issue. You need to do all the work to effect change.

[Hummingbird]

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.

 

Ok, let me explain again: The controls are not going to an issue with the F-16 because it's not like the other fighters/aircraft ingame, and that is because the F-16's FLCS sets a limit to how fast & far the control surfaces can at all move in accordance with speed & AoA, and this is irrespective of how fast full deflection is commanded by the pilot. (Also I doubt my joystick can command full aft stick faster than the pressure sensitive F16 stick anyway)

 

In other words if the F-16 pilot commands full aft stick then said command will run through the FLCS which then translates that into movement of the stabilizers for max allowed AoA at that specific speed, which up until 9.3 G's is 15 deg AoA.

 

In short, I litterally can't command the F-16 to exceed its limits irrespective of what controller I'm using, because in the end its the FLCS which decides how much the control surfaces will at all move.

[NineLine]

Ok, let me explain again: The controls are not going to an issue with the F-16 because it's not like the other fighters/aircraft ingame, and that is because the F-16's FLCS sets a limit to how fast & far the control surfaces can at all move in accordance with speed & AoA, and this is irrespective of how fast full deflection is commanded by the pilot. (Also I doubt my joystick can command full aft stick faster than the pressure sensitive F16 stick anyway)

 

In other words if the F-16 pilot commands full aft stick then said command will run through the FLCS which then translates that into movement of the stabilizers for max allowed AoA at that specific speed, which up until 9.3 G's is 15 deg AoA.

 

In short, I litterally can't command the F-16 to exceed its limits irrespective of what controller I'm using, because in the end its the FLCS which decides how much the control surfaces will at all move.

 

And again, the FLCS cannot be programmed for a desktop controller because a desktop controller can do different things depending on the make, model and even stick to stick. You like evoking Mover, as him if the controls he has used for DCS are anything like what he has used in the real jet. So no, its just like a WWII bird, the stick in a P-47 doesn't move like that of one on your desk, so you will get some variation, even though the FM is programmed to interact like a real P-47 stick, anything short of that will have variations that will throw off your comparison.

 

You are not flying a real F-16, you are flying as close as we can to it, we cannot control what flight stick, gamepad or keyboard you use. We cannot account for that, there are no documents that allow us to program an FM to match all those variables, again, like a WWII aircraft, the FM expects the exact same controls as is in the real thing, its why we try to add things like control freezing, etc.

[Noctrach]

@Nineline however despite everything, doesn't the combination of evidence from both ED and other users lead to the conclusions that:

 

a) As per your own test, the G-onset at higher speeds is slower than the one in the real Viper (greater than 10% margin)

 

b) As per user tests (need verification) the speed at which the DCS Viper attains 9G is higher than the real deal? (greater than 10% margin) Taking into account EM graphs are interpolated data.

 

 

That seems significant.

 

 

G-loc should imo not be considered "cheating" for these tests as it does not impact the control response in sim at all. It merely impacts the visuals and/or complete lack of control in case of blackout/redout. So if anything a G-loc free test would give a better representation of the in-sim FM response as it is not prematurely terminated due to pilot loc? (Less variables in testing)

 

Controls wise: wouldn't any FLCS just take the relative position of commanded input to the minimum and maximum possible commanded inputs and split the difference? I.e. regardless of exact input device, 60% commanded input results in the same FLCS response?

 

Are you saying a force-sensing stick will create a noticeably different experience and FLCS response in DCS than a regular joystick?

 

 

edit: What you are saying about controls sounds illogical to me, any flight control system in a real jet has a minimum and maximum deflection. This can very easily be linked to the minimum and maximum outputs of any controller, as we're just talking a min/max digital output value, an incredibly easy abstraction to make.

What you cannot simulate are stick forces in response to pressure differentials on the control surfaces, or vibrations and butt-sensor telemetry. This is what makes real flying different and why any pilot (including Mover) initially struggle to adjust to sims. They are missing a ton of senses they would have in the real jet. However, this has nothing to do with stick deflection vs aircraft response. I'm not sure what you are trying to convey here. Is the FLCS response non-linear?

Edited July 26, 2020 by Noctrach

[NineLine]

@Nineline however despite everything, doesn't the combination of evidence from both ED and other users lead to the conclusions that:

 

a) As per your own test, the G-onset at higher speeds is slower than the one in the real Viper (greater than 10% margin)

 

b) As per user tests (need verification) the speed at which the DCS Viper attains 9G is higher than the real deal? (greater than 10% margin) Taking into account EM graphs are interpolated data.

 

 

That seems significant.

 

 

G-loc should imo not be considered "cheating" for these tests as it does not impact the control response in sim at all. It merely impacts the visuals and/or complete lack of control in case of blackout/redout. So if anything a G-loc free test would give a better representation of the in-sim FM response as it is not prematurely terminated due to pilot loc? (Less variables in testing)

 

Controls wise: wouldn't any FLCS just take the relative position of commanded input to the minimum and maximum possible commanded inputs and split the difference? I.e. regardless of exact input device, 60% commanded input results in the same FLCS response? Again, and confirmed by Wags as he deals with SMEs face to face, this is the biggest issue for real pilots in DCS.

 

Or are you saying a force-sensing stick will create a noticeably different experience and FLCS response in DCS than a regular joystick?

 

A) as I disclaimered, I am probably not flying the F-16 like a real life F-16 pilot, we have already verified it with a pilot with 4200 hours in a real Viper. He trumps me, I admit to my short comings as a Viper pilot up front.

 

B) I cant verify much from the user test, he isn't giving me much, again I defer to our SMEs, until users can give me definitive specs on all testing variables.

 

C) G effects being turned off is a cheat, period. It what helps the FM act more like its programmed to act, removing those moves the bar farther away. It would be like turning off control freezing on the Me-109, it doesn't make testing easier, it makes it more unrealistic.

 

D) A force feedback stick would get you closer, but again... stick to stick can differ we pay a lot of money for these sticks on our desktop, but we still have to consider its not the real thing, and compensate.

Edited July 26, 2020 by NineLine

[Hummingbird]

First of all remember we're currently discussing two issues in this thread:

1) G-onset rate

2) Instantanous turn rate

 

For example the two videos I posted just before was to illustrate the subpar ITR performance, not the G-onset rate (although the latter one, test 3, can be used for that too)

 

Hence it's important to distinguish which one we're addressing when.

 

So you based your flight exactly the same as what you based the real-world footage on? Can I see that? I mean you should be able to do side by side. All things equal, load out, fuel load, etc.

 

You used that graph on Hoggit so that is what I am trying to match against when I saw how close my flight was to the real-world data that is why I opened up the thread again.

 

So show me all the work for that specific graph you used to say the FM is wrong. I mean that's what I am trying to get at here. Why could I fly so close to it, but yours shows so far off?

 

I know you guys think I am being mean when I ask for this, but its not the case, its everything Wags, Yo-Yo, and everyone else will ask me for if I bring them an issue. You need to do all the work to effect change.

 

I hope you understand Im not trying to be mean either, I am merely desperately trying to get this issue deeply looked into as the real Viper is clearly pulling G's quicker than the DCS one.

 

Regarding G-onset rate:

The graph I posted at Hoggit was meant to illustrate the issue, not to be pin point exact (hence the lack of a grid), to make sure the problem is understood. I simply timed how long it took to reach 9 G's, and how linear the evolution was in the real footage and ingame recordings.

 

In these two vids for example I try the level break turn (in a clean Viper) as in the RL footage, and here the sudden pause around 5.5-6 G really shows up:

[NineLine]

First of all remember we're currently discussing two issues in this thread:

1) G-onset rate

2) Instantanous turn rate

 

 

No, we are not, I only tested G-onset, I only recorded that, and only submitted those values back to you as best I could with what little info I had to go on.

 

G-Onset to me appears to be really good, and honestly I am not surprised, it was confirmed by someone much more skilled in Viper piloting than me.

 

If you want to talk turn rates, that's another thread, but be warned, I have seen the curves and our SME has already signed off on those as well.

 

So stick to G-Onset, and answer what I need to see.

 

How did you make that curve you posted on Hoggit? And why do my results match so much closer?

[Ramsay]

D) A force feedback stick would get you closer, but again... stick to stick can differ we pay a lot of money for these sticks on our desktop, but we still have to consider its not the real thing, and compensate.

I don't see how a force feedback stick would get closer ?

 

The real F-16 side stick is force sensing with very little movement, it's more like a Saitek X-65F than anything else.

 

https://www.amazon.co.uk/Saitek-Flight-Combat-Control-System/dp/B003BXVIQS

[Hummingbird]

If you want to talk turn rates, that's another thread, but be warned, I have seen the curves and our SME has already signed off on those as well.

 

Alright, but they don't match, and anyone can test it. Just do a some max rate level turns, doesn't take an experienced pilot to do that:

 

Testing footage:

4ORuLOpyAVs

6s7XfzAVrrk

 

 

 

I'll stop talking about ITR in this thread now.

 

So stick to G-Onset, and answer what I need to see.

 

How did you make that curve you posted on Hoggit? And why do my results match so much closer?

 

 

As mentioned I made it by timing the G onset rates in the videos I posted, i.e.:

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

 

Btw, I've asked TheFighterPilot, who is also a RL F-16 jockey, to weigh in on the matter.

Edited July 26, 2020 by Hummingbird

[Krippz]

I think the OP brings up valid points.

[Hummingbird]

Btw, I urge everyone to test this themselves.

[NineLine]

I mean the end of the story here is I was able to fly, with in acceptable range of the supplied graph, again noting that I am not an expert F-16 pilot, and solidly backed up by our own SME, whose opinion should matter as much, if not more than any other Viper pilot (more because we know these are controlled tests to verify our FM).

 

I have shown its possible at this point to fly close to expected results, and I know I would only get better the more time I spent on it. I see no reason to keep going around. If a real SME, wants to put in the effort, I am sure he can reach out to Wags and discuss. In my mind I am not seeing anything compelling to continue this.

[NineLine]

Btw, I urge everyone to test this themselves.

 

And I urge everyone to show their work, a vague graph and a statement that its wrong doesn't enact change I am afraid, remember this is being tested by real-world pilots. Please respect that as well.

[Hummingbird]

I'm sorry, I tried to illustrate and prove the issue as best I could, and I believe I did more than just show a graph and claim something is wrong.

Edited July 26, 2020 by Hummingbird

[Jester2138]

NineLine, when you say ED's SME has "signed off on it as-is," what does that mean exactly? Did he scientifically test it to verify the FM was correct, as Hummingbird is doing? Or did he fly around a bit and just feel it? Speaking as a real-world pilot myself, the FM is not valid just because a pilot played for a bit and said "Yeah that seems about right." Scientific testing is the only valid evidence here, and Hummingbird is providing that.

[NineLine]

NineLine, when you say ED's SME has "signed off on it as-is," what does that mean exactly? Did he scientifically test it to verify the FM was correct, as Hummingbird is doing? Or did he fly around a bit and just feel it? Speaking as a real-world pilot myself, the FM is not valid just because a pilot played for a bit and said "Yeah that seems about right." Scientific testing is the only valid evidence here, and Hummingbird is providing that.

 

To be honest I would hope that the testing is done much better than what we have seen here from both me and Hummingbird, I wouldn't call this scientific to much degree. They are not watching Instagram videos and playing around a bit. This is a person paid to test an give feedback to make the F-16 as real as possible. As I understand the entire day was spent on this session, and G Onset was tested. This isn't the only checks that go into our FM, but one of many as we tried explaining in our recent FM doc

 

Nothing in this thread really has been enough but to have it checked, the frustrating part is it was checked, but everyone only wants to find fault with our methods and results, when if you look at what as been given here, really only shows improper or incomplete testing, again not being mean, I still haven't seen the work up to make it, and much of the data points were left off. Its frustrating to be the heavy on this, and now spent about 24 hours on it which honestly the update from Wags should have been enough. Especially with me backing it up. And now, I am typing this from a water park with one eye on my daughter, living the life ;)

[Hummingbird]

I just fail to understand how I have messed up my level turn tests, both for ITR, STR and G-onset.

 

I mean I am definitely not saying I am = F-16 pilot, but at the same time I don't see why it takes an F-16 pilot to do a proper test of ITR, STR & G-onset rate in a simulator as long as you understand the dynamics at work. I simply don't see where my testing is flawed to the point that it isn't reflecting max performance in level turns (instantaneous & sustained) as well as in onset rate.

 

I also don't understand why the FM would be "signed off as is" if even ED's own paper on the FMs shows the DCS F-16 to be underperforming in STR below 0.6 mach (where most dogfights end up) and at peak rate:

 

 

 

Point being what's then stopping the same from being the case with the observed & measured issues with G-onset rate?

Edited July 26, 2020 by Hummingbird

[NineLine]

Which is funny because over on Hoggit you suggested Wags and his SME must have tested the wrong thing and thought they we're testing onset, or that they did it wrong, so you put me in a tough place. Anyways I am getting the stink eye from the wife, so I am off for a bit. Have a good weekend.

Edited July 26, 2020 by NineLine

[Hummingbird]

Which is funny because over on Hoggit you suggested Wags and his SME must have tested the wrong thing and thought they we're testing onset, or that they did it wrong, so you put me in a tough place.

 

and I still feel the same, i.e. either:

1) something wasn't done right/thuroughly enough or

2) they might have looked at the wrong EM charts etc (human error) or

3) they're accepting discrepancies within a much larger threshold than I am (I hope not)

 

And I don't say this to be disrespectful, I've just tested it so much by now that I can't understand how they could come to different results unless it's for one of the above reasons, as I'm sure I did everything to get peak performance at the specified conditions (and I documented it too) - something it doesn't take a RL F-16 pilot to do, it just demands thuroughness and a relatively basic understanding of flying & aerodynamics.

 

In short, we're all human, F-16 pilots too, so just because a real F-16 pilot looked at it doesn't exclude human oversight. (And yes I know this goes for me as well, I just can't see where I could've possibly gone wrong in my tests at this point)

 

Anyways I am getting the stink eye from the wife, so I am off for a bit. Have a good weekend.

 

It wasn't my intention to ruin a day at the park NineLine, so please don't hesitate to wait with responding to issues such as these until you have time. I bring up these issues out of passion for the aircraft & sim, not to be annoying.

 

A good weekend to you as well.

Edited July 26, 2020 by Hummingbird

[SCPanda]

This is a great discussion. Thank you Nineline and ED for reopening this thread!

[Contact409]

Which is funny because over on Hoggit you suggested Wags and his SME must have tested the wrong thing and thought they we're testing onset, or that they did it wrong, so you put me in a tough place. Anyways I am getting the stink eye from the wife, so I am off for a bit. Have a good weekend.

 

Any chance to bring Wags to here to show us how he and SME did the test ? How they draw the conclusion of "correct as it is"

[NineLine]

As mentioned I made it by timing the G onset rates in the videos I posted, i.e.:

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

 

Your tests and the example video you are comparing to don't match at all. How is that supposed to be a good comparison to chart when you are not flying remotely like the example video?

 

I mean you keep saying it's well documented by you, but I haven't seen it yet. Your chart has no real data points, no time, only the 9 G point... it's really hard to compare anything with your testing, you for some reason avoid showing the controls during your test, you are almost dragging a wing in the water compared to what the Instagram video shows... what am I missing here.

 

But if you want to continue to suggested we are making all the mistakes, and everything you are doing is perfect, please show it. I am willing to listen and have been asking over and over.

 

If you can document it here like you said you have, great otherwise I will bump the two posts with the answer to the concerns about G-onset to the end of this thread and close it again, realize it's only me asking, the team is quite happy with where it's at. I won't keep going back and forth, the only answer you have given so far is ED must have tested it wrong because you see nothing wrong with your tests.

 

I'll visit this thread back tomorrow night or Monday and see if you were able to transcribe your documentation that you keep. Hopefully more complete than what you have given. Thanks.

[NineLine]

Any chance to bring Wags to here to show us how he and SME did the test ? How they draw the conclusion of "correct as it is"

 

You are asking me to ask Wags if he is willing to prove that his 4200 hour SME knows how to check G-onset? Based on nothing solid I can see so far?

 

I will need a little more solid proof before I step on that landmine. Especially when I showed my simple test already backing up our claims.

 

I am willing to go to Wags on it, but I have not been given anything to do so yet.

[Hummingbird]

Your tests and the example video you are comparing to don't match at all. How is that supposed to be a good comparison to chart when you are not flying remotely like the example video?

 

I mean you keep saying it's well documented by you, but I haven't seen it yet. Your chart has no real data points, no time, only the 9 G point... it's really hard to compare anything with your testing, you for some reason avoid showing the controls during your test, you are almost dragging a wing in the water compared to what the Instagram video shows... what am I missing here.

 

But if you want to continue to suggested we are making all the mistakes, and everything you are doing is perfect, please show it. I am willing to listen and have been asking over and over.

 

If you can document it here like you said you have, great otherwise I will bump the two posts with the answer to the concerns about G-onset to the end of this thread and close it again, realize it's only me asking, the team is quite happy with where it's at. I won't keep going back and forth, the only answer you have given so far is ED must have tested it wrong because you see nothing wrong with your tests.

 

I'll visit this thread back tomorrow night or Monday and see if you were able to transcribe your documentation that you keep. Hopefully more complete than what you have given. Thanks.

 

Please watch the videos I sent NL, and note the time it takes to reach 9 G in them:

 

Then compare it with the RL footage (time it as well) :

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

Note: the 1.2 to 9+ G pull happens near the end of the video where he banks hard left.

 

Now compare the times for when 9 G is hit between DCS & the RL footage. That's all I did, and I measured 3.1 sec from 1.2 to 9.0 G in the RL footage and 6-8 sec ingame depending on entry speed, the last bit from 8.5 to 9.0 taking ages.

 

The hoggit onset graph was based on the measurements from that, but it was never meant to be precise, just an approximate to illustrate the issue, hence the lack of a grid or actual time figures. In short the graph was there solely to illustrate the issue to make it easy to understand incase my the written explanation wasn't.

 

Now I can redo the tests with stick inputs visible if you wish (you're gonna have to tell me how though as I can't find that option), the results will be the same however as I wasn't trying to cheat anyone by not pulling full aft stick or something (what would I get out of that?).

Edited July 26, 2020 by Hummingbird