Phantom vs XXX

[SgtPappy]

21 hours ago, renhanxue said:

The document doesn't say. If I had to guess I'd say 70% fuel because that's the standard used for the in-flight diagrams in the SFI, but it's far from certain. The aerodynamics compendium has two parts and the graphs in the other part frequently use specific weights to illustrate some point, but then it's always stated what that weight is. So, I really don't know.

According to the aerodynamics compendium the RM8A (on the AJ 37) starts compressor stalling at around 20° alpha, leaving only a small margin of error from the limit of 18°. I don't know how the DCS implementation does these days. The RM8B in the JA 37 is significantly better; the alpha limit is 23° and the engine reportedly does not start compressor stalling below 25°, which is the point where the aircraft starts losing longitudinal stability and you risk superstall or spin. That means the AJS 37 should not be able to reach the JA 37's level without compressor stalling even if you exceed the 18° limit significantly.

I don't have much to add; if anything I think you might be slightly too generous to the AJS 37 in terms of sustained turn rate, but that's just gut feeling, no numbers to back it up.

It's really a pity that we don't have a good JA 37 simulator...

 

Thank you for chiming in.

It's too bad that we don't know the weights but still great that we have that data at all. I suppose we'll see how they compare once the F-4 shows up. 

If ever a JA 37 was made, it would be quite a beast.

[Hummingbird]

A JA37 would be amazing

[Smyth]

Since I already posted a version of the Phantom vs xxx turn performance summary I was working on for myself in other threads, I think it is only fair to put the full version here for the sake of everyone else interested in this topic.

This summary is strictly based on what I can find in publicly-available documents from the original operator (US or USSR), scaled for fuel state and altitude. That correction uses formulas I have validated against available data for aircraft where turn performance is charted at different weight and altitude. The derivation is not overly complicated, but too long for one forum post.

Obviously this is not an "official" document despite the data sources, and should be taken for no more or less than whatever it is worth. However I am very confident that this is better than trying to guess based on comparing the un-altered primary sources with their different formats, units, and assumptions. For example the USSR typically uses 40% or less fuel in their charts, while the US uses 60% or more. These are both totally valid assumptions (representing either the end of a dogfight, or the start of one) however a direct comparison is not valid, since turn performance scales quite directly with weight, and fast jets carry a huge fraction of their total weight in fuel.

Finally there are three things DCS players should understand before I post this chart:

1. Data for the Mig-23 at 35 degrees wing sweep does not seem to exist. It's pretty easy to show based on back-of-the envelope calculations from aspect ratio what kind of difference that might make (measurable but not enormous), however that breaks the intent of this comparison to use only official data, so it is not included.
2. This chart is only relevant to the real MiG-21, and not to DCS, as anyone familiar with the F-5E vs. MiG-21bis balance in this game should quickly notice.
3. It is always subject to change if any better data is made public. Someone immediately corrected my F104 post with better data, so maybe (hopefully) that will happen here rule 1.16 permitting.

With all that said, here it is:

Questions, corrections, criticism, and flame wars are all welcome. Enjoy.

Post Script: there is "official" data (from McDonnell) that says the F-4E turns better than this, but I did not use that because I don't have proof it is correct relative to the USAF manual. I may be a Phantom Fanboy, but I try to be objective.

Edited March 1 by Smyth
Reverted MiG-21 weight to match DCS 2/29/24, Corrected F-5E altitude 12/24/23

[Kalasnkova74]

On 7/4/2023 at 10:47 AM, Smyth said:

I don't have any real-world data to add on this one, but I do have an observation from the game. Surprisingly enough, the Mig-19 has worse ITR than the Mig-21 and other supersonic jets. I don't have the module, but from watching experienced pilots in cold war multiplayer, the winning Mig-19 strategy is to patiently build an energy advantage with sustained turns and loops. Other aircraft like the F-5 can not only survive at close quarters, but pose a serious threat until the -19 powers away in the vertical. It is a very different fight than an F86 or Mig-15.

The MiG-19 fielded by most nations is not the same version as is in DCS. The MiG-19P’s heavy radar inhibits the Farmers turn performance (and acceleration) relative to the -C model flown in Southeast Asia, Pakistan and elsewhere. The MiG-19C is a much more dangerous foe close in, as the 555th TFS found out the hard way over North Vietnam. 

[Smyth]

12 hours ago, Kalasnkova74 said:

The MiG-19 fielded by most nations is not the same version as is in DCS. The MiG-19P’s heavy radar inhibits the Farmers turn performance (and acceleration) relative to the -C model flown in Southeast Asia, Pakistan and elsewhere. The MiG-19C is a much more dangerous foe close in, as the 555th TFS found out the hard way over North Vietnam. 

That's all true, although I would argue the -19P is still more dangerous because Aim-9B make it more difficult for faster aircraft to extend, and the radar only adds <<5% to the loaded weight of a MiG-19. The USSR either didn't realize how lethal a -17F or -19S with R3S would be, or did realize and didn't want to give that to their less-trusted allies. I know the US was 100% expecting to find MiG-17F with missiles in SEA.

I should have clarified that I don't think low-speed turns with a -19 are a really good idea in any version of the Phantom, and certainly not the hard-wing flavor. The slatted -E will just pose a non-zero challenge to a -19 driver in a scissors which is maybe a bit unexpected compared to similar-looking 50s swept-wing transonic jets.

As seen in the chart I posted above, the F-5E that faces the MiG-19 in DCS currently has really exceptional instant turn rate for a supersonic fighter. Even then the F-5E is quickly destroyed by an F-86 or MiG-15/7 in a slow turning fight, while a MiG-19 of any version takes much more time.

[SgtPappy]

On 10/14/2023 at 1:39 AM, Smyth said:

Since I already posted a version of the Phantom vs xxx turn performance summary I was working on for myself in other threads, it think it is only fair to put the full version here for the sake of everyone else interested in this topic.

This summary is strictly based on what I can find in publicly-available documents from the original operator (US or USSR), scaled for fuel state and altitude. That correction uses formulas I have validated against available data for aircraft where turn performance is charted at different weight and altitude. The derivation is not overly complicated, but too long for one forum thread.

Obviously this is not an "official" document despite the data sources, and should be taken for no more or less than whatever it is worth. However I am very confident that this is better than trying to guess based on comparing the un-altered primary sources with their different formats, units, and assumptions. For example the USSR typically uses 40% or less fuel in their charts, while the US uses 60% or more. These are both totally valid assumptions (representing either the end of a dogfight, or the start of one) however a direct comparison is not valid, since turn performance scales quite directly with weight, and fast jets carry a huge fraction of their total weight in fuel.

Finally there are three things DCS players should understand before I post this chart:

1. Data for the Mig-23 at 35 degrees wing sweep does not seem to exist. It's pretty easy to show based on back-of-the envelope calculations from aspect ratio what kind of difference that might make (measurable but not enormous), however that breaks the intent of this comparison to use only official data, so it is not included.
2. This chart is only relevant to the real MiG-21, and not to DCS, as anyone familiar with the F-5E vs. MiG-21bis balance in this game should quickly notice.
3. It is always subject to change if any better data is made public. Someone immediately corrected my F104 post with better data, so maybe (hopefully) that will happen here rule 1.16 permitting.

With all that said, here it is:

Questions, corrections, criticism, and flame wars are all welcome. Enjoy.

Post Script: there is "official" data (from McDonnell) that says the F-4E turns better than this, but I did not use that because I don't have proof it is correct relative to the USAF manual. I may be a Phantom Fanboy, but I try to be objective.

Thanks for making this wonderful summary.

I have a couple of questions though. You mentioned that the MiG-21 Russian manuals usually use 40% fuel load. I used the same manual for my MiG-21 comparison and I see that the sustained turn diagrams at least list the MiG weight as 7500 kg with 2x R-3S which, in the game anyway, equates to about 55% fuel and 2x R-3S. But perhaps I'm missing something. I would be glad to be wrong.

I also noticed that completely clean and with fuel set to 50% on the F-5E and MiG-21bis (infinite fuel in order to test STR), my buddy and I found that the MiG-21bis had a slight edge which seems inconsistent with the real data, as you have mentioned. What I wonder though is if you know whether the F-5 is underperforming or the MiG is overperforming. I haven't managed to find time to do the tests yet. Interestingly, if the MiG has any pylons, the F-5 will beat it pretty consistently.

[Smyth]

On 10/15/2023 at 1:05 PM, SgtPappy said:

You mentioned that the MiG-21 Russian manuals usually use 40% fuel load. I used the same manual for my MiG-21 comparison and I see that the sustained turn diagrams at least list the MiG weight as 7500 kg with 2x R-3S which, in the game anyway, equates to about 55% fuel and 2x R-3S. But perhaps I'm missing something. I would be glad to be wrong.

In mission editor, 50% fuel, 2x R3R, and cannon ammunition = 7703kg. 50% fuel with no external stores is 7480kg.

From "Pilot's Flight Operating Instructions" the operating weight with full internal fuel is 8620kg - this matches DCS mission editor. Weight with 2xR3 (or R13M, or 100kg bomb) is given as 8880kg in the manual. DCS mission editor gives 8840kg with R-3R specifically, which is consistent (because manual uses an average of a few loadings).

IRL manual gives 2750 litres of usable fuel at 0.775kg/l (RT Kerosene?) and DCS manual lists 0.78kg/l (TS-1 Kerosene?).

8880-0.5*(0.775*2750) = 7814, or 8840-0.5*(0.78*2750) = 7767

This is where 7800kg comes from in my comparison.

The 7703kg weight in DCS mission editor seems to be something fishy going on with fuel mass. Either the jet burns fuel at a different density than ~0.78g/cm, or the mission editor is showing total fuel, which is different from all other DCS modules that show usable fuel. If it is the second case I have no idea if the engine cuts out at 0% per mission editor or somewhere above, so what any of this really means I have no idea.

I originally used the 7700kg figure in my chart, but reverted to 7800kg after checking my calculations repeatedly. The conclusion remains basically the same in both cases.

2/2024 note: after reviewing more sources, I concluded the value of 0.78kg/l in the manual is probably wrong, and reverted to 0.83kg/l

 

On 10/15/2023 at 1:05 PM, SgtPappy said:

I also noticed that completely clean and with fuel set to 50% on the F-5E and MiG-21bis (infinite fuel in order to test STR), my buddy and I found that the MiG-21bis had a slight edge which seems inconsistent with the real data, as you have mentioned. What I wonder though is if you know whether the F-5 is underperforming or the MiG is overperforming. I haven't managed to find time to do the tests yet. Interestingly, if the MiG has any pylons, the F-5 will beat it pretty consistently.

I have not done any detailed sustained turn testing myself for these aircraft. @totmacher's excellent compilation of automated testing puts their peak sustained turn rate pretty close to reality though.

My concern with the MiG-21bis flight model in DCS is not the high-speed sustained turn behavior. It is the excessive maximum lift, plus low-speed energy retention due to implausible AoA behavior.

Edited March 1 by Smyth
added note on fuel density after comparing more sources

[SgtPappy]

3 hours ago, Smyth said:

In mission editor, 50% fuel, 2x R3R, and cannon ammunition = 7703kg. 50% fuel with no external stores is 7480kg.

From "Pilot's Flight Operating Instructions" the operating weight with full internal fuel is 8620kg - this matches DCS mission editor. Weight with 2xR3 (or R13M, or 100kg bomb) is given as 8880kg in the manual. DCS mission editor gives 8840kg with R-3R specifically, which is consistent (because manual uses an average of a few loadings).

IRL manual gives 2750 litres of usable fuel at 0.775kg/l (RT Kerosene) and DCS manual lists 0.78kg/l (TS-1 Kerosene).

8880-0.5*(0.775*2750) = 7814, or 8840-0.5*(0.78*2750) = 7767

This is where 7800kg comes from in my comparison.

The 7703kg weight in DCS mission editor seems to be something fishy going on with fuel mass. Either the jet burns fuel at a different density than ~0.78g/cm, or the mission editor is showing total fuel, which is different from all other DCS modules that show usable fuel. If it is the second case I have no idea if the engine cuts out at 0% per mission editor or somewhere above, so what any of this really means I have no idea.

I originally used the 7700kg figure in my chart, but reverted to 7800kg after checking my calculations repeatedly. The conclusion remains basically the same in both cases.

 

I have not done any detailed sustained turn testing myself for these aircraft. @totmacher's excellent compilation of automated testing puts their peak sustained turn rate pretty close to reality though.

My concern with the MiG-21bis flight model in DCS is not the high-speed sustained turn behavior. It is the excessive maximum lift, plus low-speed energy retention due to implausible AoA behavior.

Thanks for clarifying. I must have been looking at just fuel weight or something, I'll double check.

I can't believe I haven't seen this test data before. For those interested, here's the link https://dcs.silver.ru/. This is what you are referring to, right?

[Smyth]

41 minutes ago, SgtPappy said:

I can't believe I haven't seen this test data before. For those interested, here's the link https://dcs.silver.ru/. This is what you are referring to, right?

Yep, that's it. I neglected to include the link.

[Kalasnkova74]

On 10/14/2023 at 5:46 PM, Smyth said:

That's all true, although I would argue the -19P is still more dangerous because Aim-9B make it more difficult for faster aircraft to extend, and the radar only adds <<5% to the loaded weight of a MiG-19. The USSR either didn't realize how lethal a -17F or -19S with R3S would be, or did realize and didn't want to give that to their less-trusted allies. I know the US was 100% expecting to find MiG-17F with missiles in SEA.

I should have clarified that I don't think low-speed turns with a -19 are a really good idea in any version of the Phantom, and certainly not the hard-wing flavor. The slatted -E will just pose a non-zero challenge to a -19 driver in a scissors which is maybe a bit unexpected compared to similar-looking 50s swept-wing transonic jets.

As seen in the chart I posted above, the F-5E that faces the MiG-19 in DCS currently has really exceptional instant turn rate for a supersonic fighter. Even then the F-5E is quickly destroyed by an F-86 or MiG-15/7 in a slow turning fight, while a MiG-19 of any version takes much more time.

The Soviets didn’t export their best equipment to Hanoi primarily because of China. All equipment destined for that theatre had to go through Chinese territory, and Beijing was “skimming off the top” for their own R&D. Knowing this and other concerns like exposing technology to the Americans, they did not field their best hardware such as the SA-6 Kub or MiG-23. 
 

As to the low speed turns aspect…horizontally speaking, you’re correct. However correspondence between the Israelis & TOPGUN in the mid-70s revealed there are ways a Phantom can kill a MiG (or any other lightweight jet)  in the dreaded low & slow regime. I won’t be sharing those findings here- but you’ll see it if you get gunned in a slow speed scissors by my HB F-4 in the coming years.

Edited October 16, 2023 by Kalasnkova74

[Harlikwin]

On 7/9/2023 at 2:20 PM, Hummingbird said:

A JA37 would be amazing

Yeah, but sadly not happening due to DL issues. 

[Harlikwin]

On 10/13/2023 at 11:39 PM, Smyth said:

Since I already posted a version of the Phantom vs xxx turn performance summary I was working on for myself in other threads, it think it is only fair to put the full version here for the sake of everyone else interested in this topic.

This summary is strictly based on what I can find in publicly-available documents from the original operator (US or USSR), scaled for fuel state and altitude. That correction uses formulas I have validated against available data for aircraft where turn performance is charted at different weight and altitude. The derivation is not overly complicated, but too long for one forum thread.

Obviously this is not an "official" document despite the data sources, and should be taken for no more or less than whatever it is worth. However I am very confident that this is better than trying to guess based on comparing the un-altered primary sources with their different formats, units, and assumptions. For example the USSR typically uses 40% or less fuel in their charts, while the US uses 60% or more. These are both totally valid assumptions (representing either the end of a dogfight, or the start of one) however a direct comparison is not valid, since turn performance scales quite directly with weight, and fast jets carry a huge fraction of their total weight in fuel.

Finally there are three things DCS players should understand before I post this chart:

1. Data for the Mig-23 at 35 degrees wing sweep does not seem to exist. It's pretty easy to show based on back-of-the envelope calculations from aspect ratio what kind of difference that might make (measurable but not enormous), however that breaks the intent of this comparison to use only official data, so it is not included.
2. This chart is only relevant to the real MiG-21, and not to DCS, as anyone familiar with the F-5E vs. MiG-21bis balance in this game should quickly notice.
3. It is always subject to change if any better data is made public. Someone immediately corrected my F104 post with better data, so maybe (hopefully) that will happen here rule 1.16 permitting.

With all that said, here it is:

Questions, corrections, criticism, and flame wars are all welcome. Enjoy.

Post Script: there is "official" data (from McDonnell) that says the F-4E turns better than this, but I did not use that because I don't have proof it is correct relative to the USAF manual. I may be a Phantom Fanboy, but I try to be objective.

So whats the difference between the black and red lines?  And the 3/4 different graphs (is that alt?) 

With regards to the whole ML/MLA/MLD story. AFAIK, the ML manual is an early one, and it was revised for MLA with better data, also various things like G limits got upped between each one of those iterations, partly due to some changes in the airframes, but also from experience. From what I have anecdotally heard the 33 degree wing sweep got you like ~1-2 DPS more than the 45. Same thing for stuff like AOA limits as well. 

At any rate, that more or less means the F4 and 23 are relatively close performance wise for STR and ITR. Which is what RL testing from Israel anecdotally said (though they said their MLA/MLD (aerodynamically supposedly an MLA) the 23 outdid their F4. 

Also for all the GCI guys, while ofc the Soviets had GCI and used it doctrinally, they were basically horrified by how badly the Mid East air forces performed and basically setup their own version of Top-Gun, to teach BFM out in Khazakstan in the 70's. So more or less an analog to the US Vietnam experience. And by the early 80's you had experienced mig-23 pilots beating rookie mig29 pilots there, just like early on  F15's got beat by F4's or 104's in DACT. 

 

 

Edited October 16, 2023 by Harlikwin

[SgtPappy]

42 minutes ago, Harlikwin said:

So whats the difference between the black and red lines?  And the 3/4 different graphs (is that alt?) 

With regards to the whole ML/MLA/MLD story. AFAIK, the ML manual is an early one, and it was revised for MLA with better data, also various things like G limits got upped between each one of those iterations, partly due to some changes in the airframes, but also from experience. From what I have anecdotally heard the 33 degree wing sweep got you like ~1-2 DPS more than the 45. Same thing for stuff like AOA limits as well. 

At any rate, that more or less means the F4 and 23 are relatively close performance wise for STR and ITR. Which is what RL testing from Israel anecdotally said (though they said their MLA/MLD (aerodynamically supposedly an MLA) the 23 outdid their F4. 

Also for all the GCI guys, while ofc the Soviets had GCI and used it doctrinally, they were basically horrified by how badly the Mid East air forces performed and basically setup their own version of Top-Gun, to teach BFM out in Khazakstan in the 70's. So more or less an analog to the US Vietnam experience. And by the early 80's you had experienced mig-23 pilots beating rookie mig29 pilots there, just like early on  F15's got beat by F4's or 104's in DACT. 

 

 

 

All the plots are at 1000 m altitude. Each column of black graphs represents the aircraft on the top of the column (F-4E slats, F-5E, MiG-21bis and MiG-23ML respectively). Each row of red graphs also represent the planes in aforementioned order. I can see how the latter isn't actually labeled but if you look closely, you'll see how - for example, in the 3rd row - the red graphs match the final 3rd column MiG-21 graph. The black graphs in the 3rd row are therefore comparing the F-4 and F-5 to the MiG-21 turn rates.

[Harlikwin]

28 minutes ago, SgtPappy said:

All the plots are at 1000 m altitude. Each column of black graphs represents the aircraft on the top of the column (F-4E slats, F-5E, MiG-21bis and MiG-23ML respectively). Each row of red graphs also represent the planes in aforementioned order. I can see how the latter isn't actually labeled but if you look closely, you'll see how - for example, in the 3rd row - the red graphs match the final 3rd column MiG-21 graph. The black graphs in the 3rd row are therefore comparing the F-4 and F-5 to the MiG-21 turn rates.

Ah if thats the case I think I misunderstood them. One "interpolated" graph I have but don't think I can post has the 23MLA doing 14.5 dps STR at 45 sweep. It is interpolated/translated onto a western chart for comparison so I'm not sure if its correct. And if its true it "matches" what the Israelis have anecdotally said about the 23 and F4, but well, anecdotes. There is another nice one from a dutch F16A pilot that flew the MLD as well and said it outperformed the viper in BFM which is kinda eeeh, but does match what the soviets thought about it. 

All this being said, I think on release people will be surprised that the F4 can actually dogfight, and doubly true for the 23 since basically all of the western lore is about literally the worst first gen mig23 that ever flew, and while its somewhat accurate, its in no way accurate when discussing the ML/MLA/MLD versions. That being said I'll also say the same is likely true for optimistic assessments of how they though the 23 would fare against 4th gens. 

At the end of the day for DCS purposes, its all down to how well the Aero will actually get modeled, and how well the systems get modeled. Like if Jester 2.0 can pick out a gnats ass in lookdown, well thats likely a problem. Same thing if the MLA radar isn't somewhat janky but functional in lookdown.

Oh one last comment I saw mentioned some issues with instability at high speed with the 23. That was a problem with early gen1 23's but it was also at high mach, like 2.0 and above. And it was solved in the 2nd gens for the most part from my understanding you can see a big difference in the vertical stab between the two. 

Edited October 17, 2023 by Harlikwin

[Kalasnkova74]

19 minutes ago, Harlikwin said:

 

All this being said, I think on release people will be surprised that the F4 can actually dogfight, and doubly true for the 23 since basically all of the western lore ….

 
I doubt this. We commenters here are a tiny minority of the people who will buy both modules.
 

Remember the majority of DCS players got their start flying something advanced , like a MiG-29/ F-16/ F/A-18, Su-27 ,F-15 or similar. If you fly an F-4E the same way you’d fly an F/A-18 , your flight is ending with a ride on the ejection seat. Playing the horizontal game with an F-4E (or MiG-23 for that matter) is a ticket on the pain train. 

Getting A2A kills with the F-4 requires finesse and tactical understanding of using the jets power in the vertical. That is just not how DCS players do BFM with 4th Gen stuff, except maybe the sharper F-15E players. 
 

I suspect when the F-4E drops we’ll see a DCS reenactment of the errors US crews made between 1965 and 1969:people getting shot down playing 1 circle nose pointing contests. After a few months people will understand that’s a bad idea and start learning how to properly fight the jet, but on release it’s gonna be a great day to be a Mirage F-1 or MiG driver. 

[Harlikwin]

1 hour ago, Kalasnkova74 said:

 
I doubt this. We commenters here are a tiny minority of the people who will buy both modules.
 

Remember the majority of DCS players got their start flying something advanced , like a MiG-29/ F-16/ F/A-18, Su-27 ,F-15 or similar. If you fly an F-4E the same way you’d fly an F/A-18 , your flight is ending with a ride on the ejection seat. Playing the horizontal game with an F-4E (or MiG-23 for that matter) is a ticket on the pain train. 

Getting A2A kills with the F-4 requires finesse and tactical understanding of using the jets power in the vertical. That is just not how DCS players do BFM with 4th Gen stuff, except maybe the sharper F-15E players. 
 

I suspect when the F-4E drops we’ll see a DCS reenactment of the errors US crews made between 1965 and 1969:people getting shot down playing 1 circle nose pointing contests. After a few months people will understand that’s a bad idea and start learning how to properly fight the jet, but on release it’s gonna be a great day to be a Mirage F-1 or MiG driver. 

Well. I'm not gonna disagree about the average DCS MP players skill issues. But it will get sorted out. Most of the better players on the CW servers already handle F5's and Mig21's decently enough, though they are in the minority. Those will be the guys that take to the F4 or 23 like ducks to water. But your average DCS 4th genner, yeah they will have a bad day.