streakeagle

TK went from one extreme to the other: the original AI always acted surprised (more like comatose) unless tasked with some sort of anti-air mission AND at the waypoint where the mission was supposed to take place. That caused tons of complaints from just about everyone that ever played the game from the original Walmart release of SFP1. So, rather than make some complex AI that carefully evaluated what each pilot should know and how they should react to that knowledge, TK went for the simple solution: make the AI see everything all the time. Blind spots specified in data.ini files are worthless: not used by the game. I like to design historical missions from Vietnam using the Red Baron reports, but I can't simulate surprise. The moment a mission starts, all the enemies approaching from blind spots are automatically called out and either evaded or attacked. Aircraft with bubble canopies are penalized by this behavior. However, this all-knowing capability makes almost every dogfight fun/challenging at the expense of realism. Despite the cheats and limitations, SF2 AI is pretty much the most interesting I have ever fought against. Give the AI a good turn fighter, or better yet an aircraft that has power and can turn, and the AI gives me a decent run for the money!

Projected schedule from ED announcement: August 2014 – Fw 190 D-9 Dora October 2014 - BF-109K December 2014- Spitfire IX March 2015 - P-47D-30 May 2015 - Normandy Map with period AI units Other features like the Me.262A-1 and AI-only B-17G are also in development we but do not have a delivery data estimate at this time until these other features are further along. Fw190 might actually be delivered per the schedule, but that is easy because they waited until they were nearly finished to announce these tentative release dates. I know the Bf109K looked like it was coming along nicely, but I would be surprised if we got it before Christmas 2014 given past performance. At that pace, maybe Spit by spring or summer 2015 instead of the Thunderbolt. Leaving the P-47 to pop up during the summer or fall 2015? Normandy by the following Christmas 2015? Perhaps I am being to generous in sliding the schedule by only 6 months. Individual DCS aircraft tend to overshoot their projected release date by a year or more and multiply that by the number of airplanes after the nearly finished Fw190: 3 years to finish the main planes. Hopefully the map actually comes out sooner. I can see the Me262 and B-17 being 4 to 5 years away at the present rate of development.

Per Wags@ED, $40 backers like myself should get their Fw190D license ASAP. 1 down, Spit, Thunderbolt, and Bf109 to go. Dare I hope for the Me262? But how many months or years will it take to finish them all and the Normandy map?

If you have the fuel to do it, having the speed/power to avoid the enemy until he runs out of fuel works in reality, too. Duke Cunningham's 1v1 with a MiG-17F in the vertical didn't end in a win because he outflew the MiG, it was because the MiG dove away and tried to extend, presumably to disengage. A number of MiG-21s went down to lack of fuel in the Middle East. Unlike SF1 AI, SF2 fighter AI does have a tendency to perform evasive maneuvers even after running out of fuel and attempting to head home. But I can assure you that I was hearing stall sounds and/or departing to get my nose on the MiG-15s in an F-101. I think a falling refrigerator can turn inside a Voodoo!

There is no other sim that supports the Century series so well. DCS will never have this planeset (at least this century!). When I hosted and played SFP1/WoX online, quite a few of the people joining were somewhat new. So, I frequently flew the F-100D, F-104G, and A-4E while they learned to use the F-4 or flew much more maneuverable MiGs. With the advent of WoV, the F-105D was added to my stable of aircraft I could fly against newbies. Aircraft with major limits are fun and challenging for me. Though, it is nice if they have at least one strength that can keep you alive or even allow you a shot at winning.

In what other sim could you have this fight over late 1950s Europe? It was 2 vs 2. My wing died almost immediately as the AI isn't really programmed well to fly planes that go fast but can't turn. It was a game a patience trying to set up a shot without departing controlled flight. Altitude was always plenty high up for easy recovery. I came in fast from above, but throttled back and if necessary applied brakes to reduced range rate for targeting/firing. Eventually, the AI cooperates if you have enough fuel to wait it out. Took a while to bag the leader, but the proficiency gained doing so made it far easier for me to set up a firing solution on the wingman. As presently modeled in SF2, the Voodoo is extremely difficulty to employ in a dogfight, but has quite a bit of power. With well trained pilots and sound tactics, it could even be reasonably effective.

SF2 AI most certainly has some cheats. It can see you even when you are low and on its 6 o'clock. It knows when you are targeting it. But then TK made some cheats for the player after SF2NA: player missiles are more effective than AI missiles, particularly with regards to decoys. But SFP1 AI was just plain (plane?) stupid from WoV until WoI introduced what would evolve into SF2 AI. From WOI forward, missiles and radar locks had problems when used at low altitudes. Given a choice between SFP1 and SF2 for single player, I take SF2. For all of its faults (and there are many which didn't exist in SFP1 or earlier releases of SF2), the final SF2NA game is still the best for dogfighting with jets, particularly of the Korea to Vietnam era. My favorite aspect of flight sims is the flight model. SF2 still holds up fairly well in this area. But ED has finally surpassed SF2 in the area of detailed/realistic flight modeling. The DCS flight models are sheer joy whether it is a prop like the P-51D, a jet like the A-10C, or a helicopter like the UH-1H. DCS flight models are not dumbed down to make them more accessible. But, if you flew in the First Eagles beta testing or flew the various marks of the Lightning in the expansion pack, then you know that TK could make extremely challenging and realistic flight models with no changes to the SF flight engine. I primarily fly DCS now because of my love for detailed/realistic flight modeling. But I won't stop flying SF2 until someone gives me a sim that models F-4 vs MiG-21 combat better. An interesting comparison is to fly SF2 F-15A vs MiG-23MLD and then fly DCS:F-15C vs MiG-23MLD. Aside from the big difference in the terrain rendering (DCS is easily a win even with highly modded SF2), the main difference is in the AI performance. Despite the cheats or perhaps because of them, SF2 is a bit more fun: less predictable and better tactics. But, the solution is to play DCS online with skilled human players, which SF2 cannot do at all. SFP1 was fun in multiplayer, but was nowhere near as flexible and capable as DCS multiplayer. But in recent years, I don't do multiplayer at all, which leaves SF2 as a viable game for me. I have never stopped playing SF, just slowed down quite a bit while a wait for DCS to surpass it. Aside from AI issues, DCS needs a much more diverse flyable planeset and a lot of new terrains with appropriate period objects. These problems are being solved slowly but surely, but not any time soon. So SF2 is going to stay on my hard drive for some time to come. I wish TK would continue working on it.

That doesn't shock me. The first time I went into LOMAC to build a custom mission and discovered that it didn't include the MiG-21bis, I was shocked. LOMAC/DCS included the F-4 and F-5, but not the MiG-21? That was a crime! Even though I am not much of an A-G guy, I love the Skyhawk and if I am happy with the flight model, I will fly it frequently. At some point, I won't have enough time to enjoy everything I have bought. Presently, I timeshare between the P-51D, F-15C, UH-1H, and Mi-8. I fly a bit of MiG-29/Su-27/Su-33 as well. I also flip between basic combat missions and start-up/take-off/landing practice. I have a very simple reason for not flying the A-10C that much: I like to fly it with the Warthog HOTAS, but my setup is dedicated to using my F-4 Phantom B-8 stick, which just doesn't work with the stick-hat intensive A-10C control scheme.

Where on Earth is the info showing that the US claimed kills based on 2-6 hits of 0.50 cal on gun camera footage? The 792 to 78 ratio is a simple total of official "verified" MiG-15 kill claims versus official admitted F-86s for the entire war, not one specific time period. When the "official" USAF numbers are used over time periods, it gets pretty ugly for the USAF during the final months. The USAF was flying predictably and the MiGs adjusted their tactics to spank them bad, diving from above to catch aircraft departing with low fuel. My focus on the 0.50 cals "ineffectiveness" is against the Russian pilot's bragging about MiG-15s watching the 0.50 cals bounce off of them, which is an easy way to explain how the USAF thought they got 792 kills when they only fatally destroyed about 200. But that would indicate a really poor standard of verifying a claim. In Vietnam, someone had to witness the target hit the ground or enter a cloud in a non-recoverable attitude/altitude, such as vertical at low level, or the gun camera had to show fatal crippling such as wings being blown off or the pilot ejecting. A lot were verified by calling out visuals on the parachutes of the downed pilots. Given the similarity in the appearance of the F-86 and MiG-15, I could believe that there were more than a few confused situations that would have made friendly fire likely. I would hate to think that it actually resulted in downed aircraft more than a handful of times. As my Vietnam era aircraft/aces Osprey library is nearly complete, I am going to start collecting the Korean era aircraft/aces books and get more up to date analysis as well as some entertaining anecdotes. Osprey books may not be perfect, but there are few other sources that go into such detail about each enagement. I just got the USS Constitution vs HMS Guerriere duel book and was thoroughly impressed by the quality and quantity of information concentrated in such a small book in a format that was easy and entertaining to read. I had previously read every book I could get my hands on about the USS Constitution and "Preble's Boys" and still learned quite a bit more.

Simply pulling g is being evasive against guns with lower rate of fire and lower muzzle velocities. It takes skilled gunners to pull the necessary lead with the necessary precision to use slow firing, low velocity, heavy cannons. Play any decent WW2 sim and try using German cannons, especially the 30mm versus using 0.50 cals or high velocity Hispano 20mm cannons. The MiG-15 was a hotrod design, but with several key flaws. As with many US planes, its key flaw was simply being large and heavy. If the 0.50 cals were so truly ineffective and the larger cannons so effective, how could the US still get a better than 1:1 ratio against comparably trained USSR pilots who had the acceleration at high altitude to have energy advantages in speed and altitude? If the USA had successfully mounted the 4x20mm package prior to the war, would the kill claims of 792 been true? What is not really clear is exactly why the overclaims were so high. If the rules for making a claim are good, you don't get credit unless reasonably certain from gun camera footage or an independent witness that the aircraft was fatally disabled or completely destroyed. Based on recent studies comparing North Vietnam's claims and losses to the USA's claims and losses, there was very little overclaiming and even some underclaiming of enemy kills. The problem area was identification of the cause of a loss. If in doubt, the USA would claim flak or SAMs, which in a war with relatively few air-to-air fights and losses, artificially inflates the kill to loss ratio.

Interesting information. While the kill ratio drops from 10:1 down to 2:1 (actually a bit less at 1.8:1 overall), keep in mind that the aircraft are almost equal and in the key areas of climb, acceleration, and ceiling at high altitudes where combat took place, the MiG-15 had almost everything in its favor. The armament comparison fails to mention how the MiG-15 guns lethality was degraded by rate of fire and accuracy against an evasive target like the agile F-86. So, given all the limits the F-86 pilots were up against, a 2:1 kill ratio indicates that their training/skill/tactics were still the key difference, as further reflected in the lower 1.3:1 ratio against veteran Soviet pilots. In Vietnam, we had numbers and to some extent aircraft quality on our side, and still only averaged 2:1 due to a decrease in USAF pilot quality. With the same principal fighter (the F-4) or even one with slightly lower performance (the F-8), USN pilots could get a much better kill ratio on the order of 4:1 when trained (all F-8 pilots and later F-4 pilots), or do about the same 2:1 as the USAF with intercept-only trained F-4 pilots. In the end, the hardware mattered little, it was the pilots and air combat doctrine that had the greatest effect on the kill ratios. Political restrictions in both Korea and Vietnam kept the USAF from doing as well as it did in World War 2. You don't win air superiority by engaging in large furball dogfights or small duels, you win it by bombing/strafing the air fields, which allowed prop planes to overwhelm even the much superior Me262. Unable to cross the Yalu in Korea and just flat out denied the right to take out airfields in North Vietnam were costly restrictions.

Got the answers I wanted from RAND, and like Vietnam, there was much to be learned from studying the enemies reports on kills and losses: Recent Scholarship Reveals MiG-15 / F-86 Exchange Ratio Much Closer Than Traditionally Thought • For decades Western sources reported that USAF F-86s achieved kill ratios as high as 14:1 against the MiG-15 – Lop-sided kill ratio claimed to be the result of superior USAF pilot training, experience and tactics – Research conducted since fall of the Soviet Union casts doubt on these claims • Indicates actual number of MiG-15s shot down was just over 200 vs. almost 800 claimed by USAF • Overall kill ratio likely closer to 1.8:1 with F-86 kill ratio against Russian flown MiG-15s likely 1.3:1 • Why the big difference between USAF claims and actual MiG kills? F-86 Armament a Key Factor • F-86 designed as an air superiority fighter – Primary mission to fight other fighters – Designers believed six M-3s .50 in machine guns would be sufficient armament • Developed versions of M-2 .50 in machine guns of WWII with increased rate of fire • Fired 43 gram (1.5 ounce) projectiles with ~ 1 gram of incendiary composition in nose –Effectiveness reduced above 35,000 ft (where most Korean War engagements took place) • MiG-15 designed as a bomber interceptor –Carried heavy cannon armament • NR-23 23mm cannon fired projectiles weighing 175 grams (6.2 ounces) with 19 grams of HE – NR-23 hit ~ 6 times as destructive as .50 in hit • N-37 projectiles weighed 729 grams (25.6 ounces) with 49 grams of HE – N-37 hit ~ 18 times as destructive as .50 in hit … and So Was MiG-15 Design • MiG-15 was ruggedly built – Self-sealing fuel tanks – Rear armor – Thick bullet-proof windscreen – Jet engine much less vulnerable to battle damage than piston engines of WWII fighters – Kerosene-based jet fuel less likely to ignite when hit than gasoline – In interviews after the end of the Cold War Yevgeni Pepelyaev, successful MiG-15 pilot of the Korean War stated: The US Browning .50-calibre guns bounced off our aircraft like peas! It was routine for our aircraft to return home having taken forty or fifty hits. – One crash landed with 200+ hits and was repaired and back in the air in 8 days • Postwar USAF study concluded: – On average an F-86 needed to fire 1,024 M-3 machine gun rounds to kill a MiG-15 • About 64 percent of an F-86’s normal ammunition load • Required just over 8.5 seconds for 6 M-3 machine guns to fire 1,024 rounds • Bottom line – lots of MiG-15s were hit, damaged and seemed to fall from the sky, but lived to fight another day

If you take the total MiG-15 claims by US pilots of 792 and divide it by the 78 Sabre losses admitted by the USAF, you get an overall kill ratio of 10:1. Of course, MiG pilots claimed to have killed about 600 Sabres. When you read detailed histories, the kill ratio bounced up and down throughout the war from quarter to quarter as tactics and hardware evolved. The ratio was actually going against the USAF towards the end when the MiG-15 pilots had perfected attacking patrols just as they were getting low on fuel and heading for home. If the Wiki entry is accurate: http://en.wikipedia.org/wiki/North_American_F-86_Sabre Not having read either of those sources, consider the implications if both of those 2:1 estimates are correct. Assume the US is correct in assessing 78 losses in air-to-air, then F-86 pilots only shot down at most 156 MiGs. From my studies of Vietnam losses that have been supplemented by the other side's data, I would bet that the USAF always assumed a loss was anything but an air-to-air kill unless a first hand eye witness said otherwise. Pilots that are shot down don't always know how they got shot down or don't survive even if they knew. In Vietnam, air-to-air kills were relatively rare compared to AAA and SAMs, so if in doubt, it could be assumed that a missile hit was a SAM or a cannon hit was AAA. Not only was that a legitimate assumption, it also has the added benefit of making the air-to-air kill ratio look a bit better than it might otherwise be. I would be curious to know exactly how those two references arrived at a 2:1 ratio. I am buying the Osprey book [32] to find out ;) Suddenly all those years of having the "10:1" in Korea versus the "2:1" in Vietnam have been erased. While the kill ratio in Vietnam could have and should have been better, it is suddenly no worse than Korea historically. By the way, the RAND report is interesting and can be read here: http://www.mossekongen.no/downloads/2008_RAND_Pacific_View_Air_Combat_Briefing.pdf

After flip-flopping a couple of times, I finally backed the DCS:WW2 project at the minimum level to get all of the planes including the Me262. So at some point, I should get access to the Fw190D9 and I look forward to trying out its flight model and dueling AI P-51Ds for the flip side of the dogfight I have been doing for months.

Are you ready for the DCS summer? Hawk, F-86F, MiG-21bis, and FW190D9 if all goes well?

The Skyhawk has been a work in progress that I was following in the DCS forums, but it is now publicly associated with VEAO. If the VEAO Hawk is any indication of the kind of product the A-4C Skyhawk will be, we are in for a treat. However, for me, the most important part is the flight model. The Hawk will be released with an SFM. I don't know how much time and effort it will take VEAO to develop its own EFM comparable to an ED PFM. I can only hope the Skyhawk will follow the same path: release it with an SFM to get it into players' hands then offer a PFM quality EFM as an addon when it becomes available. This business model may work better for 3rd parties, at least until they can get a handle on building detailed and accurate EFMs. For comparison, Belsimtek has gone straight for the PFM on its helos and provided the PFM for the F-15C. So, it is possible to produce/release PFM quality from the start with a bit of debugging after the beta release. So, overall, I am optimistic about the possibility of a DCS:A-4C Skyhawk from VEAO with the realization that there isn't even any kind of projected release date. I hope somewhere along the way a MiG-17F, even as an AI-only enemy, joins the party to provide a suitable opponent. Of course, the AI MiG-15 bis being supplied as an enemy of the F-86F will have to do in the mean time. Is it too much to ask for a Vietnam era aircraft carrier and terrain? I have visions of having the scope of Wings Over Vietnam combined with the detail/realism of DCS. DCS:Vietnam? Of course, I may be long dead and cremated by the time Edge and 3rd party terrains are released!

Unlike "the last of the gunfighters", the Thud drivers actually got plenty of guns kills.

There are certain aircraft that have always been on my "flyables" list for any and all combat flight sims. TK and/or the modders for SF sims pretty much covered them all, but nowhere near the level of detail that DCS does. So, I am hoping DCS duplicates that success.

The water shader caused CTDs when exiting missions, and people are still having that problem. The final patch theoretically fixed that and some people are doing okay, others still have problems. Of course, DX9 has to be installed, which Win 7 does not have by default.

Great news! I live in Orlando about 30-40 minutes east of the theme parks. I have annual passes, so keep me informed about your Disney plans and if you don't mind I will come and meet you at one of the parks as time permits (unless I end up traveling for Thanks Giving).

TK doesn't want to even try to jump through the hoops Steam requires and appears not to like their business model anyway. After having been burned by his experiences with old school big-name distributors, I am sure he likes the end result of being an independent: Every single penny you earn goes back to you. But Steam provides a great way to access a much larger audience, so any losses in profit per unit sold might be corrected with volume.

As far as I know Strategy First never had any rights to distribute anything after SFP1. WOV and WOE were backed by Bold Games, but obviously TK retained some sort of distribution rights because he has them for sale at his online store. But maybe Strategy First bought rights from Bold Games or TK? But why now? Apparently, the version being distributed is the same as the online store: 083006. This has caused problems since it is inherently incompatible with WinVista/Win7, but it is still a good patch level to start at since it gives you the option to maintain compatibility with all of the old mods. While it may not be selling like a new game, the Steam user base is so large that it has sold some copies: there are a few posts in the Steam forums. The posts involve how to patch WOE to get rid of the CTD after exiting a mission in Vista/7 and to check out CombatAce and Capun's site for mods. Of course, they also include comments like "it looks and plays like a 1999 flight sim" and about the limitations of multiplayer. If TK had bit the bullet and jumped through the legal and financial hoops to put the entire SF2 series on Steam, would the exposure have generated enough sales to justify the costs? Is it too late for TK to put SF2 on Steam and make any profit? It is sad that TK can make a better living with a free-to-play ad-supported mobile version of SF than the original PC game. Although, TK is probably happier with the game play of the mobile versions since it leans way more heavily towards the arcade "fun" game that could be played and enjoyed by anybody that he wanted SFP1 to be.

Your Sabres still look and work great in SF2. Nothing to feel inadequate about. One reason I feel TK never ventured into releasing a Wings Over Korea game is how well the free mod was done, and aside from the terrain, it was clearly centered on your F-86 variants.

I saw a tweet about this on the home page of Third Wire and looked into it. It appears that Strategy First cut a deal with someone to put Wings Over Europe on Steam for $9.99 and is currently on sale for $8.49. I am not sure why they would do that or who they made the deal with, but it leaves Third Wire hanging in the wind with its online store price of $14.99 Is there a glut of WinXP users that need this one particular older release but won't pay $15 for it? I know Strategy First never stopped selling SFP1 so maybe their market experience taught them they could make a bit more with WOE? Does TK get a cut this time? I don't see where this is going to make anyone any money, but good luck to all parties involved. Maybe the Steam exposure will help stir up new interest in the SF2 series.

I can't wait! It will be torture trying to decide whether to fly the F-86 or MiG-21 and also trying to stay proficient on the P-51D, F-15C, UH-1H, and Mi-8. Of all of them, the Mi-8 is the most challenging just transitioning into a hover before landing. I think I have almost got the landings down on the P-51D despite the difficulty of trying to do a soft-touch three-point landing with a tail-dragger.