If you are not interested in the modus operandi and the details of the results and prefer a brief overview, feel free to jump to the last chapter of this article.
Modus Operandi
The values collected include details about the fighter and the target, average and peak altitude both absolute and relative, launch and travelled distances, and speed details including peak and average. I also approximated terminal data, id est when the AIM-54 should go active or, as it happens, when the AI reacts to an ARH missile. This is an approximation because I did not collect such a parameter directly. Long story short, I took the average speed and altitude in the neighbourhood of the activation time. I converted Mach into TAS using the OAT at the determined altitude, and from there calculated VC. Then, I moved back from the impact distance and grabbed the related missile speed. The result is quite accurate, besides when the speed changes drastically in a very short period, often at very short range. Nevertheless, all terminal speed approximations are subject to the same algorithm, so they can be compared without issues.
Data and Results
Now that we know what we are looking at, let’s move to some examples, starting from the heavens above.
45,000ft
Let’s start by high and far. At 45,000ft and 80nm, with the Tomcat launching at M1.2 and the target flying at M.8, the Phoenix is in an ideal environment, where the air offers less drag, something this chunky missile suffers a lot otherwise.
The new Phoenix, identified as “b2 26”, since this was the second internal variant of 2026, wins under every parameter. It climbs to slightly higher altitudes, and reaches higher speeds. Both the terminal speed and the impact speed offer a solid increase of M.3. These parameters are now M2.79 and M2.44. Moreover, the missile arrives circa 10s earlier, something that is always good, moving the total flight time below the 2-minute mark, at 01:57.
There is not a lot to point out here, but observe the effects of the tuned rocket motor and avoidance manoeuvre, which make the acceleration smoother and more efficient.
Now, what happens if we Tomcat is flying at a lower speed, at the lower edge of the transonic region, say M.9? The first thing we notice is that the altitude reach is not as high. It is actually lower than the peak of the pre-update AIM-54. This translates in less energy in the terminal phase, and worse performance. Interesting, right?
Let’s quickly check the 60nm and 40nm datasets, same speeds and altitudes. In these scenarios, the trends continue, and the new Phoenix works better than the older when tested at identical speeds.
Overall, the first impression of the new Phoenix is excellent, thanks to the new rocket motor, and reduced drag.
At 20nm, one of the most important changes appears: the Phoenix now lofts within 21nm. Although not as impactful at 45,000 ft due to the properties of the atmosphere at almost 14km, this will change as we move towards the ground.
Note how, no matter the launching speed, the Phoenix is still accelerating past its activation range, impacting between M3.2 and M3.5.
Pretty good, so far.
35,000ft
Going down, we check the same set of distances at 35,000ft. Compared to the 45,000 ft set, we immediately see how the altitude reached by the new AIM-54 is lower than the older update. On the other hand, whether launching supersonic or subsonic, the outcome is better with the new Phoenix, with an impact speed between M1.9 and M2, and a terminal speed between M2.3 and M2.5.
At 60nm, the difference in terms of envelopes and speeds is minimal. At 40nm, the speed characteristic starts to resemble a triangle, losing that soft curve and acceleration as the missile transition from cruise to terminal. Moreover, the difference altitude-wise between old and new is widening. This means that the Phoenix now has slightly less energy in the last part of the trajectory, as it has less energy to covert from altitude.
At 20nm, we see that the launching speed is significant, but the slight climb plus the rocket motor improvements are sufficient to make the new missile the clear missile once again. This trend is constant throughout the tests.
25,000ft
Down another notch, we land at 25,000ft. This is the last time that 80nm is tested, as such a long range at 15,000ft is just too far.
Here we see a reiteration of the previous trend, as the new Phoenix flies lower than the older, but we also notice something new: the missile launched at a lower speed climbs higher. The launch at M1.2 still has the advantage at 80nm, but at 60nm, the new Phoenix performs worse than the older missile. Truth be told, the difference is minimal, but it is also a first sign that things are not as easy as the previous tests showed. As we know, in fact, the Phoenix’s secret is how high it gets, and how it then converts altitude into energy. This is how it reaches impossible distances for any contemporary missile.
In fact, there is an interesting point at circa 01:10 where the speeds reverse, and the fastest becomes the slowest and vice versa. This could be explained by observing the altitude curves: at 01:10 we begin the proper dive onto the target. It is no coincidence that the missile that maintains or even gains more speed is the one that reached a greater altitude.
I guess you are now wondering if the old Phoenix behaved in the same way when launched subsonic and supersonic. The answer is no-ish. It certainly changed slightly but, as the plethora of old tests I made and this quick one shows, the difference is not as marked.
At 40nm we see the trend continuing. The new Phoenix launched at M1.2 arrives slightly faster and at higher terminal speed. However, since the dive is not as pronounced, the impact speed suffers. Nothing too egregious, so far.
Lastly, the new loft mechanic at ranges below 21nm, along with the other improvements, allow the Phoenix to become a serious threat. Both terminal and impact speed see a substantial increase.
15,000ft
Moving towards thicker air, we start to see the impact of the lower altitude reached by the new AIM-54 Phoenix.
In primis, the 80nm test stops making sense, but the missile still reaches 70nm somehow. The three curves now visibly diverge, and the higher altitude peak reached by the Phoenix thrown at a lower speed benefits it the most. As unintuitive as it sounds, flying at a lower speed helps the missile more than flying faster. Which is good, I guess, as flying supersonic with such a heavy payload becomes quite taxing fuel-wise.
At 60nm, the trend is exacerbated even more. Now, the M.9 shot is faster both in terms of terminal and impact speed. Funnily enough, the old Phoenix was actually faster when activated, but it impacted at the lowest speed of the three, probably due to its incorrect drag value.
40nm, more of the same. The speed curves not have given up on the triangular shape. The “bulge” that was once convex, is now concave. This is really not good in terms of missile performance, as it makes the missile potentially easier to defeat kinematically.
At fairly short range, 20nm, the new Phoenix still easily wins, due to the already mentioned behaviour.
Note, however, that here the launch speed still reigns, even if the M.9 shot reaches a slightly higher altitude.
5,000ft
Down in the weeds, because weeds can totally grow up to a km and a half, shooting at long-range targets become unfeasible. The longest distance I tested is 50nm, and the Phoenix arrives there at a subsonic speed, making it de facto a non-threat besides niche scenarios.
Even at 40nm, we see a similar phenomenon. Speeds are slightly higher, but still vastly too low to be effective. Once again, shooting at M.9 results is slightly better performance due to the higher altitude reached by the AIM-54.
Quickly moving to the 30nm range, we see more of the same. The target is still too far, and even the combined closure rate of M2 is not sufficient to make the Phoenix a real threat.
Arriving once again at the 20nm set, we see again the new AIM-54 combined with the higher launch speed winning the comparison. The curves representing the speeds of the three missiles are quite peculiar and show the drastic impact of the rocket motor running out close to the 30s mark. Although the new update manages to loft the ’54s to an altitude greater than 3 km or 10,000ft, that “slice” of the atmosphere still induces quite a lot of drag on the missile. The subsequent dive does little to recuperate the invested energy. Still, it is better than the older version, and it can surprise a low-SA target, especially if the distance is closer to 10nm, rather than 20nm.
A Mess, But a Welcome One
As we have seen, the new Phoenix is better in certain combinations of range and altitude, but the speed at which it is fired also have a non-marginal effect. To help you understand which set works better, I put together the following tables. They represent the performance of the new Phoenix in different scenarios against the old AIM-54 present in DCS. Each arrow represents 10%, with the 5x representing 50% or more. The equal symbol includes ±2.5%. For example, two arrows pointing up mean that the advantage of the new Phoenix is between 10% and 20%. As a reminder, the old AIM-54s were shot at M1.2.
When we look at the results we see a stark division between 25,000ft and 35,000ft, and lower altitudes. Above such a threshold, the atmosphere’s characteristics offset the lower altitude reached by the missile, and at 45,000ft, the Phoenix gets actually higher than the old. Moreover, flying faster has a greater positive impact overall.
The next area we observe is the 20nm range. Here the new Phoenix wins easily, no questions asked. In both this scenario and the previous, flying faster helps to obtain better results.
Lastly, the area between 30nm, 5000ft and 80nm, 25,000ft is home to a series of unexpected results, where the new Phoenix performs worse if launched faster. I suppose that the reason is the lower altitude reached, which cannot be compensated enough by the improved drag and rocket motor performance.
Flying at subsonic speeds have a non-dramatic but visible positive effect, again, possibly correlated to the altitude factor.
Conclusions
The AIM-54 Phoenix has always been a complicated missile that rewarded proficiency as a Radar Intercept Officer. Now, it has become even more challenging. However, by combining what I described so far and remembering the areas where the combination of parameters is more effective, RIOs can still guide their stick monkeys to acceptable results.
One of the several topics I have not discussed yet in detail is the manual loft. I have data ready to be released, but I would rather not add more numbers to this discussion, I doubt anyone is still following at this point! Ergo, manual loft will be covered in a separated brief video and related article. If you can’t wait, here is a small spoiler for you: yes, manual loft helps a lot.
Leave comments and feedback in the Discussion Forum thread.