The original F1 was around 30+ Million horsepower and there was five of them lit up for take off. I was at the Cape for a early launch and I was 5 miles away and it was scary. A friend Bill Dodge was one of the Lunar Lander engineers and I was his escort. Rocket thrust doesn't relate well to horsepower but Bill told me that 30 million was about right. Maybe Brett can update us.
Ed
Horsepower is not a good way to measure it in this case, because the thrust is more-or-less constant but the velocity changes. Figure 1.5 million lb thrust (which is the early low rating, it eventually got up to 1.8 or so at sea level on the later models), and the first stage accelerates it to 7500 feet/sec. 1500000*7500 = 11.25 billion ft-lb/sec = 20.45 million HP for each engine at the end of the first stage burn.
The reason that HP is not a good measure of a rocket engines performance is that if you just care about the engine itself, it's not representative. At liftoff the HP being generated is miniscule, and at cut-off it's 20.45 million HP, because the rocket velocity matters. If you just launched the first stage by itself, the final HP would probably be three times as high with exactly the same engine performance. Running on a test stand, the engine HP is *zero* regardless of the thrust.
This gets into the extremely gross misunderstanding of what horsepower actually means among most people and is rampantly misapplied in stunt discussions. If nothing is moving there is NO work being done, and since horsepower is the rate at which work is done, if it's not moving there is NO horsepower.
Rocket engines are rated by thrust and specific impulse. The thrust is obvious, it's how much force is applied at the mounts. The specific impulse (ISP) is how much momentum changes for a unit mass of propellant. The momentum is the thrust x time (figure lbF and seconds). also called the impulse, units of lbf-sec, and the propellant mass is lb, so if you ignore niceties of comparing force to mass, the units of ISP is seconds. It's a measure of fuel efficiency for a given thrust.
A really good chemical engine like the upper stages of the Saturn V have an ISP of about 425 seconds. The first stage has a lower ISP of only 263 seconds. The difference is the fuel - the upper stages use liquid hydrogen and the first stage uses RP-1 (more-or-less ultrapure Kerosene). Both use liquid oxygen as the oxidizer.
This seems like a mistake and that they could have made it work better by using liquid hydrogen in the first stage, too, and nearly doubled the performance. But that is not the case. Liquid hydrogen is very efficient in terms of mass but it's also not very dense. So to get enough of it, the tanks have to be very large AND have to have insulation, which weighs enough more than it offsets the fuel efficiency. All that Orange foam that they spray on the Shuttle external fuel tank (and that sheds, and ended up destroying Columbia) is because of the liquid hydrogen.
Kerosene isn't as fuel efficient but it is dense and "space-efficient" so the tanks can be much smaller. For the upper stages it's a different story because they had progressively less dead-weight payload so the tank mass didn't matter as much. Still, the tank design and construction, particularly the second stage, were very difficult problems to solve and ended up being the limiting factor in the Saturn V development.
The thrust and ISP also change as the external air pressure changes. You get the highest ISP if the pressure at the exit is the same as the atmospheric pressure. The engine outlet area determines the exit pressure, so in a vacuum you want a GIGANTIC (or really, infinitely large) exit. Engines designed for sea-level operation have small exit bells. Since the pressure during the first stage operation goes from sea level to essentially a vacuum, there is a compromise size that works best, and the thrust and ISP change as the flight goes up.
Brett