[ravi-delia]

Not so much for takeoff! Most rocket designs better than chemical rockets trade off thrust for specific impulse. That's an improvement in orbit, since delta-v is delta-v. But imagine a 10kg rocket- it's receiving ~100N of gravity. If your engine doesn't put out 100N of thrust you'll just sit there on the launch pad. As you pick up speed you no longer have to deal with that (after all, LEO has basically the same gravity and doesn't have to burn against gravity at all) but when you're launching off something other than a point mass, some of your thrust has to go towards ensuring you don't hit the planet, or you will not into space today.

The practical designs we have for NTRs are solid core, which after long effort got up to a thrust to weight ratio of 7:1, meaning they could in principle carry up to 6 times their weight and accelerate up in Earth's gravity rather than down. Chemical rockets can get 70:1. No one ever had plans to use NTRs in lift platforms- instead they could serve as more efficient upper stage engines, for orbit-orbit transfer burns and the like. In principle there are engines which are technically NTR and offer much better performance, but no one's ever gotten a working prototype. Also you probably wouldn't want to launch with an open cycle rocket, since the open part describes how the radioactive fuel is ejected out the rear. Unfortunately, with the technology we have, we have to make tradeoffs between efficiency and thrust. For the lift stages chemical rockets are, for now, unrivaled.

(Unless of course your nuclear propulsion is of the more, shall we say, entertaining variety. Project Orion has its proponents...)

[lumost]

When discussing potential alien civilizations, one can’t discount the existence of civilizations which exist on substantially more radioactive planets.

If the background radiation of earth was 100x higher, would we care about an Orion launch? Or a small nuclear exchange…

[Dylan16807]

The more fuel you have to pile onto the rocket, the less the weight of the engine matters.

Using the chart in the accepted answer, launching with chemical engines takes 50 thousand tons at 3x gravity and 3 million tons at 4x gravity.

Now consider a theoretical engine that has a 7:1 thrust to weight ratio at 1G but sips fuel. Take a 25 ton engine, strap 10 tons of fuel to it and 1 ton of payload. Watch it go to orbit on a single stage.

A real NTR doesn't save nearly as much fuel, but it can still be useful in certain ranges.

[jowea]

I can't help but think that any species insane enough to use Orion drives in the first stage probably already found a way to blow itself up before it gets to that point.

And maybe I'm taking Terra Invicta too seriously but maybe they would wait until they figure out nuclear fusion and have more options.

[hermitcrab]

I once got to briefly discuss project Orion with Freeman Dyson at a book signing. IIRC (it was a long time ago) he said that :

- he thought it could be made to work

- all big engineering projects (dams, skyscrapers etc) kill people

- putting all that radiation into the earth's atmosphere couldn't be justified