Preface.
A long running problem with E : D isn't that an RNG is used, it's that the RNG is obvious and in your face. It's just a fact of life, unless you're going to simulate an entire galaxy, an RNG will be needed at some point. The good news is that this is being addressed, for example, the new way unidentified signal sources work. It used to be very obviously random, which was immersion breaking (and made missions that needed them as much fun as trying to find your lost car keys). Now your search for your target USS is actually directed, and it's a great deal more satisfying and a lot more fun. Unfortunately, to replace this obvious and in-your-face RNG, we now have randomness with the engineers. Layers of randomness (randomness in finding the materials - at least most of the time, then randomness in the effect you get, then randomness in the special you get).
A possible way to address this.
Engineer mods should be completely deterministic. Instead of those indicators moving up and down after the "pull of the handle of the slot machine", you get to set the sliders. When you move one slider up, another (or more than one) of the sliders should move down as the consequence. There should be other consequences to certain modifications, especially if you push the mod to the module really hard. If you push a module really hard, the only non-deterministic effect should be *reliability*. The harder you push it, the less reliable it is. Think of it like overclocking your computer. At moderate levels of overclocking, there's not much of a consequence - you may need a slightly better heatsink and fan. Push it harder still, and now you really have to deal with the extra heat somehow. Push it further still, and then reliability will begin to suffer. But the overclocking itself is always deterministic - you'll always get the CPU frequency you select, the main tradeoff (heat) can be predicted. Reliability will be a lot less deterministic. Perhaps your machine will run fine, or perhaps the OS will blue screen of death just at a vital moment. Some processors will overclock great, and some others with the same speed grade and from the same manufacturer won't overclock as good.
The other consequence when you start asking for something too special - and which has already been considered for the special effects - is engineer reputation. If you're really wanting an engineer to push a module to its limits, you're being kind of a pain in the backside to the engineer, so it'll cost you some reputation if you're trying to really push it hard.
This means there are tradeoffs you must consider, this means that you won't get the "One True Build" due to reliability concerns if you try to make (for example) a weapon too good, but it will mean it's not a frustrating big "pull of the handle of the fruit machine" exercise where you can spend significant time looking for materials only for the RNG to give you a useless "upgrade".
A long running problem with E : D isn't that an RNG is used, it's that the RNG is obvious and in your face. It's just a fact of life, unless you're going to simulate an entire galaxy, an RNG will be needed at some point. The good news is that this is being addressed, for example, the new way unidentified signal sources work. It used to be very obviously random, which was immersion breaking (and made missions that needed them as much fun as trying to find your lost car keys). Now your search for your target USS is actually directed, and it's a great deal more satisfying and a lot more fun. Unfortunately, to replace this obvious and in-your-face RNG, we now have randomness with the engineers. Layers of randomness (randomness in finding the materials - at least most of the time, then randomness in the effect you get, then randomness in the special you get).
A possible way to address this.
Engineer mods should be completely deterministic. Instead of those indicators moving up and down after the "pull of the handle of the slot machine", you get to set the sliders. When you move one slider up, another (or more than one) of the sliders should move down as the consequence. There should be other consequences to certain modifications, especially if you push the mod to the module really hard. If you push a module really hard, the only non-deterministic effect should be *reliability*. The harder you push it, the less reliable it is. Think of it like overclocking your computer. At moderate levels of overclocking, there's not much of a consequence - you may need a slightly better heatsink and fan. Push it harder still, and now you really have to deal with the extra heat somehow. Push it further still, and then reliability will begin to suffer. But the overclocking itself is always deterministic - you'll always get the CPU frequency you select, the main tradeoff (heat) can be predicted. Reliability will be a lot less deterministic. Perhaps your machine will run fine, or perhaps the OS will blue screen of death just at a vital moment. Some processors will overclock great, and some others with the same speed grade and from the same manufacturer won't overclock as good.
The other consequence when you start asking for something too special - and which has already been considered for the special effects - is engineer reputation. If you're really wanting an engineer to push a module to its limits, you're being kind of a pain in the backside to the engineer, so it'll cost you some reputation if you're trying to really push it hard.
This means there are tradeoffs you must consider, this means that you won't get the "One True Build" due to reliability concerns if you try to make (for example) a weapon too good, but it will mean it's not a frustrating big "pull of the handle of the fruit machine" exercise where you can spend significant time looking for materials only for the RNG to give you a useless "upgrade".
Last edited:
