Which in itself comes from QA-Jonny in the bug report.
The real problem with RNG (and a solution!)
- Thread starter Spyro Jance
- Start date
Which in itself comes from QA-Jonny in the bug report.
Its funny, because i've had people tell me they consistently get below average rolls on engineers, when my own experience is that i consistently get above average rolls, which is in line with comments from the devs where they said there is a bias towards good rolls.
And i wonder, how many people here are exaggerating their point to absurd levels in order to make their point seem valid.
Yeah, its understood there are levels of RNG that are annoying. Some of it annoying for me as well. But when the exaggeration gets to silly levels, it just makes a mockery of the actual issue.
The randomness gods/goddesses can be quite cruel sometimes I must admit, I think people blow it way beyond proportion though.
Like the probes are a bad example because its an item that people typically collect one of and then never touch again once they are done with it, so you get statistics of 1,
Iv in my play time collected exactly 4 probes. My experience is varied.
1st time, i think i got a hit within 3-5 mins
2nd time, I got a hit quite honestly within 30 seconds of entering orbit
3rd time, it took about 40 minutes
4th time, it was about 10 minuets
So yeah all over the place. But say if you are that lucky person who got it after 30 seconds, as a none critical thinker you would simply say "Oh not sure what you guys are moaning about, I got it right away"
So its the same for engineers, Iv done a fair amount of it, and I can honestly say that there was only one session in which I was not really impressed with any of the rolls in comparison to some id done before (with the FSD range increase) Point is that a roll over stock will always give you an improvement for good and bad, its about accepting that.
Anyway - I find (quite clearly) these discussions quite frustrating because its very clear that people don't actually understand stats at all... then again, i guess i shouldn't be shocked by that, we are a society in which the anti-vaxxers actually get listened too.
Like the probes are a bad example because its an item that people typically collect one of and then never touch again once they are done with it, so you get statistics of 1,
Iv in my play time collected exactly 4 probes. My experience is varied.
1st time, i think i got a hit within 3-5 mins
2nd time, I got a hit quite honestly within 30 seconds of entering orbit
3rd time, it took about 40 minutes
4th time, it was about 10 minuets
So yeah all over the place. But say if you are that lucky person who got it after 30 seconds, as a none critical thinker you would simply say "Oh not sure what you guys are moaning about, I got it right away"
So its the same for engineers, Iv done a fair amount of it, and I can honestly say that there was only one session in which I was not really impressed with any of the rolls in comparison to some id done before (with the FSD range increase) Point is that a roll over stock will always give you an improvement for good and bad, its about accepting that.
Anyway - I find (quite clearly) these discussions quite frustrating because its very clear that people don't actually understand stats at all... then again, i guess i shouldn't be shocked by that, we are a society in which the anti-vaxxers actually get listened too.
Nonsense. Especially the part about CPU production. What has that to do with compelling game design?
The lower-to-mid grade mods, especially G3 and G4, definitely seem to be positively weighted. I've had some astonishingly good G3 and G4 rolls when building reputation, almost (but never quite) good enough to keep and forget G5s altogether, and very few if any really bad rolls. The G5s themselves are another matter; while they have a result range that is wholly positive, there doesn't seem to be any bias towards the top end of that range. Most of my really good G5s have been the result of hugely fortuitous secondary effects on mediocre primaries.
At least the probability distribution (bias aside) is basically linear. Can you imagine if the probabilities were, as they arguably should be, a normal distribution centred on the middle of the blue zone with "crap rolls" and "god rolls" massively improbable outliers? Sure, it would reduce the number of rage-inducing "wasted" rolls, but the min-maxers would be doing their collective nuts trying to push a roll towards the very edge of that bell curve.
Maybe we should be grateful for what we have.
I am going to go contrary to apparent popular opinion.
I love RNG in ED. It's perfect. I wouldn't change it. I'm not even joking.
It is really the reason I keep playing, and I am more and more convinced the same is true for many other players - they just don't realise it.
Just look at the countless videos on-line - of people going through pain after pain for engineering rolls - or mining materials - and the elation we get when it suddenly goes right. That sudden buzz justifies the next set of torture. And actually I start to enjoy the torture because of the anticipation. Disappointment redoubles my efforts - and redoubles my hours in the game. The RNG-God may not look kindly on me today - but I will just pray even harder...
If we could just move a set of sliders - I'd have stopped playing ED months ago. I spend hours upon hours playing against RNG.
BUT perhaps that is the point? Perhaps what we have here is a community of players addicted (through gritted teeth) to play a game, looking for that almost impossible perfect roll. If they had that roll given to them - they would no longer need to keep playing - they'd be free? They could play a different game?
Just my farthing or groats worth .... not worth a penny - I'll make tea.
I love RNG in ED. It's perfect. I wouldn't change it. I'm not even joking.
It is really the reason I keep playing, and I am more and more convinced the same is true for many other players - they just don't realise it.
Just look at the countless videos on-line - of people going through pain after pain for engineering rolls - or mining materials - and the elation we get when it suddenly goes right. That sudden buzz justifies the next set of torture. And actually I start to enjoy the torture because of the anticipation. Disappointment redoubles my efforts - and redoubles my hours in the game. The RNG-God may not look kindly on me today - but I will just pray even harder...
If we could just move a set of sliders - I'd have stopped playing ED months ago. I spend hours upon hours playing against RNG.
BUT perhaps that is the point? Perhaps what we have here is a community of players addicted (through gritted teeth) to play a game, looking for that almost impossible perfect roll. If they had that roll given to them - they would no longer need to keep playing - they'd be free? They could play a different game?
Just my farthing or groats worth .... not worth a penny - I'll make tea.
You mention "addiction" and that's pretty close. "Fun" is something else.
[video=youtube;tWtvrPTbQ_c]https://www.youtube.com/watch?v=tWtvrPTbQ_c[/video]
Please mate, i spent 9k rolls doing a 6A thrusters... RNGneers is the worst possible game mechanic.
It is not what keeps me in-game but push me out. The only reason i play time to time is hoping to have the roll i want and being done with RNGneers to finally play the game for fun.
- RNG grinding is tedious and boring
- RNG rolling is frustrating because it is wasting the time you spent doing tedious and boring grind.
These 2 links explain why RNGneers have to be reworked and RNG removed :
https://www.reddit.com/r/EliteDangerous/comments/6ft2z2/serious_engineers_and_the_grind/
https://www.reddit.com/r/EliteDangerous/comments/6g1ji5/serious_engineers_ironically_killed/
Last edited:
All of this sounds good, and voices many of the things people have complained about, but actually misses the real problem:
Computers Can't Generate Random Numbers
But don't take my word for that - let's see what Steve Ward, Professor of Computer Science and Engineering at MIT’s Computer Science and Artificial Intelligence Laboratory has to say - take it away Steve:
https://engineering.mit.edu/engage/ask-an-engineer/can-a-computer-generate-a-truly-random-number/
And now you know, and knowing is half the battle. The other half, the important half, is to continually and kindly urge Frontier to give us something better than pseudo-random numbers or random numbers based on the particle count in the Heisenberg Compensator at a non-specific time of day, but actual systems of development and improvement (Engineers) and more balanced and thought out (loot tables/mission rewards).
Computers Can't Generate Random Numbers
But don't take my word for that - let's see what Steve Ward, Professor of Computer Science and Engineering at MIT’s Computer Science and Artificial Intelligence Laboratory has to say - take it away Steve:
https://engineering.mit.edu/engage/ask-an-engineer/can-a-computer-generate-a-truly-random-number/
And now you know, and knowing is half the battle. The other half, the important half, is to continually and kindly urge Frontier to give us something better than pseudo-random numbers or random numbers based on the particle count in the Heisenberg Compensator at a non-specific time of day, but actual systems of development and improvement (Engineers) and more balanced and thought out (loot tables/mission rewards).
Blackcompany
Banned
Exactly.
This was supposed to be a space sim. Time for the RNG to go away. Try replacing it with deep, compelling game play Mechanics.
Blackcompany
Banned
RNGineers are what drove me over the edge with this game. I finally decided to try them...
...and discovered they are nothing more than poorly disguised mobile game wait walls and slot machine gambling. It's sad that what was advertised as a space sim has come to this, mechanically. It shouldn't have been this way.
The biggest problem with RNG is it's used too much in everything.
It makes the game feel lazily-made, first off. And in the case of engineers, it literally marginalizes player effort as the possibility exists for a player to get all the need for a Class 5 mod, and end up with results that would have left them better off trying an inferior range multiple times instead. And don't even get me started on random drop chances.
It makes the game feel lazily-made, first off. And in the case of engineers, it literally marginalizes player effort as the possibility exists for a player to get all the need for a Class 5 mod, and end up with results that would have left them better off trying an inferior range multiple times instead. And don't even get me started on random drop chances.
The OP sounds good on paper but I personally disagree with it.
It's difficult to explain why in the engineering example without a wall o text .. but illustration is what can happen in SRV where you might fall down a ravine from which there's NO escape. Say your ship can't recall down there, or you recall your ship and it lands on bumpy ground, you can't get under to dock SRV. This is chaotic but presents interesting problems to solve, maybe with some epic mountaineeriing.
So for me all RNG means is your plans can go wrong, forcing to think around the OCD and I found it interesting myself, when a combination of mods added up to too much power from the powerplant, so you have to rethink the great roll you thought you had. It's not quite the example the OP is using but I'm trying to say that there's interest in things being sub-optimal. All the OP wants is a God Roll, turning engineers into station outfitting where if you grind enough rolls (direct alternative to credits) you get your AAA grade, guaranteed? (Why not forget it? Engineers are better used, imo, for a tuning 'edge' anyway unless you actually like looking for mats).
RNG introduces chaos, might take some practice to live with but it's more interesting because you never get the same roll twice, no two loadouts are identical and we don't end up with fleets of identically optimised ships. My 2c.
It's difficult to explain why in the engineering example without a wall o text .. but illustration is what can happen in SRV where you might fall down a ravine from which there's NO escape. Say your ship can't recall down there, or you recall your ship and it lands on bumpy ground, you can't get under to dock SRV. This is chaotic but presents interesting problems to solve, maybe with some epic mountaineeriing.
So for me all RNG means is your plans can go wrong, forcing to think around the OCD and I found it interesting myself, when a combination of mods added up to too much power from the powerplant, so you have to rethink the great roll you thought you had. It's not quite the example the OP is using but I'm trying to say that there's interest in things being sub-optimal. All the OP wants is a God Roll, turning engineers into station outfitting where if you grind enough rolls (direct alternative to credits) you get your AAA grade, guaranteed? (Why not forget it? Engineers are better used, imo, for a tuning 'edge' anyway unless you actually like looking for mats).
RNG introduces chaos, might take some practice to live with but it's more interesting because you never get the same roll twice, no two loadouts are identical and we don't end up with fleets of identically optimised ships. My 2c.
[video=youtube;5cVLUPwrSmU]https://www.youtube.com/watch?v=5cVLUPwrSmU[/video]
Last edited:
If you did 9000 rolls it must have kept you in the game a little bit.
Or maybe there was an exploit that has been fixed
Or maybe there was an exploit that has been fixed
You mean you did 9000 g1 rolls? You just cheated?
Problem here is expectation and entittlement. A Grade 5 module with bad rolls is still better than a standard module. The problem is we know that for a small minority, they have better stats which causes desire for those rolls. In that case a "God Roll" is no better than a loot drop that gives you legendary gear. For example I have Grade 5 Dirty thrusters with an optimal multiplier at 34%, of coarse there are people that have 40% or better. So question is do I chase after those higher stats. There lies the issues. It's not RNG, it's not Fdev. It's your expectations. Is the game subjectively better if you can jump 50 ly or 60 ly.
Last edited:
Honestly forgot about that one. These days it's all about grinding the Rifts for legendaries, I forgot you could even use the Smith for something other than deconstruction. LOL
RNG introduces realism and unpredictable outcomes. To keep boiling soup off the same old bones: Maybe the range of randomness is too much, but the main issue is with the graphic illustration of it: The bars make the randomness look enormous, regardless of whether it is actually so.
I like that high level mods can go horribly wrong and fantastically right, and that the way they eschew whatever balancing there was really needs to be considered for other rolls.
So maybe change the interface. Make the visuals look more like it is a tuning shop and the changes indicated make them proportional to the actual changes. Perhaps balance the range in the higher end, if class 4 - 5 mods are too wildly variable.
But don't listen to the min-maxers that exaggerate their issues and won't make do with anything less than perfect rolls. If this is supposed to be a simulation, there should be unpredictable elements and slight variations between individuals of the same type of space ships.
S
I like that high level mods can go horribly wrong and fantastically right, and that the way they eschew whatever balancing there was really needs to be considered for other rolls.
So maybe change the interface. Make the visuals look more like it is a tuning shop and the changes indicated make them proportional to the actual changes. Perhaps balance the range in the higher end, if class 4 - 5 mods are too wildly variable.
But don't listen to the min-maxers that exaggerate their issues and won't make do with anything less than perfect rolls. If this is supposed to be a simulation, there should be unpredictable elements and slight variations between individuals of the same type of space ships.
S