can we use this technology to do scientific experiments(example : making The Large Hadron Collider, New medicines's reactions against virus and bacteria.)

maybe you can make the argument that the virtual world's model is based on real life physics and science, so virtual world can't be use for experiment, and i agree with most of this.

But can't we use fundamental laws of physics to see the outcome to any kind of reaction? by doing that we can get some result which maybe perfect or maybe not.

This is just one scenario in which we are using same fundamental laws as ours, what if we modify those laws and see results. This just an example what we can do with it. we can make various models for different kind of experiments(example : model that contain only laws or result of string Theory , or model which only contain work of biology(various system: nervous system , immune system etc)

  • 4
    $\begingroup$ What kind of "virtual world technology" are you talking about here? $\endgroup$
    – Cadence
    Feb 28, 2019 at 6:01
  • 1
    $\begingroup$ We can use computers to model things and tweek the variables, yes. What's the actual problem that you are trying to solve - can it be defined by mathematical relationships or what are you asking? $\endgroup$ Feb 28, 2019 at 10:42
  • $\begingroup$ @Agrajag according to all physicists Everything in the Universe Is Made of Math so yes we need to use maths in all scenario. $\endgroup$
    – earthling
    Feb 28, 2019 at 10:51
  • 1
    $\begingroup$ There are plenty of phenomena we don't have a single explanation or a single best model for. We don't know all the answers, but at least we can focus our efforts and seek an answer - but first it is necessary that the question be defined. I say again, what is the problem that you are trying to solve? $\endgroup$ Feb 28, 2019 at 11:07
  • $\begingroup$ We can extrapolate on this question by looking at times gone by: 100 years ago, we did not know about DNA - there was some known rules about how we pass on traits, though. A simulation using those known rules would never have featured the four nucleotides. But - the same simulation might have been able to simulate some evolutionary workings. --- > A simulation will never be able to uncover rules more basic than those put in, but it might uncover some that are more meta. $\endgroup$
    – bukwyrm
    Feb 28, 2019 at 13:42

4 Answers 4


Alright, first of all we normally discourage the asking of multiple questions on a single post as this site works on a single question, single answer model. That said, the answer to all your questions is the same;


I could break them all down individually but in the end they are all just facets of the same problem, which is that you can't live in a virtual world completely and survive. You still need to eat. You still need shelter. You still need a job so that you contribute to a real economy because you're going to consume real resources out of that economy and you cannot help but do so.

Time is subjective to be sure; 'time flies when you're having fun' after all. But, time is a real resource in the real world, and the brain may well be capable of being stimulated really quickly, but that doesn't slow down time as the real brain perceives it. It just gives you more to do in the time you have on hand.

Economies are the same; your body needs real food and shelter. Living in a virtual world without nourishing your body means you'll eventually starve to death. In point of fact, dehydration will get you first, but the point is still valid; you need access to the real world to support your real body. Food. Shelter. Medicines. More. To access those you also need a real job. In other words, you're expected to contribute to the economy at least as much as you consume from it.

This leads to the question of war. If food or medicine or petrol or gold or uranium or whatever you need is scarce, you need it and your neighbour has it but won't give you any, war is inevitable. A virtual world doesn't change that fact for the same reasons mentioned above.

As for the science in the remaining two points, the important point to remember about a virtual world is that it can only model what we tell it to. More fundamentally, it can only model what we already know about. Science is about discovering what is unknown, or refining what is known to make it more precise or more accurate. Therefore, it is literally impossible to develop new science via a virtual environment because it can only model what is known, using the models we currently have. If a new scientific discovery turns those models on their heads, the virtual reality cannot model the new understanding without being reprogrammed to do so. If it isn't, then any effect we would observe in that virtual environment would not reflect the new understanding and would merely reinforce the thinking as laid down in the model it currently contains.

This is not dissimilar to questions that sometimes arise on games like World of Warcraft or EVE Online as to whether, with that many participants, they should create their own virtual country. If you can eat the food contained in EVE, use the medicines there, and only consume services that the game provides, then I say fill your boots.

For the rest of us who need to eat, wear clothes, live in houses, drive cars, and seek treatment from real doctors, we need to get busy contributing to the economy that will support those needs.

That's the one here in the real world.


When I was in the University and was attending the course on Engines, the professor, one of the inventors of the Common Rail for Diesel engines, was busy in a research to optimize the design of injectors nozzle.

They had a nicely made set up with which they could take pictures inside the cylinders during combustion. They usually did some simulation based on known physics laws to optimize the design, and then run a real implementation for confirmation.

Well, the images of the simulations showed a nice injection splitting in 6 jets, atomization of the flow and combustion. Then the reality of the combustion showed a messed up reality, with unbalanced jets, partial combustion and what more. His self mocking conclusion was: when reality doesn't fit your simulation, change reality.

The "problem" with simulation is that they do what we think they should do based on the model we have fit in. So, if we built a simulation of dynamics based on the Scholastic theories (the light things go up, the heavy things go down) we would see exactly that. In other words, if we create a simulation where 2+2=5, the simulation will always tell us that 2+2=5. So we could grow confident that 2+2=5, we see it in all the simulations!

According to Popper a scientific theory is such if it can be falsified: it shall be possible to make an experiment for which the prevision of the theory fail. And the experiment has to be made in the real world, where Nature can do what she pleases, not in a simulator where we tell a model what to do. So, in the real world I would take 2 apples, put them together with 2 more apples and count that I have a total of 4 apples. Whoops, my theory said I should have 5 apples! My theory is wrong!

Virtual models can do a lot of previsions, but then the reality is what tells us which are correct and which not. Remember Galileo's Eppur si muove!. The model cannot contain what we don't know.

  • $\begingroup$ I think it's depends on what kind of model we are making for simulation , it would take lot of hard work if want make even small system. $\endgroup$
    – earthling
    Feb 28, 2019 at 7:44
  • $\begingroup$ @earthling - please use this site as intended: One Question - Multiple Answers. Please do not fundamentally change what we are talking about, otherwise all the hard work in the answers is for nothing. Also, please see the 'Tour' for this site. Also, please incorporate your own research into your question BEFORE the answers come in. $\endgroup$
    – bukwyrm
    Feb 28, 2019 at 13:38

Since you're talking about a written work, rather than asking about real world theoretical, the answers are

  1. Yes. Though it's important to state that you don't have to. If you want lag to be a common experience in your virtual world, there's plenty of precedent. But it could also proceed at the speed of thought.

    Note that to satisfy readers like Tim B II, you'd need to either talk about how the overhead activities the participants need to engage in and how the time they take balance off with the gains they get within the virtual world, or you need to come up with something like the sustenance pods from the Matrix. But you'd probably want to reference an earlier source than that movie due to IP and stuff.

  2. Yes. Though it's important to state that you don't have to. If the designers of your virtual world technology prefer a dystopian virtual world, they could even make it worse.

    But a great deal of our economics are wrapped up in things that can be automated and entertainment - which the virtual world technology can directly address. Virtual world technology could also theoretically harness the brain power of participants, giving them a way to create wealth while engaging in the same activity that reduces their need for more traditional entertainment.

  3. That's up to the author. Depending on implementation details of the virtual world, wars could be made impossible, people inclined to war against each other could be put into separate virtual realities so they wouldn't need to butt heads, so there's no inclination for war. People inclined to war could get stuck in the virtual reality so they can't come out and fight... Or there could be a war about that precisely - outrage at people who've had friends get trapped in the virtual world.

  4. Probably not. Virtual reality is more akin to navel gazing, more or less. It could be used to facilitate people collaborating on discussing stuff. But adding an abstraction layer is the opposite of escaping an abstraction layer.

    But just because it doesn't sound plausible doesn't mean you couldn't write around that difficulty. If the world of your story is secretly a virtual reality simulation itself, a massive virtual reality built within it could conceivably overload the outer virtual reality, or could possibly trigger glitches to show up in the outer reality - especially if the minds of the people in the real world that are participating in the "real" world are being harnessed for some of the processing needed to implement the "real" world. By having people get used to using their minds to change things in a world they know is fake, if the world that they think is real is fake, too, those skills could translate. If you wanted.

  5. Sort of yes, but not really. Virtual reality is awesome at modelling physics experiments if your team takes the time to be all fiddly with the math. But the thing is, when it comes down to it, math is only a model of the real world. Your virtual reality could theoretically copy the mathematical world to a T - but it's no better of a model than math is. Anything that you could do to improve the virtual reality model would necessitate an improvement in the mathematical model. If you managed one by chance, as soon as it's determined to actually be an improvement - the mathematical model just improved, because your virtual reality is all math.

    That having been said, the mathematical model is vital to all physics research, and a good virtual reality would be a very good tool to help make the mathematical model more accessible. But a not so good virtual reality - at least in this sense - would not be helpful. For example, I've heard of scientists using Second Life to collaborate on their research. Minecraft seems like a much less plausible science collaboration tool.

    Virtual reality could also make enhanced communication possible - either by allowing one to experience more time, or by removing language barriers - possibly even allowing direct mind to mind data transfers without bothering with messy human languages. This enhanced communication could also be used for enhanced experimental control in the real world. If a user in the virtual reality experiences 10 seconds for each second that passes in reality, and that user was given a virtual interface that controlled an experiment in the real world, they could potentially react something like 9-20 times as fast to the events of the experiment. (There would be some lag translating what happened in simulation to implement the actual control. But this is balanced by the fact that you don't have to care about the signal propagation time from their brain to their hands to actuate the control. If the simulation can react based on their actualized will, it could activate the control quicker than a reflex could fire.)


Someone gotta program the thing

Since we can't program unknown rules into the simulation, so the idea of discovering new rules inside it become a bit silly. We might also be wrong in what we currently believe to be the rules, which introduce bias to the virtual world we create.

As such, it's great for confirming theories within those rules. If we have a starting point A and end point B, we can check if our current rules allow us to go from A to B.

Working within the rules

However, that doesn't mean we can't use virtual worlds to discover anything. We already do use heavy machine learning and simulation to discover combinations of materials, proteins or elements that result in new, interesting and desired properties. This then need to be confirmed in the real world, as the simulation might be wrong.

When it's right, then the we essentially discovered it within the virtual simulation. When it's wrong, we update the model with new data and try again.

  • $\begingroup$ you can discover new rules base on proven rules or facts $\endgroup$
    – earthling
    Feb 28, 2019 at 8:43
  • $\begingroup$ @Earthling Say you don't know light behave both as a wave and a particle. How do you suppose the simulation starts to treat light as a wave, when all code define it as a particle...? $\endgroup$ Feb 28, 2019 at 11:31
  • $\begingroup$ @Earthling you can infer rules based on data from real-world experiments yes, at which point you simulate them to test if they hold up. That's different, because you write the rule into the simulation, you don't discover them within the simulation, and you'd try to confirm it in the real world afterwards anyway. $\endgroup$ Feb 28, 2019 at 11:37

Not the answer you're looking for? Browse other questions tagged .