that depends on how much data AR and VR use. there used to be a standard called UMS or universal matrix standard. basically a library of animations, sounds and similar that was used to present the matrix. then came VR2.0 with its sculpted systems that had custom stuff.

there is also the human brain to factor in. often you do not need neon tiger detailed down to the pixel. it may well be that instead the simsense can trigger the brain to think about neon, tiger and similar, and the brain will create the experiences itself.

in some ways thats how written text or speech works. as long as we both understand the same language, i do not have to provide a image of said tiger, i just have to describe it using words and you will get a mental image of it.

hell, look at vector based graphics. or you average web page. rather then define every last pixel, one use codes to define where lines and similar start, end and how they are to be reshaped. the software that reads the file then use that to generate a image based on said codes. as long as all programs interpret the codes the same way, you get the same image.

one may say that a sculpted system use the same UMS codes in the background, but present a different interpretation of them. that may explain the conflict rules between a sculpted system and a reality filter in older SR rules.

It's too early for me to look stuff up in the book, but I'm pretty sure the book says that they had to make security less realistic so that hackers could breach security in minutes or seconds instead of months or weeks.

im sure that if they could kill you using software and get away with it, they would invest in IC rather then fixing buffer overflow errors (99% of the security problems come from those. and yes, i made that statistic up on the spot)...

thats more or less what they said about encryption in the faq...

thats what you get when management get mixed up in the decisions best left to the IT department...

or in other words, blame it on crash 1.0 and the corp led rebuilding of the net...

Ha! I'll only be 97!

To be fair, I fully expect 97 to be the new 67 in 2070.

Oh yeah, that's why I just hold my nose and look the other way when dealing with the rules. Because this is one of those cases where Realism™ would suck royally.

Maybe, I would guess 87 as closer to the new 67. I imagine that it is reasonably possible that I'll make it to 102, I had 2 grandparents live into their 80's. However However their functioning level, physical and mental, for the last 2 to 3 years of their lives is why I through in the qualifier about being able to evaluate.

Actually it is absurd. You can postulate quantum computers that break some public key systems (but not all) but assuming that they can be run an a handheld takes some doing.

Anyway, even with that, it doesn't allow you to attack symmetric key systems (like AES) very effectively. Even throwing absurd computing power at at you can't break a 256 bit system in less than geologic time. Somewhere I have the spreadsheet, but if you assume computers are a trillion times faster each, with 1024 dedicated crypto processors on each of 1024 processors with 1024 nodes attacking you can break it in a mere 23,000,000 years. It's 8*10^45 years if you go to 512 bit keys.

Assuming you do the keys right, which is actually a huge assumption. Crappy keys makes attacks much more possible. And most people, left to their own devices, will choose crappy keys.

I'd personally vastly prefer a hacking model that was based on social engineering like that, along with installing funky hardware or malware once you manage to gain access to areas you should never be allowed into.

An example:
Banking on Security

It is. We've done it. But in general it's not an effective way in SR4.

In SR4 hacking is so effective (per RAW) that it's almost never worth the effort. If nobody can ever catch you when you drive a bulldozer through the front door and haul off the safe why should you choose an approach where you have to worry about security systems, locks and armed guards?

It probably comes down to error-rate. Even the most well crafted Decision Tree/expert System will pick the wrong pre-scripted action to take a certain percentage of the time, especially in the situations when, based on the inputs the expert system was programmed to consider relevant, there's not a dominant choice between several pre-scripted actions.

By putting a human in the loop with the expert system, and exposing the human to all the incoming inputs, not just the ones that the expert system driving the persona is coded to consider to be important, you gain the benefits of the human's skills and intuition for those times when the expert system is having trouble picking the best response, while still getting the speed benefits of expert-system driven pre-scripting.

As for why those computer-code based inputs must be filtered through VR Iconography, instead of being presented directly to the human in the loop, it comes down to the sheer volumes of information needing to be presented.

At least, those were the explanations that I gave in previous editions. In SR4, where you have the capability of coding agents that are essentially just as skilled in hacking as their human counterparts, it would seem that processing power has finally progressed to where an expert system can be complex enough to make the right decisions as often as a human would. It's one of the reasons that I'm not too troubled by the fact that using AR can be just as fast (and in some cases, faster) than using VR.

Except for that annoying page 217, where they say that this precisely what they don't do.

"The vast majority of Matrix activity (data traffic,
background processes, etc.) is highly uninteresting and would
quickly overwhelm your senses, so the bulk of it is fi ltered out.
Instead, basic AR Matrix perception is usually limited to a very
narrow subset of things, such the icons of nodes/users you are
interacting with, menus, dots, arrows, and any display features
you call up."

That's AR Matrix perception.

"Virtual reality is a drastically different experience than AR." SR4, p.228

Though to be fair, previous edition's descriptions of VR mentioned similar "filtering" of unimportant items. However, even if you assume that the expert system is considering every single input that hasn't been filtered out as unimportant, having a human in the loop adjusting the weight and importance that the expert system gives to certain inputs, in real time, based on a "big picture" view, can still result in higher performance than you'd get from an untuned expert system performing alone.

In previous editions, the difference between an expert system running by itself and an expert system that presumably made up the bulk of the Master Persona Control Program, running with a human decker fine-tuning the end results was pretty wide; you could send the expert system by itself, but you wouldn't have the same effectiveness of sending in a live decker immersed in full VR.

In SR4, with the increased processing power behind agents, taking the human out of the loop is becoming more and more of a viable tactic.

I'm not really saying either way. I mean we cant currently directly connect our brains to the internet. While VR technology exists its very crude and limited, bandwith is also limited etc.

I mean its -feasible- I suppose. Which in a way is saying it's 'realistic'. Though without actually having a real VR system to hook directly into the human brain directly through the neural pathways. It cant really be said if you can actually kill some one with a simple program, affect their brain with a virus or what ever.

Theoretically it -might- be possible. There's probably other theories that attempt to debunk this idea.

So all I could really say is its... 'sort of' realistic the way it's represented. Assuming the technology esists, and it actually works the way it 'theoretically' works under SR then yes it is. But we dont have anything actually like that right now, only vague similarities.

Not to mention the SR system cuts out alot of stuff that exists/can be done in the real world as far as the internet is concerned. This is however for the sake of simplicity. And certain parts of I I consider unrealistic, such as how you can crack a SOTA encryption program in a matter of seconds. If it was so damn good you might neve rmanage to crack it. Of course maybe you actually happened to manage to steal a copy of the proper decryption program. *shrug*

So to I guess sum it up another way, for the 'feel' of shadowrun and whats considered posisble in that setting, yes its got at least a 'semirealistic' feel to it. But it cant possibly be actually realistic (At least not yet) since in many cases there isnt any 'real' technology behind it, real technology is simply the basis behind the idea that is the matrix.

Don't forget the unlikely but canon "one generation knowledge/technology setback" caused by the crash of 2029.

(I still love how one of the characters in Shadowplay simply can't believe that the Corvette he's sitting in was built in 1990 - from his 2050s perspective technology couldn't have been so sophisticated back in the 20th century...)

VR (at least in the hot sim kind) also contained text stating that some kinds of data was translated as other sensation then audio, visual and similar. the user chould maybe pick up the tell-tale signs of a trace by what subtle sensations it would generate in hot sim...