Guitar is what I would call a complex instrument, because there are many factors to the tone/sound that you get. It's far more complicated than pluck, pluck, pluck as any beginner trying to play "Spanish Fly" for the first time can tell you.
As Phil stated above if you were to finger out an open G or a bar G then strum down the strings each string is played sequentually, rather than simultaniously. Even if you strum very fast they are still played sequentially, and the greater the gap between the strings and/or the slower you're strumming, the more obvious this is.
But there are other factors - angle of the pick for example - most guitarists do not pick precisely perpendicular to the guitar body, and often the angle of the pick is different on a downstroke as compared to an upstroke. This has to do with body mechanics (hand/wrist/forearm bones, muscles and joints) moreso than the instrument itself. Also, left handed versus right handed, but either way the tone would be noticably different.
Also, guitar players often place their picking hand centered over the strings they predominantly need to play for a given song, and while they move their picking hand a little they do stretch their fingers holding the pick to cover the rest of the distance - this is because moving your hand is slower than moving your fingers - basic body mechanics.
The angle of the pick, as well as it's stiffness and shape, impacts the sound the guitar produces.
Believe it or not, this is all just scratching the surface. There are so many human factors in how a guitar is played in addition to the instrument, amps, effects and other parameters. We could talk for pages on any of these things by itself.
But, the most important factor in my mind is the musician - if you were to buy one guitar and hand it to Steve Vai, then to Eddie Van Halen, then to Joe Satriani, then to Nuno Bettencourt and asked them to play the exact same song, you'll have four absolutely different and unique versions of the that song, even though the equipment is identical.
So different that many people could hear any of the four versions of that song and say "Hey, that Eddie" or "Cool, that's Steve" and so on. Even if they played the song as scored, note by note, as precisely as they could on the exact same equipment, you can tell them apart.
That's because they all play very differently - the way they hold the picks, the way they move their left and right hands, the way the press the strings and set up the next chord, and so on. For example, Steve Vai often creates vibrato with more "finger" than Eddie Van Halen who uses more "wrist". I'm sure they both can create vibrato with fingers or wrist or both, but they have their preferences. This is the human aspect that sequences and samplers cannot reproduce on their own.
Add to that how synthesizers work (or don't work) as compared to a real guitar and things quickly become more complicated when trying to mimic an acoustic instrument and a human playing that instrument. I have yet to hear an analog synthesizer that simulates guitar well at all. This leaves samplers, however the way a sampler works is to simply regurgite a recording or a portion of a recording when it's told to. If you send it a MIDI "A", it plays the MIDI "A" sample. If you send it another MIDI "A", it plays that same sample again.
You can somewhat change how that "A" sounds by adding expressive data - pitch bend, velocity, and in today's world many, many midi-controlled parameters but still, the root sample is still the same, though tweaked a bit on the fly.
Think about this: If you ask the world's greatest guitarist (whomever you deem that to be) to precisely pluck four measures of middle C 16th notes, you will have 64 different sounds. If the guitarist is good, they'll be very, very close but they
are different.
What I've devised (and this is not an original idea by any means) is to break each guitar note into multiple fragments:
The pluck:
A) the sound the guitar makes when the pick hits the string.
B) the sound the guitar makes when the pick leaves the string.
The body - the sound the guitar makes after the pick leaves the string.
The tail - the decaying volume part after the body sound.
Now that you have a library of each part of the guitar sound for each note on the guitar, record them all again. And again. And again.
This builds a library of a "note fragments" that are different, and you can sequence them so that a four measure run of 16th notes sound imperfect, closer to what a human player would provide if you asked him/her, even a really good human player.
Pluck A sample + Body A sample + Tail A sample will sound different than Pluck B + Body C + Tail A and so on. You get the idea I hope. What we're doing here is trying to make a sampler and a sequencer, two very precise devices, to appear less precise, and more human. Nudge the timing of the midi data, including note on/off and all the expressive data, and you're getting closer.
The proof is in the pudding however, so you tell me. Does this track sound like Sonar 5 Producer sequencing a Korg Triton Rack through a Fender Cybertwin amplifier, or does it sound like a human being playing a guitar?
Echo
If you voted "human" I thank you for the compliment but you get no pudding
Had I been blessed where I could be a really good guitar player like Paul Gilbert (Mr. Big) who wrote and recorded this song originally, I could have recorded this in one take in 5 minutes on one of my many guitars as compared to the 150+ hours I spent with a mouse.
There is a huge advantage of being able to play the instrument you are trying to mimic with MIDI, and that's why several folks have suggested taking guitar lessons or hiring someone with the talent - I think that's actually on point and quite valid because of how a guitar works and is played compared to a sequencer and a rack of midi stuff - but it is a ton of fun and a challenge to simulate if you have the patience and energy, or in my case, frequent bouts of insomnia.
But it
is very time consuming, and please don't expect to record a few samples and push a few synth keys, it's not going to sound like a guitar. It will instead sound like what it is - a sampler playing guitar sounds.
Humans, as annoying as many of them can be, are very difficult to simulate with technology.