I had a long discussion with a friend about this, and we ultimately agreed that the 100 player idea is broken. It will be an enormous headache, or alternatively damage the game play to a significant enough degree to put off many players. But I'll do something that the other QQ'ers on the Star Citizen forums have not done... and that is ... explain why. (*gasp*) Let's assume the instance is limited to 100 players. A new one is dynamically created when the 101'st player tries to enter the same region, hence the instances 'stack' over top of each other where there are 100+ users. Where they are less, it is a single instance. This initially appears to make sense, but you soon run into some very odd and complex issues relating to balancing. A) What is 100 players? Is it 100 Constellations manned by 4 people each? So it's really 400? Or is it 25 Constellations manned by 4 players each, making it 100? This is a huge deal when it comes to large fights. B) How do you intend to ever have proper space battles? If your fleet of 100 friends is heading towards another fleet of 100 enemies, and you have your friends as your POI's (at least the 'Fleet Commander' would) - then you would never meet in battle. You'd be split into two paralleled instances and never fight. C) As a branching off of 'B' - what if the instancing system is faulty and only 10 friends get dropped into a fleet of 90 enemies? How is that remotely balanced, when the intent was for 100 v 100? D) The game would appear to limit the fights to 50 v 50. But how does the instancing system make sure the teams are not lobsided, it doesn't seem fair to split your team along arbitrary lines if you each have defined roles in your fleet. E) What happens if your friend gets blown up, and some unlucky freighter now gets popped into the instance? You have no back-up because your support will jump/warp into a new empty instance. Once again, how can you possibly hope to find a balanced match? The 100 player limit is far too low. F) People seem to take this 100 player limit as something positive... well, Chris had to put a positive spin on it. But having been developing with Cryengine 3- what you are likely seeing is Chris hitting a fundamental, underlying, engine limitation rather than a 'feature'. Nobody in their right mind, who claims to want to 'design their dream' would ever arbitrarily cut down the notion of 'open world' without an underlying limitation. E) The game is slated for Nov 2014... we will experience at least 1.5 - 2 generations of GPU and CPU during that time (Haswell -> Broadwell for example). We will have shifted to DDR4, and SSD's will likely be far more common. You cannot apply today's processing requirements to a game slated for release in 2 years, on an engine notorious for being beautiful but resource intensive. F) EVE Online is single-shard. Can handle 45,000+ players on the shard, and can support battles of up to 1000 players, tracking individual ships, projectiles, and now missiles (with their trajectories) live. Time dilation doesn't kick in until you go North of ~1100+. CCP games is by no means a big company, they run on a tight budget, have had to deal with layoffs and financial difficulties in the recent past, yet they manage to run a server farm that supports all of that, and an amazingly detailed world. Once again- Chris is hiding something more fundamental to the engine. G) While the game is high-poly (9 mill poly for the carrier). You will never have that amount of polys in your field of view. Look at the way the cockpits are designed. They are deliberately big, with small windows. Modern day GPU's now have 2GB frame buffers on average, and I don't see why that wouldn't continue to increase as we head towards 2014. Caching the textures into the GPU shouldn't be as big of a deal as they make it out to be. H) I don't buy the network issue argument. FireFall (an FPSMMO) is surprisingly small for a game of its type. Why? Because the game servers *stream* the textures live as you move around the world, consuming up to 7-8Mbit/sec of bandwidth (you can adjust this setting in your options from 0.5 - 9 iirc). Chris mentions the difficulties in computing the trajectories and physics of each ship as it moves around the world, citing games like Battlefield 3 as an example of a very intense 64 player environment. But the dedicated servers are not particularly stressed by this (having run several). So do I see an issue with the 100 player cap? Yes. Very much. It will be a major hindrance in core systems (PvE/trade centric) and will negatively impact any notion of a 'fleet' fight. You have to keep in mind that Chris constantly cites the poly counts as being one of the core issues: ... "Even if you had enough internet bandwidth to handle the data going back and forth and a super computer for the server there’s no PC, even with quad SLI that could render that many spaceships with Star Citizen’s fidelity." So how does he intend to counter 9 mill poly capital sized ships in a single 100 player instance? Each capital ship will have 2-5x the poly count of a fighter. And be manned by several players (or their NPC counterparts). Will he then introduce a capital ship limit, making the game even more restrictive? And how does he want to involve NPC ships in this mess when we will have problems with just the players alone? The issue is far more fundamental then most players realize. As I see it now; this is not an open world by any means.