Half Life 2 ending...........spoiler
Discussion
Furyous said:
What a let down ! Really enjoyed playing this the whole way through.Just finished it tonight.Cant believe how weak the ending was.
Is Episode whatever worth playing ?
F
Is Episode whatever worth playing ?
F
Must admit, I'm waiting for the reworked ending.
G-Man appears, starts his "Time" monolouge.
Breaks in to song and dance routine... "It's just a jump to the left..."
ThePassenger said:
Furyous said:
What a let down ! Really enjoyed playing this the whole way through.Just finished it tonight.Cant believe how weak the ending was.
Is Episode whatever worth playing ?
F
Is Episode whatever worth playing ?
F
Must admit, I'm waiting for the reworked ending.
G-Man appears, starts his "Time" monolouge.
Breaks in to song and dance routine... "It's just a jump to the left..."
Totally agree, worked my arse off, and compared the the end you get in HL1, it was a bit lame. Obviously HL1 was something that felt really new, and the end suggested HL2 was in the pipeline. I guess we all knew another HL was on the way so they could do little in that way, but the lack of imagination seemed like a poor return for the lost hours of my life I poured into the game 
wo was me 
I didn't think EP1 was all that special. It added an extra edge, but apart from acting as a vehicle in getting the player to EP2, I can't really see what point it served (well, apart from to make money for them!)... It didn't further the storyline.
BUT if you just want more of the same sort of thing, then it is good as will keep you entertain for a few hours (HL2 took me about 32 hours to finish, EP1 took me about 4 hours in one evening to finish
).
wo was me I didn't think EP1 was all that special. It added an extra edge, but apart from acting as a vehicle in getting the player to EP2, I can't really see what point it served (well, apart from to make money for them!)... It didn't further the storyline.
BUT if you just want more of the same sort of thing, then it is good as will keep you entertain for a few hours (HL2 took me about 32 hours to finish, EP1 took me about 4 hours in one evening to finish
).I played HL1, loved it, and replayed it many times... many years later played HL2, enjoyed it, but only played a few sections through again and then that was it.
I can remember more of HL1 now than HL2, and it's 8 years since it came out! Thats how much impact it had at the time...
HL2 did little new. Visuals and interactivity with the immediate environment jumped leaps and bounds, but the AI and constrained path feel were no further on from HL1... infact the sensory realism just made the lack of choice and freedom stand out even more
I want to have more choices on where to go and what to do. No longer should an indestructible piece of wood stop Gordon Freeman, when every other piece of wood will break with the crowbar!
Working out a puzzle to get past an electrified fence which takes me 30mins, when in theory I could put those boxes upto the wall, but a magical screen prevents me from doing it! It worked in HL1, it had to, but surely we have the ability to have some open planned gameplay now?
Dave
I can remember more of HL1 now than HL2, and it's 8 years since it came out! Thats how much impact it had at the time...
HL2 did little new. Visuals and interactivity with the immediate environment jumped leaps and bounds, but the AI and constrained path feel were no further on from HL1... infact the sensory realism just made the lack of choice and freedom stand out even more
I want to have more choices on where to go and what to do. No longer should an indestructible piece of wood stop Gordon Freeman, when every other piece of wood will break with the crowbar!
Working out a puzzle to get past an electrified fence which takes me 30mins, when in theory I could put those boxes upto the wall, but a magical screen prevents me from doing it! It worked in HL1, it had to, but surely we have the ability to have some open planned gameplay now?
Dave
Mr Whippy said:
HL2 did little new. Visuals and interactivity with the immediate environment jumped leaps and bounds, but the AI and constrained path feel were no further on from HL1... in fact the sensory realism just made the lack of choice and freedom stand out even more
This is one of the major problems with designing modern games. Building the actual 3D engine is relatively easy, even if you add on a superb physics engine and HDR engine on top... that's a piece of cake. Content sucks up vast quantities of time and money; it's also what sucks up the vast amount of space when installing a game.
Take HL2 for example, towards the end when your in the Citadel you get to see textures of City 17 and it's a big place! Even attempting to build half of that would have meant the game would still be in development, you'd need to model it, texture it, play test it to make sure everything works... and lets be honest if your walking down a boulevard of shops you don't want all of them to be the same and you'll want to go inside (same with buildings in general) so that's even MORE modelling work that needs to be done.
In some respects the procedural generation routines being shown off in Spore are the future of how games are going to be built; it has to be otherwise your looking at longer and longer lead times, higher costs of development (and thus higher RRP) and as far as us gamers are concerned it'll be sequel after sequel.
AI, is one of the trickiest things to program. It's a nightmare of code jumping all over the shop at random (XB360 and PS3 have problems with code doing this, but that's an argument for another time) you don't just need to program in basic senses (sound reception and directional identification, vision) but you also have to build in 'dumb'.
Seriously, if you make every Combine soldier a perfectly trained and well drilled marksmen with 20/20 vision then the player is going to get mown down in 5 seconds flat, you have to build in the fact that sometimes the AI doesn't hear the player coming or doesn't see Gordon dash in to a doorway, then add in 'fog of war' type scenarios, panic and the like and you can see how messy that code is going to be.
Even the best AI's I've come up against have a pattern that I can learn and adapt to, even the best adaptive routines that ramp up AI difficulty can't honestly copy with a player who dynamically changes tactics (i.e. flips between gung-ho and sneaky when the game isn't expecting or designed for that).
Again, the procedural code ideas might well improve on this and allow for some genuine tactics to be deployed by the computer opponents rather than the game simply reacting to the player.
Yep, it is a BIG challenge.
I've made bits and pieces of content from Quake1 through to HL1 and then moved to making cars for some sims along with tracks too.
Even within the main program I currently build for, over the last 4 years I have gone from building for 5k poly for the main car bodies and maybe another 5k here and there in interiors and wheels etc.
What is scary is that the latest gen PGR3 stuff is upto 40k polies!
What worries me even more is that these artists are made to build these models quickly and with a huge detail level, and I pick holes all over them! PGR1 was bloody terrible really for accuracy, though as an overall package it was damn nice
Dunno what I'm trying to say. Perhaps the best games will be those made by communities. They have relatively infinite resources, and the skill sets are getting very high too! rFactor was made as a very well rounded engine over delving into trying to provide content!
NFS Carbon for instance looks great, most of the NFS series have done, but they are wrapped in what I call a "dog turd" of a package. NFS HP2 and Most Wanted must have had artists flogged to get that level of work done, but then it was all packaged corporately to appeal to the lowest common denominator! How disheartening for the artists to see their work used in a package that doesn't do it justice Of all the games NFS4/5 were two of the best. Network, internet, hundreds of modes, tracks, gameplay types, damage, worthy AI etc etc etc, replay modes galore, SOOOOooo much stuff to do with the content which at the time was stunning!
But NFS HP2? No cockpits, no replay track cams, half the play modes, no real support for steering wheel inputs, WORSE AI? Helicopters dropping bombs? What a step backwards? Loosing features?
I feel the same is visible in most new games. The artists can push the boundaries but the people behind the other areas like storyline, AI, interactivity etc, simply cannot achieve enough to do it justice. Perhaps eye candy takes precedence over the rest of the bits, it sells so why not.
If it looks good and sells then thats all that matters. If it has about 2hrs gameplay time till your bored they don't care much any more. Gone are the days of small groups making a living. Now it's BIG business the games market... profits probably come before long term gamer satisfaction!
rFactor/WoW and other games like that seem to be the way to go. Test Drive unlimited is clever. An open world that people can all interact upon, and then just release new content as and when at a small price... seems to be working well for now
Dave
I've made bits and pieces of content from Quake1 through to HL1 and then moved to making cars for some sims along with tracks too.
Even within the main program I currently build for, over the last 4 years I have gone from building for 5k poly for the main car bodies and maybe another 5k here and there in interiors and wheels etc.
What is scary is that the latest gen PGR3 stuff is upto 40k polies!
What worries me even more is that these artists are made to build these models quickly and with a huge detail level, and I pick holes all over them! PGR1 was bloody terrible really for accuracy, though as an overall package it was damn nice
Dunno what I'm trying to say. Perhaps the best games will be those made by communities. They have relatively infinite resources, and the skill sets are getting very high too! rFactor was made as a very well rounded engine over delving into trying to provide content!
NFS Carbon for instance looks great, most of the NFS series have done, but they are wrapped in what I call a "dog turd" of a package. NFS HP2 and Most Wanted must have had artists flogged to get that level of work done, but then it was all packaged corporately to appeal to the lowest common denominator! How disheartening for the artists to see their work used in a package that doesn't do it justice
Of all the games NFS4/5 were two of the best. Network, internet, hundreds of modes, tracks, gameplay types, damage, worthy AI etc etc etc, replay modes galore, SOOOOooo much stuff to do with the content which at the time was stunning! But NFS HP2? No cockpits, no replay track cams, half the play modes, no real support for steering wheel inputs, WORSE AI? Helicopters dropping bombs? What a step backwards? Loosing features?
I feel the same is visible in most new games. The artists can push the boundaries but the people behind the other areas like storyline, AI, interactivity etc, simply cannot achieve enough to do it justice. Perhaps eye candy takes precedence over the rest of the bits, it sells so why not.
If it looks good and sells then thats all that matters. If it has about 2hrs gameplay time till your bored they don't care much any more. Gone are the days of small groups making a living. Now it's BIG business the games market... profits probably come before long term gamer satisfaction!
rFactor/WoW and other games like that seem to be the way to go. Test Drive unlimited is clever. An open world that people can all interact upon, and then just release new content as and when at a small price... seems to be working well for now
Dave
Valve are apprently reading some serious upgrades to the source engine - if you've got a multi-core CPU, the engine will start using the other idle core to start processing AI, to make it more 'reactive' - should make it more interesting!
But as a result, Episode 2 (and TF2 and portal ) have all been put back a few months. bah.
But as a result, Episode 2 (and TF2 and portal
) have all been put back a few months. bah.Tonsko said:
Valve are apprently reading some serious upgrades to the source engine - if you've got a multi-core CPU, the engine will start using the other idle core to start processing AI, to make it more 'reactive' - should make it more interesting!
But as a result, Episode 2 (and TF2 and portal ) have all been put back a few months. bah.
But as a result, Episode 2 (and TF2 and portal
) have all been put back a few months. bah.Valve's AI routines need the overhaul, the Combine soldiers in HL2 seem to be two modes. Random Blitzkreig or Ignore everything, zero tactics whatsoever.
Valve saying AI is updated is akin to Polyphony Digital saying GT4's AI was updated. Basically it wasn't but someone altered a few coefficients because that afternoon they thought they were better...
Call me skeptical but most AI is all caveat based. I remember a long while ago reading an article in Edge magazine about the AI for the Ferrari Challenge game for the PS2. Using survival of the fittest and generational "growth" of superior drivers... all really interesting.
I don't pretend to be great at AI or programming, but several friends work in and around this area of the industry, and I've had long discussions with them. It is damn hard to do it well, but it is possible. I honestly feel that too much emphasis is put on eyecandy for the screenshot/demo brigade.
I still have Tornado (10meg overall) from DID as my flight sim of choice. Old, functional graphics, limited sounds, but oh the immersion in the gameplay! I can see they spent 90% of the time thinking about the game and not the graphics!
Dave
Call me skeptical but most AI is all caveat based. I remember a long while ago reading an article in Edge magazine about the AI for the Ferrari Challenge game for the PS2. Using survival of the fittest and generational "growth" of superior drivers... all really interesting.
I don't pretend to be great at AI or programming, but several friends work in and around this area of the industry, and I've had long discussions with them. It is damn hard to do it well, but it is possible. I honestly feel that too much emphasis is put on eyecandy for the screenshot/demo brigade.
I still have Tornado (10meg overall) from DID as my flight sim of choice. Old, functional graphics, limited sounds, but oh the immersion in the gameplay! I can see they spent 90% of the time thinking about the game and not the graphics!
Dave
Mr Whippy said:
I don't pretend to be great at AI or programming, but several friends work in and around this area of the industry, and I've had long discussions with them. It is damn hard to do it well, but it is possible. I honestly feel that too much emphasis is put on eyecandy for the screenshot/demo brigade.
Yep. Graphics are the most important thing in the eyes of the publishing companies these days because that's what sells.
If you came out with a game based on the original Neverwinter engine (not the tweaked NWN 2 one) that had a brilliant plot, good looking characters, well done animation sequences and the like; every reviewer would slam it because it doesn't have graphics to rival *insert modern game* which means it's pap.
Personally I'd happily play something using the original Unreal Engine if it meant the enemies were acting intelligently, by that I also mean not like the 13yr old bunny hopping speed freaks I find on-line either
Gassing Station | Video Games | Top of Page | What's New | My Stuff