Posts by Defrag

Quote

what you're saying is if you're weak and succumb to the propaganda (i'm sorry but it is), then yes, you will be able to shoot people without any moral obligations.

I wouldn't really characterise it as 'propaganda'. It's not really an ideological thing -- it's simply teaching someone to be better at their job via responding to an input with an targeted response.

Quote

if you have absolutely no inclination whatsoever, you will be smart enough not to be put into battle in the first place.

If you're a soldier who is conscripted / drafted and you want to survive, it is in your best interests to soak up the training and give yourself the best chance of surviving. I agree that this sort of conditioning is quite scary, but war is fucked up in general. Even famous people and established sportsmen often have to do their service in countries that have mandatory service; it's not like everyone who doesn't want to fight can just duck out of it and they'll be excused. I would never want to join the army and potentially have to hurt or kill others, but if I was drafted and there was the possibility that I would be placed in situations where my life and the lives of my friends would be at risk, I would want to come out alive, even if it meant the enemy (faceless or otherwise) would not.

You are definitely correct regarding the potential for post-traumatic stress, guilt and various other mental issues that come after a tour of duty, but you have to survive the battlefield first and worry about that later. There's various things that can be done to manage it, but it's still an ongoing area of research. In typical human fashion, we are pretty damn good at conditioning to kill & destroy, but repairing the damage is almost like an afterthought.

I'm not trying to say that this sort of training is moral or immoral, just that is necessary if you want to be a 'good' soldier.

Quote from Thrik

No amount of mandatory hardening (which is also wrong IMO) can make you a killer if you have no such inclination, unlike those who sign up of their own accord.

I don't really agree, Thrik. Unless you're a serious peacenik who actively ignores / rails against all training offered to you, you will be trained to do your job -- the job being killing other people. Once you drill something enough times while aiming at human-shaped targets using real weapons in a stimulus-reaction scenario, the action (shooting at the target) becomes automatic. There is no "oh the humanity" moment prior to a well-trained soldier pulling the trigger.

Modern day soldiers are basically trained to run on autopilot as if everything is just one big target range -- you can't be a good soldier without it. If you read some accounts of WW2, you'll see why this is. Back then, it is estimated that only a fraction of soldiers actually tried to shoot at the enemy. Imagine how quickly you'd lose a modern war if most of your soldiers did that and the enemies are going all out to kill you...

Most armies have learned from the 'problems' of outdated/ineffective training methods and now have solved the 'problem'. Knowing that your enemy isn't a faceless machine doesn't come into it. You won't be thinking when you're pulling the trigger, you'll just be pulling the trigger and moving on to the next target. Kinda scary, but that's the reality of it.

My 2p, mostly just recycled out of advice given to me in the past.

• 
  [*:18aqogx3]Ditch the splash, or change it to incorporate a quick summary of your skills (i.e. a summary of key skills from your CV). I'd personally just ditch the splash and make the work page be the default.
  [*:18aqogx3]Make the image thumbnails bigger. People want to see what they're clicking on Also, the website itself is quite narrow. It only takes up about 1/3 of my screen width (1600x1200). Once you start adding more content, it may be cumbersome to navigate.
  [*:18aqogx3]Brand your images to include your name and website url. It doesn't have to be some obtrusive thing, just a small text layer in PS. You can actually create a layer with your info in PS, then create an action to go through all images in a dir and brand them automatically -- it saves a lot of time and effort doing this.
  [*:18aqogx3]You mention max & static props etc. in your CV, but there is (as far as I can see) nothing model-wise on your site. Modelling is quite specific so I definitely think you need to show some props if you put the skill in your CV
  [*:18aqogx3]Put your skills close to the start of your CV with the key skills listed first rather than positioning your skills at the foot of the page. I'd be tempted to remove specific level mentions from your CV and just say that you worked on several maps in your spare time -- they can see this stuff on your site. Keep the mod mention, though.
  [*:18aqogx3]Putting negative things on your page isn't such a good idea. An employer probably doesn't want to hear about abandoned maps regardless of whether you learned anything from it unless your page already has a lot of quality content (currently you have two maps on your site, and one of them is abandoned). Personally, I would not put an abandoned map on a portfolio site. If you need that map to fill it out, then it probably means you need to make some newer, cooler things. Just keep churning out the content.
  [*:18aqogx3]If you have a recent, non-mod map to pimp, it's probably a good idea to include a download link. I usually end up losing my old work, but fresh stuff should be available to DL if someone wants to take a look
  [*:18aqogx3]Register an easy to remember domain and some cheap hosting (I'd recommend 34sp.com -- they've been fantastic for me)

Anyway, hope this helps. Read this site for excellent advice; it really helped me. http://www.thejonjones.com/2005/10/07/y ... pels-jobs/

I followed most of these rules and, despite only applying to one company (got in a while back), I got emailed out of the blue by a recruiter the other week regarding various programming jobs (probably half spam, half serious, but still -- my name got out there despite doing no advertising/soliciting whatsoever), so following this guy's formula definitely is good.

Good luck!

Where's Chris Morris when you need him? A BrassEye video games special would've been epic.

This looks f to the uckin' awesome. I'll pester my buddies to get it on our server (if it's not there already).

Oh and JeanPaul, I seem to remember trying to get people to stop being overly-critical of something you posted (the big tubes) a while back, so why are you being so pedantic about 0.0001% of the map? Did Valve invent trusses and applying textures to brushes, too? It's a totally pointless gripe.

Played a version of this on a pub and it was fun. Next time I play it, I'll try and think of something to say other than "it is fun", though

Btw that interview was pretty interesting -- especially the part about generalised computing potentially sounding the death knell for the likes of DirectX. I've heard this a few times, but it seems to be becoming more and more of a possibility due to the rise of multiple core CPUs. I do agree that it's a shame that the PC market has become a minefield in terms of disparate target audiences-- it's currently high(ish) end and then everything else. You either make a game that is dripping in bells and whistles and exclude most PC owners from playing it (Crysis..), or you have to make compromises to guarantee people with slower PCs can run it. How much you can compromise depends on how much money you have to throw at it, and whether the design even allows certain things to be culled in the first place.

If we do go back to old school software renderers (OK, a bit more advanced ), that will solve a lot of problems in one stroke. If you take some of the cheaper CPUs on the market at present and then compare it to the high end, although the difference is large, it is absolutely nowhere near as large as when comparing CPU and graphics at the same time. I'm just glad stuff like physics cards never got a foothold; that would've been another thing to pin down the PC market and further fragment the situation. Anyway, if newer CPUs all start to accumulate more and more cores, it won't be long before everyone has some pretty decent grunt even in a shitty PC that was never bought to play games. If everyone is using much the same thing in terms of functionality (you can do most things on a cheap CPU versus an expensive CPU, it usually just comes down to speed), then it's a lot easier to manage.

If we go back to having no dedicated graphics acceleration, the PC (once again) becomes an almost inadvertent games platform. I remember back when I was in high school around the time that 3D accelerators were just arriving. I swapped my friend his PS1 for my PC games for a few weeks. His parents had bought a PC for office stuff, yet it could still play the games I lent him more than adequately. There is absolutely no way that would happen now -- it'd be some atrocious integrated graphics chip that died the moment you started the game. Back when I got my first PC, everyone could play games on their PC even with a worthless packard bell. Anyway, that's part of what the PC is missing right now -- the ability to just start playing games on any PC without having to wander through a hardware minefield.

If everyone has to start rolling their own rendering software (or paying licensing fees to companies who write their own) it could be ... interesting. I don't think DirectX and co will disappear for the foreseeable future, but standardised, free APIs seem to be something we just take for granted these days. If you have to write your own renderer or be left in the dust by Carmack & co. it could get interesting...

Looks good. The credit tab makes you look really childish though, regardless of whether there's any truth in it (which there is [the pipeline / tools being sub-par])

Quote from dux

Shame it's Dell

What's wrong with dell? The PC I use at work is a Dell and it's silent and a total beast I wouldn't buy one for my home PC because I like building my own, but their high-spec range is very nice.

Quote

One thing I'd love to see is an ability to automatically cap (or sew as it's refered to) any side of a displacement without requiring vertex manipulation to negate texture distortion. Let me know if you want some pics explaining what I mean.I'd also be happy to take this thing through it's paces if you want a beta tester, I love displacements

Yes, I'd like to see some pictures as that sounds like a useful feature

Regarding testing it -- as soon as I have a version that has subtle (as opposed to very obvious) bugs, I'll put it up for consumption

Quote

One important point would be, to be able to manually control the subdivisions horizontally and vertically.

Unfortunately I can't do this; Hammer's displacements are equally subdivided. I could probably allow you to split geometry to get the same(ish) effect, but I'm not sure you can get exactly the same shapes if you have to do this.

Congrats, guys. id software, eh? Not bad, Method. Not bad

Can you be more specific? If it's terrain based then it's probably not suitable for the editor I'm making.

I'm going to preface this post with an "it ain't done yet and won't be done for a while yet" message, so this excuses me when it comes to laziness.

To learn some C#, .NET & XNA a while back, I started working on a little Bezier patch-based displacement editor (think a simpler Q3/Doom3 patch editor). Its working title is the "Bezier Bongo Displacement Machine Extraordinaire".

The idea is simple -- you can build shapes using Bezier patches -- this lets you make really cool curved geometry (e.g a hallway arch that curves in two different ways) without resorting to the tediousness that is nudging potentially hundreds or thousands of vertices. Source obviously doesn't support Bezier patches, so the tool is going to tessellate the patches (already done and using Hammer displacement style triangulation) and spit out all of the vertices into a .vmf so you can load 'em in.

What I'd like to know is what you good folks would like to see in such an editor, both in terms of allowing easy manipulation of objects (groups of patches and individual patches) and sub-objects (patch control points). This could also be in terms of prefabs -- pipes, toruses, curved walls, floors etc. I've got plans for a prefab creator amongst other things, so if I get suggestions I should be able to add 'em. Please note that since I'm using Bezier curves, this application isn't really geared towards making anything but smooth geometry. Unfortunately it cannot support things like terrain generation. I could possibly have a bash at that in the future, but right now I'm focussing on curved geometry .

I plan to release the source once it's (relatively) finished, so other people can contribute and add their own features if needed, too. Given that I used it as a way to learn new languages and frameworks, it's probably a bit rough around the edges.

Current features

• [*:2os40wm9]Form-based. Has toolbars, resizable elements and all that good stuff.
  [*:2os40wm9]Patch creation. You can re-tessellate patches via subdivision just like hammer.
  [*:2os40wm9]Prefab patches. You can create arches, curved walls and a few other things. I plan to re-do the interface and put all of this stuff in a Hammer style "creation" form, complete with dynamic preview.
  [*:2os40wm9]Patch manipulation. You can select multiple patches, control points etc. and drag them around. You can also nudge using the arrow keys. You also see the control cage of the Bezier patch; this makes it easier to figure out which 'line' or section a point is in.
  [*:2os40wm9]Multiple views. Same as Hammer -- it has top, front, right and perspective camera views. Currently I can only show one at a time. I will eventually display more than one simultaneously.
  [*:2os40wm9]Orbiting camera option. This is a 3D package style mode where you focus the camera on the selection and then orbit around a selected object (bit buggy but hey).
  [*:2os40wm9]Customisable grid, including grid snap. I've got it set up to look almost exactly like hammer minus the highlighted additional grid lines every x units, but I can add them at some point

Stuff not yet in

• [*:2os40wm9]File saving and loading. I plan to allow saving and loading XML files so you can save your application settings & creations in the application's own format rather than just exporting to .vmf (so you could work on some geometry, save the bezier file and come back to it later).
  [*:2os40wm9]Exporting to .vmf (doesn't look like fun, in fact I'm dreading it )
  [*:2os40wm9]Textured mode. Texcoords are generated, but I'm just using wireframe at the moment.
  [*:2os40wm9]Undo support. Some sort of undo stack would be nice, but I have absolutely no idea how to go about creating this. Maybe I will do a bit of research soon as undo is very useful.
  [*:2os40wm9]Bug squashing. There's quite a few bugs and inconsistencies which have to be ironed out (e.g. in the perspective camera mode, nudging vertices is a bit dodgy). Moving things around is also a bit dodgy as the top left of the selection will move to the mouse cursor rather than dragging naturally. I will sort this soon I hope. The most annoying thing is that XNA is right handed which has confused my poor left-handed Direct3D addicted brain a lot. I'll have to eliminate all inconsistencies in my code to sort out winding order problems. Left handed forever! D:

Drawbacks: Since I'm using 3x3 patches (9 control points per patch) it is not possible to represent curves perfectly. What this means is that curves will 'bulge' very slightly. If you survived it in Q3/D3, then I'm sure it won't be an issue. The reason I went with 3x3 is that it is much easier to actually control the end result. 4x4 allows the creation of curves that are almost indistinguishable from the real thing, but they are trickier to edit (you have to use a formula to get the control point values correct!) I've set the application up to be flexible, so I may be able to allow these to be used in future if people demand it, though.

What it does mean is that if you make curved geometry in my editor, anything that must align with it perfectly (e.g. floor curve and the wall going around the curve) will probably have to be made using my editor, too (unless you like nudging vertices by 0.00001 units in hammer ...).

Here's a couple of screenshots of what I've got so far:

Screenshot one

Screenshot two

This is also a good read

http://www.3d-mike.com/game/lighting.htm

I bitched about Valve's tools all the way through making FF, but ultimately Valve is one of the coolest companies out there when it comes to support. Moreover, they are constantly upgrading the source engine so they have to support multiple versions which makes things more complicated. I spent about 5 hours a couple of weeks back trying to get shader compiling working, but I would fix one problem (such as the aforementioned sourcesdk/bin/ep1/bin problem) and then encounter another one. I'd then try and fix it myself despite having zero python experience, turn to the web for help and find nothing. Nothing in the wiki, either. It's incredibly frustrating, but you really just have to look at the complexities of supporting modding these days and cut them some slack IMO.

Also: Rule #1 of development: Everyone hates the tools they work with*

*May not strictly be true. However, when it comes to general usage, people are more inclined to find things to complain about rather than praising the things they take for granted.

I thought the characters looked good. They have .. erm.. character!

It's looking amazing, but did anyone notice a few framerate issues in the vids?

p.s. 'The old guy in the flat cap in the Steinway Beer Garden video looks like Fat Rab oot o' Consolevania / VideoGaiden

Whoever wins, the consumer gets screwed.

Going by the amount of their games I've played in the last ~8 years, Valve.

I think it's quite funny / humorous rather than malicious. The likes of myg0t just invade servers, play annoying shit over voice and try to be as obnoxious as possible. These guys seem to grief with a wink and a smile