dwringer

Hello all; I've been an enthusiast of BIS since OFP:Elite on my XBOX, and recently just picked up TO:H to add to my collection. When I bought it I went through Steam, so I should have the latest version/latest patch (everything says it's updated). One thing I noticed in the latest patch notes is a fix for an intermittently disappearing vehicle interior. I just wanted to report that I have been experiencing this bug while using the latest patch. The flickering is very brief, and only happens every few minutes - I guess based on what's being drawn at the moment. FWIW I'm on a Geforce GTX 680 with 301.42 drivers, on Win 7 Pro. *EDIT: Well, as a follow up: IT would appear that the Steam version of TO:H is still 1.05. That's kind of irritating since the patch has been out for days; the website discourages you from trying to install the patch ONTO the steam version, and the literature all claims the Steam version will update automatically. Guess I'll just keep holding my breath for them to update it; would be nice if the Beta Patch support were continued until the FULL Release of a new patch so I wouldn't be stuck with an old version right now.

***UPDATED: 2/12/2011*** 'mode' variable now takes up to three positions to the right of the decimal place. Script is MUCH more predictable and configurable; zones can be initialized to guarantee the same building placement as far as has been tested. Since I was always confounded by the unintuitive process of setting up and debugging a Town Construction Set module every time I wanted to make a certain kind of random town, I figured some of the process could be automated within reasonable constraints to generate reasonably appropriate towns for a variety of situations. Therefore I came up with, and present to you all, a very rudimentary script to set the initialization variables for Town Construction Set modules with only a single, higher-level line of code within the editor. To do this, make a file called "tginit.sqf" and paste the following code into it: ////////////////////////////////////// /// Town Generator Initializer /// /// Author: Darren Ringer /// ////////////////////////////////////////////////////////////////// // // // Instructions: place this file // // into your missions folder. // // Place a Town Construction set // // module on the map, and in its // // initialization field, put: // // // // nul = [this,<mode><.xyz>,<radius>] execVM "tginit.sqf" // // // // Where <mode> is determined as follows: // // 0 = sparse buildings // // 1 = loose suburban arrangement // // 2 = village-like arrangement // // 3 = dense, city-like arrangement // // This integer can optionally be followed by up to three // // decimal places (x, y, and z above), as follows: // // x :: if nonzero, scales the number of buildings, so // // mode 0.5 would make half as many buildings as // // would mode 0. // // y :: if nonzero, specifies a proportion of the number // // of buildings to be randomized, so mode 2.03 would // // generate 70 to 100% as many buildings as mode 2. // // z :: specifies proportion of random deviation from 1.5 // // (the default) as a multiplier to make trash and // // misc objects based on the total number of // // buildings created. // // and <radius> is simply the maximum distance from // // your Town Construction Set module at which you want // // buildings to be created. // // // // NOTE: Certain modes will not work below a certain radius // // size. In these cases, the mode will automatically // // get set to a compatible (denser) mode. No mode has // // been tested to be effective below a 50m radius. // // // ////////////////////////////////////////////////////////////////// // For example, // nul = [this, 1.735, 350] execVM "tginit.sqf" // in the init field of a Town Construction Modle // will create a town of radius 350 with mode 1 (scattered // buildings), 70% as many buildings as would normally be // in a mode 1 r 350 zone, 30% of which may or may not // exist, and 0.75-2.25x as many misc objects as // buildings. _tcs = _this select 0; _mode = _this select 1; _radius = _this select 2; _size = 2 * _radius; _miscrand = ((100 * _mode) % 1); _mode = _mode - (_miscrand * 0.01); _houserand = ((10 * _mode) % 1); _mode = _mode - (_houserand * 0.1); _scalar = (_mode % 1); _mode = _mode - _scalar; if (_radius < 100) then { if (_mode < 3) then { _mode = 3; }; }; if (_radius < 200) then { if (_mode < 2) then { _mode = 2; }; }; if (_radius < 400) then { if (_mode < 1) then { _mode = 1; }; }; switch (_mode) do { case 0: { tgdivisor = 90000; tgspac = 22; }; case 1: { tgdivisor = 46181; tgspac = 7.6; }; case 2: { tgdivisor = 22500; tgspac = 4; }; case 3: { tgdivisor = 2400; tgspac = 1.3; }; }; sCoeff = _scalar; if (_scalar == 0) then {sCoeff = 1;}; _count = (ceil (sCoeff * ((3.14159 * (_radius * _radius)) / tgdivisor))); _vals = [ _size, (ceil ((_count - (_count * _houserand)) + (random (_count * _houserand)))), 28, tgspac, (1.5 - (1.5 * _miscrand) + (3 * _miscrand)) ]; _vars = [ "townSize", "houseCount", "houseRotation", "houseSpacing", "coefMisc" ]; {_tcs setVariable [_x, (_vals select _forEachIndex)]} forEach _vars; tgdivisor = nil; tgspac = nil; sCoeff = nil; Then in the init of each town generator you want to set up with the script, just put: [this, <mode>, <radius>] execVM "tginit.sqf" Where <mode> is an int [0-3] defining the density of the buildings (3 being the most dense), and <radius> defines the radius (NOT the "size") of the town. The module's internal variable appears to operate as a diameter and not a radius as indicated by the wiki. My script lets you think in terms of radii. :p ------------------------------- Feature added 2/12/2011: Instead of setting mode to a single integer like 2, you can now do something like: [this, 2.905, 500] execVM "tginit.sqf" This gets a little tricky, so let me break it down. If mode is represented as A.XYZ, A :: [0-3] (general zone density, 3 being the densest of the four options) X :: [0-9] (if zero, does nothing. if 1-9, scales the number of buildings 10% to 90%) Y :: [0-9] (determines what percentage of the number of buildings in zone may or may not be generated. A 4 here makes between 60% and 100% as many buildings as a 0, for instance. this precludes the zone being generated the same every time.) Z :: [0-9] (by default, misc objects are created as a quantity 1.5x that of the number of buildings. this num determines what percentage above or below 1.5x will be randomized in the same way houses are randomized by variable Y. a 5 here would use a random coefficient: from 0.75 to 2.22x the number of buildings, for instance.) Thus, the example above will make a 500 radius zone with mode 2 density, 90% the typical number of buildings in such a zone, zero randomness as far as the number of buildlings, and 50% random deviance from the 1.5x multiplier to create misc objects. The primary motivation for this change was so that zones will consistently load once you have them set the way you want them (assuming you use a 0 for the tenths place), and if you can't get a zone to load (too many rocks or mountains or something) you can just scale back the number of buildings in it until it WILL load. ------------------------------- Suggestions, comments, flames, etc are more than welcome. I understand that a lot of this could be cleaned up, improved, etc. It is very much a work in progress. But I have found it very useful already, and thought it might be of interest to others. I would like to better understand the spacing variable and its effect, as well as potentially find a more elegant mathematical solution to scaling the different modes. The values right now were obtained through nothing but trial and error, other than the slightly less rational number 46181 that I got by working from a desired number of buildings in a specific size zone (75 in a zone of radius 1050, don't ask, haha). The randomization elements cause the building placement to be slightly random, if you remove all the randomness you can consistently generate the same zone over and over. That could be handy if you want to populate some of the buildings with some exhausting use of the setpos command. One thing to keep in mind here is that you can use some very large radii. Mode 0 on a 2400m radius circle sprinkles 200 buildings over a 7 square mile area. That's great for diversifying otherwise empty landscapes, especially maps that don't have many, or any, villages on them. You can see below (hopefully, despite the terrible image quality thanks to MS Paint), an example of how this can be used. The big zone on the left is density 0; the medium red one is density 2, and the small green ones are density 3. Layering zones like this can cause errors, so you have to be careful with placement. (The colored zones are meaningless markers added for demonstration purposes only)

I'm afraid I'm out of the scripting routine too much to be of any help at the moment with this, but an idea I think could make this script very useful would be as follows. Don't allow the player to jump more than once every few minutes with that kind of power, or include some kind of time-goverened scaling factor. Perhaps if the player jumps more than once in a short time there is a rapidly diminishing effect. One could also base the size of the jump on what kind of equipment the player was carrying. Finally there could be natural variances, either a randomization factor or simply a personal decision not to allow high jump strength factors on most units (except the most elite of the elite ;p). Then, all our roleplaying-addon friends could even make it a skill where repeated use gradually increases your maximum jump strength. In essence, I agree with all of you who say this is ridiculous, but at the same time I agree with the original intent of adding this kind of functionality. I can't be the only one who finds it absurd to have died several times simply due to the player character not being able to climb a 4 foot obstacle.

Unfortunately you have to shell out the cash for a wireless USB receiver for the 360 controller to get it to work. Probably 25-50 USD. They're pretty nice to have though, in the end. If you hook your controller up with a usb cable, it will charge, but not interface with the PC (at least, there's no stock driver support. i heard a rumor of some 3rd party way of doing it, but I'm not sure if it's legit). And nope, my 360 controller doesn't work with TOH either. And it is a shame.

if you change it to an sqf file, i think that will work, the "exit" shouldn't be needed however. Also, you will have to call it with "execVM" rather than "exec" if you use sqf, because sqf gets compiled to bytecode that runs in a thread on the virtual machine (i think)

Seems like to me this IS reinventing the wheel, in a way. If a compiler can be made for C# into sqf, and you say the primary reasons are .net metaprogramming tools and the vs IDE(I presume?), why not instead work on an SQF ide and new sqf extensions, with their own compiler to turn it into regular sqf. Seems there'd be a lot less work, and a lot more results.

I am sure using some of the engine's UI scripting capabilities, you might be able to simulate some kind of notebook screen, but it wouldn't appear on the computer. It would simply be a pop up window (similar to the gps device or any of the various addon consoles out there)

After reflecting on everything, I am forced to come to this conclusion as well. The outerra dev blog (or perhaps another page I read earlier that linked to it, too lazy to look it up now) mentioned being contacted by a prominent milsim developer for the use of their technology. Hmm. Given the outerra devs are, near as I can tell, Czech, that might indicate your post is not as far out as it might seem.

I'm still not entirely convinced I've got the relationships between the people fully resolved, is there a theme perhaps with 'resting places'? Seems evident on the wiki from a few places on the list. I'm not familiar enough with all the subjects in question to say one way or the other at this point. EDIT: I think It's probably going too far into it though at this point. I think we might need to look at the evidence about Malden, Miller, the Astute, Astute:statue anagram, etc, to finally be able to put some pieces together. I don't know if there would or wouldn't be another clue after all this to help us put everything together, but the persian 'goodbye' might indicate that there will NOT be. EDIT 2: So, here's what I'm thinking currently. The events of OFP did not play out as they should have. What happened there led to something happening later in that region involving China, Greece, Persia, and/or Turkey that was not supposed to happen, and the balance of power has dangerously shifted. SO the idea is that miller was supposed to die at malden, it was supposed to have been nuked. That ties him in with the other name/date/place sets from today's hack. He would have died there, and the Soviets winning would have been temporary but prevented something much worse? EDIT 3: Err, whoops. OFP is set in 1985, so it can't be. Unless he was the child of someone from OFP.

There does seem to be a clear indication of specific events in which one party suffers or is defeated, but ends up winning a larger struggle. Beyond that I agree my above post offers tenuous connections in places.

Gamma may not affect anything, but the ingame Brightness setting affects the calculation of HDR so if you overcompensate with it for cutting the gamma TOO much you will have your whites wash out far too easily (and other problematic things). Should brightness and/or gamma affect anything other than a final whole-image calculation, AFTER post processing? no, IMO. Is that the case? Not by a mile.

There is a clear ordering intended for the phrases, I have put together the following: 2003 BELARUS DUFFY BIELSKI 2002 NOGOVA BOHEMIA TROSKA 1991 NOTTINGHAMSHIRE REYNOLDS HOOD 1962 OTTOMAN LEAN LAWRENCE 1958 CLARA CASILLAS GUEVARRA 1943 NERETVA LOHR TITO 1814 MADRID GOYA SPANIARD 1462 TARGOVISTA MEHMED VLAD Notice any patterns emerging? And I assume MALDEN and MILLER are two more words for us. EDIT: MALDEN REYNOLDS MILLER ? No idea, just grasping at straws. Haven't really been following all the bits and pieces :P EDIT 2 (since nobody else is saying anything, lol) It appears to be a relationship where in one location (first term), an entity (third term) "loses" to another power or entity, or suffers as a direct consequence of action by said entity (the second term). But in losing, is actually "winning" a larger struggle. For instance, the Lawrence of Arabia instance specifically (i think) refers to when he is tortured (and then released) by the Ottomans. Since that is really the only time in the movie he is IN the ottoman empire that I can recall. Or Vlad Tepes, being driven out of power at Targoviste, who also participated in the end of Ottoman aspirations in the region (under Mehmed). No, I have no idea about what any of that means. I just wiki'd it for the purposes of this thread. ;p

I skipped several pages when it looked like people were still fixating on the ftp. Was the FTP really enabled just one or two days ago by someone entering a secret php filename into the p8h8 server, thus disabling the Programs/ directory and activating the ftp server? Because I have seen absolutely NO evidence that this is true. I thought the FTP server was there days ago. And for all we know, that FTP server is for BIS devs to upload/change things, not for us to hack into.

Man, I KNEW i recognized the name Capt. Scott Miller from somewhere. I can't believe it took this long for somebody to find it and point it out. Many kudos, BigSlongsDaddy. (Though it doesn't fit in this context whatsoever, it would seem. Might just be a coincidence... who knows?

Analysis of CIA Cryptonyms from "Phase 5" Map So I had not realized until today that MKSEARCH was, in fact, the new name of the MKULTRA program. With that in mind, it seems like a fair bet that this is a page from a CIA intelligence estimate. Judging by the top of the page, it is regarding an operation codenamed "SDREAM", presumably an operation to do something unpleasant to the Iranians (if SD is the diagraph for Iran, and 'ream' is taken to have metaphorical significance). Finally, the bottom indicates the source of the intelligence: SDDUSTER (perhaps the CIA codename for Iran's own operation in the region), via LCIGNITE. Presuming LC is the diagraph representing China, this would perhaps indicate a Chinese captive was interrogated under the MKSEARCH program. I am making a bit of an assumption with the last part. But the addendum saying "ref. to p. 215-248; reliability disputable - SDREAM #000019" is beyond me. At first I thought perhaps it implied that the intel was gathered through some form of torture, rendering it unreliable. But this is not something I have any faith in whatsoever.

From the OFP: CWC article on wikipedia, the latest edit (just today!): The edit added a bullet point to the Reception heading (citing reviews and such): But, that IP looks to be a T-Mobile account from the UK. So this might be a complete non sequitur. Interestingly, it was done precisely 1 minute after Neopas' comment directly above, if my timezone estimations are correct.

I had dismissed it as my overactive imagination, or a desire to see something that's not there, but now that you guys mention malden, I have to say, I looked up some Arma 2 billboards to see if the ascii art from the animation resembled any of them, and the only thing I located with similar composition (if you kinda squint and use your imagination as to the color conversion algorithm they must have used to make it into text) was the billboard displayed at 0:16 on the page Which also makes me think that it would mean Greece ends up invading Malden. The reference to 1821 is to the Greek War of Independence, seizing their opportunity when the superpowers of the time were distracted.

Just saw the line 'real OO support, not bolted on' and couldn't help but laugh. The VAST majority of object oriented code out there in the business world has 'bolted on' OO support. The entire world doesn't revolve around Java. SQF has remarkable capabilities, it is just very specific to the Arma 2 engine so requires certain allowances by the programmer to constrain logic within the performance limitations of that framework.

Regardless of what it is, I'm guessing the ASCII art implies we are getting 'properly' destructible buildings. (more like Bad Company 2 for instance, but presumably with more accurate physics and materials modeling)

Amazing find on the nvidia drivers, if it turns out to be anything. That said, I put together some of the frames from http://dl.dropbox.com/u/9367994/hacker_mails.txt into somewhat of a proper order, although I got completely bored after the first five or so. I added an intermediate frame that I personally received as well. So if anyone is curious, here they are. It just shows the sphere flying in from off screen to the left and then breaking through the wall. I stopped before the wall started to fly apart because that would require some actual attention to do. http://goo.gl/pfU8v

I just wanted to say that in my opinion, the change to AI movement behavior when commanded by a player is not a fix to the issue it purports to resolve, but rather a workaround which is detrimental to the current gameplay. If i have a few of my squad members assigned to team red, and tell team red to move somewhere in "DANGER" mode, I want them to use all the covering they bloody well can to take care of the movement without any micromanagement from me. This worked extremely well. If I were to do the same now, they would simply run out into the open and get shot up without anybody doing a thing about it if I understand the patch notes correctly. I do not like this at all.

I am pretty sure I read an article somewhere that said the SVD was designed to give infantry squads an effective range of small arms fire of 600 meters, and trying to hit anybody at that distance with anything less than 7.62x54 is a crapshoot at best.

For example you could make a trigger. The activation condition could just be "Anybody Present" I guess. In the "Condition" field below, put (((boundingBox yourUnitName select 1) select 2) < 1) Then the trigger only fires if the unit is prone. It will not work with crouches though, and I guess there is no way to get it to (?) You could make it "(this AND (((boundingBox yourUnitName select 1) select 2) < 1))" if you have a trigger Activation condition that must also be met (not just "anybody present" with no regard to where)

I have seen the Chain of Command pass commands through AI leaders before, through quite a bit of tricky synchronization (well not tricky if you know the way it's supposed to be done, I suppose, but since the AI is so weird it's hard to tell IMO). Unfortunately the subordinate groups tend not to give their units the order that the leader received. Thus the group leaders all follow the orders they are given (pretty much clumping together wherever the commander of which they are subordinates was ordered to go). This is mostly useless, but the fact that the units in their groups simply stand around makes it completely useless. I guess in COOP missions this would still be ok. Or has anybody figured a way of getting AI to work such that a human high commander can usefully give orders to subordinates in a structure similar to that 2 posts above? :P

If you don't mind my asking, how? I have experimented a little with the config_events file (or whatever it's called) and event handlers in the past, but never got anything like that working. What method did you use with DAC to do that?