Miuramir

First, you're more likely to get help if you don't deliberately set your words into a difficult-to-read color.

Please post your CoH Helper output so we can get a better idea of how your system is set up. Based on a quick search, your laptop may be using "Integrated Radeon Xpress 1200 Graphics", which is something I'm not really familiar with, but hopefully someone else is. In general, laptop integrated graphics are on the weak side for CoH, and manufacturer-provided drivers are frequently old.

Something you can try in the meantime is to start CoH in "safe mode", which sets most of the graphics settings to minimum. If things are correct (although low-resolution), you can try gradually increasing settings to see what your laptop can handle.

Two possible major categories I don't see mentioned yet:

* Full End refill from Domination. Back when Dom gave a damage bonus, people had (pricey!) builds with "perma-Dom" fairly early on, which implies enough recharge to fire Domination more often than every 90 seconds; and there were characters with double-stacked Dom, which unless I'm mistaken means <45 sec recharge.

* Confuse. This is sort of a special case of pet endurance instead of your own, with the additional bonus that there's a continuous supply of End with every new foe. The non-linearity of the damage to XP works in your favor if you juggle it right; the oft-quoted "do at least half the damage, get at least 80% of the exp" point may not be the optimum, but points out the possibilities.

Plant Control gets Seeds of Confusion at level 8; you could have Seeds, Flares, Incenerate, and Fire Breath exemped all the way down to level 3 under the current system, which has some weird potential given the ultra-low-level concept listed previously. Picking Natural or Mutation and having someone mail you damage Yin-Os would allow you to be pretty alarming even at real levels not much over that.

Anecdotally, my Plant Control / Fiery Assault Dominator did not need Stamina before the Dominator changes, and I was not particularly building for it and only to the mid-20s. With the Seeds of Confusion a core part of her attack style, the addition of Consume and planned for eventually picking up Power Sink (and possibly Surge of Power, using Domination to counter the crash), and having a variety of AoEs and pets (Carrion Creepers, Fly Trap, possibly Mu Guardian), I'm pretty sure End wasn't the limiter. Oh, and Fiery Embrace may be helping as well. Post-Dom update, I did respec her to have the Fitness pool, as her attack chain sped up; but IIRC post-update-post-respec I can't recall her running out of End even with Hasten running.

The interesting question is whether with a targeted build the enemy-killing rate could be cranked up on a uber-pricy build variant of the above to be noticeably above the limits you start with here; I'm not someone who plays at those sorts of rarefied levels enough to have a good feel for it, but I suspect that in a sufficiently idealized setting it could be done.

Side notes and musing: if she "went blue" and took Fire Mastery instead, would she get Consume twice? Better or worse than Power Sink? Or go Soul and get Dark Consumption, Soul Drain, possibly Fortunata? She got sidelined before epics for largely social reasons (i.e. unrelated to build), and I've not looked at her in a while. Certainly plenty of End-management tools available, to the point that I'm pretty sure that in the upper levels the limit would be on the damage output per time rather than being End-limited. (Over half of your chart's total time to 50 is spent post-41, with potentially the first epic power picked.)

My take is closer to "menu work is a necessary evil that is better attacked sooner than later". If Option 3 would be "Half the improved textures originally planned, but everything better organized for future expansion", I'd pick it.

As a side note... I seem to recall Castle saying that when he went and designed the new NPCs (Girlfriend from Hell, etc.) that the dev-version creator with the NPC parts was far worse organized than what we see as players, or at least burdened down by far longer lists. Wouldn't it be a long-term gain in overall productivity to improve the costume design organization for everyone? Or, put another way, if the reorganization is going to be needed eventually, investing in it now will pay off more dividends than letting things go until it's creaking further under the load.

I'm not sure it's quite that simple for older cards.

Quick thought experiment (WARNING: COMPLETELY MADE-UP NUMBERS): Let's say that the default texture used 4 bytes per pixel (R, G, B, Alpha), used to be 64x64, and was upgraded to be 256x256. (This is probably about the right magnitude of change to go from the worst of the old jaggy textures to something with much cleaner curves and diagonals, I suspect. I'm not sure the visual improvement by a 2x per axis improvement would be distinctive enough to be worth the effort in doing it, and 8x per axis sounds pretty pricey in terms of system budget.)

The hypothetical old texture would then have 4,096 pixels (4k) * 4 bytes/pixel = 16k for the texture. To keep things in easy 2^n numbers, let's say a typical outfit had 32 different textures; that's 512k per costume. Let's further say there were 64 people standing around Atlas for a costume contest; if you logged in next to Ms. Liberty, that's potentially 32M of costume textures you have to load up. (Hopefully, cache re-use keeps total memory use lower than that, but the card may have to process that much in any case.) For any card capable of running CoH even moderately well, that should be pretty easy.

If we go to hypothetical 256x256 textures, they are 16x as large throughout... 256k per, 8M per costume, 512M for the costume contest as a whole. Given that you need a bunch of textures for the scenery, plus the actual poly data, all of a sudden a lower-end card needs to do a whole lot of memory moving, and you might start having to waste a lot of time paging textures in and out of the card's memory. In the worst cases, on a <2G system memory computer, you might force CoH as a whole to do more disk paging, and that *dramatically* slows down everything.

Now, in reality I seriously hope that texture compression and texture library optimization / reuse means that the actual numbers are much more friendly; on the other hand, the textures might well be more high-res than the above and/or have a lot more bytes of housekeeping data embedded to handle things like reflectivity and such. Only someone intimately familiar with the way CoH handles texture memory and paging can really answer those sorts of questions; and there are some indications that the dev team no longer actually tests performance in any detail (if at all) on configurations similar to the published minimum specs.

tl;dr: Until we see a detailed analysis or some assurances they've tested things on low-end setups, it's possible that this change will move CoH yet another step away from the "minimum" listed setup actually able to be useful. Of particular concern would be anyone using low-RAM cards on old buses (PCI) in a low-system memory computer, as cascade slowdown might cause noticeable performance loss if it causes CoH to page to disk more heavily.

A factor that applies in reality, but the devs may or may not have accounted for, is that high-power reactors of almost any sort require considerable volumes of cooling water to turn that power into usable energy. Such a plant would almost certainly be located on the ocean, a major river, or the Great Lakes. Given the additional hassles involved with dealing with salt water, fresh is preferred; and something less vulnerable to seasonal variations would tend to favor the Great Lakes.

Given the existing shape of the US high-power distribution network, Chicago would actually be a pretty reasonable choice, although Detroit might be slightly better.

Also not conclusive by any means, but the design of Praetoria is heavy on the water features; it seems unlikely that this sort of design would have been created someplace without a major year-round water source.

There are two fundamental problems here, and an additional side issue:

* The CoH team has close to a 100% failure rate at delivering on the above promises, and in every such case I can recall players have been basically told to suck it up.

The arrogance level of the art team is viewed as pretty high for a reason, with cryptic pronouncements at widely-spaced intervals rather than communication, inattentiveness to player concerns, a lack of a meaningful feedback process (especially early enough in the process to actually serve any purpose), an unreasonably high error rate (what's the odds that a patch will break a previously-working costume piece?), extreme slowness to fix errors (if a database typo causes something to go missing and it takes more than one patch cycle to fix, your process has serious flow issues), and so on.

You appear to be working to fix a lot of this, and I have hopes... but we've had plenty of people come in with nice-sounding comments who fail to deliver results. Forum talk is cheap, results speak volumes. You are going to have to show us on this one.

* The trend lately seems to be toward more elaborately designed pieces to fit a specific look or set, rather than more useful pieces that can be used for many different things.

If I can draw an analogy, it's conceptually similar to a trend in Lego sets some years ago, where the number of custom oddly-shaped plastic pieces increased dramatically and the number of general-purpose bricks declined sharply. This made building the set pictured on the box easier and faster, but reduced the utility of the set for building anything else.

Personally, I'm a bit conflicted; some of the newer pieces are much crisper looking and that's usually good; however, I find myself frustrated by the "added elements" not matching what I'm trying to go for, or each other, far more often than they help. I can't even begin to count the number of times I've wasted time swearing at the costume creator because some artist decided that one piece "needed" some sort of overlay or blending layer, which changes the color so that it doesn't match some other piece that doesn't have such a layer, and the limited color picker doesn't allow me to match them up. Similar are the pieces that seem to have some element of "Caucasian" peach-ish skintone baked into them, and look really off when used on more ethnically diverse humans let alone the wide variety of colors seen in CoH.

Then there's things like the pieces that have decor elements that limit them to a very narrow range of uses, where the basic piece would be far more widely applicable. To pick just one example I ran into recently, there's a perfectly good bicorne hat model that if done properly could have been used for all sorts of historical concepts; Napoleonics, formal diplomatic dress, naval officers, and of course pirates both realistic and silly. It might even get used for unusual concepts like a hammerhead shark headpiece. However, it's got a giant white skull and crossbones plastered across the front, totally non-removable and non-colorizeable , which limits it to a small subset of silly cartoon pirates with particular color schemes and limited imagination (and the even smaller set of people making deliberate ironic use of the above).

Plus, the trend so far has been to introduce even fewer versions of the higher-resolution pieces. For instance, I have a character that has the "Headband with wings" piece as an essential part of her look. While she enthusiastically adopted the body textures from the Mac Pack Valkyrie set when that came out, the great-looking high-res head wings that came with it don't have a "with headband" option, and due to the somewhat bizarre legacy organization of the costume pieces, can't be combined with the existing headbands; so she's stuck using the old low-res versions.

(As just one example, if the ear-covering side-head pieces such as wings, headphones, earcups with antenna, etc. were placed where they really belong in the "Ear" menu, you could mix and match them with the existing "upper face" details such as headbands any way that was needed; the requirement to even *have* a "headband with wings" option is a failure of the costume organizational system. It should be a naturally emergent case of being able to pick a headband from one menu, and side-head wings from another.)

* The organization of the costume pieces is terrible, and the user interface is poor. There's at least some hope that you might be able to argue for some UI designer time or whatever is required to make better sense of the way costume pieces are presented while doing all this, and the odds of us getting a more flexible interface going forward are increased if you have to consider multiple versions of the existing pieces.

For instance, using your example of the Tank Top piece, if you added a sub-option to it, we could start with the existing piece called "Legacy" and add a "Matte cloth" option based on the medium-light weight white cotton or poly/cotton undershirt you linked. Later on, someone might decide that this same geometry could be used for a filmier, shiny casual party top or more feminine underwear, and we'd get a third or perhaps fourth option, "Shiny cloth" and "Shiny cloth with lace" in a Valentine's pack, perhaps. At some other point, a police-themed pack might give us the distinctive vertically-ribbed undershirt worn for ventilation under body armor ("Ribbed cloth"), and an actual light body armor texture ("Light body armor"). Once you have the structure, being able to add pieces later is much easier; but if you go about replacing pieces haphazardly in situ we loose big in the long run.

Currently, you'd use the existing low-res piece for all of the above; once you start getting specific, you need to have more sub-options so that you don't make things look worse for some uses. For example, anyone currently using the tank top to represent the silk top is going to be seriously upset if you replace it with something that looks less like a sexy piece someone might wear to a nightclub and more like something someone would wear fixing their pickup in a trailer park.

tl;dr: The existing low-res pieces are used for all sorts of things that they plausibly might be; if you pick only one thing to represent in high-res and the new version is clearly that, you eliminate all of the other possibilities of what it might be. The only reasonable option that will not get players justifiably upset is to have the original general-purpose Legacy piece as an option, and add additional high-res options as you have art resources to produce them.

Intermittent lag of the sort you're describing is most frequently the result of heat problems. Many modern components will drop back to slower modes if they start overheating, causing exactly the behavior you describe.

Carefully check your intake screens, fans, and heat sinks for clogging and clean them; graphics cards with plastic covers are notorious for building up gunk under the cover where you can't see it that drastically reduces their cooling capability. Make sure wires haven't gotten in the way of your fans, and that they can turn freely. Look at your airflow layout; is your GPU sucking already hot air from after the CPU heatsink, or vice versa?

Reducing your graphics settings (particularly the demanding shadow settings) and cleaning up other things that may be running in the background may help. You can try temporary over-cooling measures (opening the side of the case and pointing a box fan at it, etc.) to see if that helps the situation.

The bit about excluding commonwealths is I believe new, and potentially confusing; I don't know of any particular legal reasons why residents of Kentucky, Massachusetts, Pennsylvania, and Virginia would be excluded, all of which are at one level or another a "Commonwealth".

While in some of the others listed it's merely an archaic formalism, in the particular case of Virginia, it's been the Commonwealth of Virginia since before it was a part of the United States. Most official documents and titles still refer to it as such; "state" is considered basically a shortened nickname.

U.S. State department definition: "The term 'Commonwealth' does not describe or provide for any specific political status or relationship. It has, for example, been applied to both states and territories."

The presumed intent is to cover Puerto Rico and the Northern Marina Islands, which are commonwealths but *not* states, but you've already listed one of the two explicitly.

There's also the issue that you forgot to close your parenthesis, so it could be misinterpreted as excluding Quebec as of Monday, etc.

The whole thing looks like it could use a more careful going over by someone who deals with this sort of thing professionally.

Most individual older and basically 32-bit programs under Windows are unable to use more than 2 GB each. There are obscure settings you can juggle to get this up to 3 GB, but it tends to result in instability at best and frequently outright failure to work; it will be years still before people are reliably writing 64-bit clean code by default.

If your primary system design purpose is playing a single copy of CoH while little to nothing else is running, 4 GB and a fast dual-core is plenty. Once you start expecting to run other things in the background, or run multiple copies / accounts of CoH at once, you'll see some noticeable improvements from 6 GB or 8 GB, and a quad-core. With current technology, you'll be limited by processor concerns or graphical rendering limits long before you need more than 8 GB of memory, even if you're someone trying to run an entire superteam.

Another way to look at it is that with current technology, few to no ordinary home user programs will benefit from more than about 2 GB per core, and only fairly intensive parallel uses will need even that (running multiple copies of CoH does qualify).

As a final note, memory is generally the easiest thing to upgrade later. If something comes up in a couple of years where you actually need 16 GB (I'm doubtful, but predicting the future of computing is notoriously tricky), it's easy to upgrade when it comes up, and probably for less than it costs today in real terms.

http://boards.cityofheroes.com/showp...8&postcount=33 and further in that thread
We've got official confirmation of the following:

* There's a new currency (or two, depending on your point of view); Hero Alignment Merits (HAM) and Villain Alignment Merits (VAM) that is planned for GR
* You will need to have purchased GR to take part
* You will need to have access to Fort Trident (Hero) or the Crucible (Villain) to spend HAM / VAM for purples; access is contingent on being a "true" Hero or Villain (ie, not a Vigilante or Rogue, and having "reaffirmed" your loyalty)
* You will be able to choose the recipe you want; not a random roll
* If you side-switch, any unspent HAM / VAM are lost
* Even if you side-switch, you will eventually be able to get locked into the new side enough to participate in the HAM / VAM system

Not yet fully confirmed with a dev post to my knowledge, but strongly implied based on reports of statements at the SDCC panels:
* You will be able to get at least some purple recipes through the HAM / VAM system
* HAM and VAM are not tradeable, mailable, etc.; they are locked to an individual character like Reward Merits, Vanguard Merits, etc.

Less strongly implied by official statements, but likely given what has been said:
* The previously mentioned series involving "tip" missions leading to a "morality" mission is likely connected to at least the "reaffirm your loyalty" part, and possibly how the "must have GR" test works, as it's been said that characters on accounts without GR will be able to get "tip" missions, but not the "morality" ones.
* At one point it was stated that there was an intended limit on how quickly characters could switch sides; if still true and depending on how it is implemented, it is likely it would affect how quickly one could qualify for HAM / VAM once GR launches, and how long it will take to "prove yourself" if you switch sides.

Usual disclaimers apply; there's already been one major feature dropped from GR (Incarnate slot), and the above could certainly change.

It gives us a base for discussion, however; IF the HAM / VAM (xAM?) system goes live as described, there will be a way for individual characters to choose to defeat enemies in such a way that they get currency that can be spent on buying purple recipes of their choice. Implications?

Some that come to mind:
* People may rearrange their methods of generating original wealth / drops to focus on HAMs / VAMs allowing them to purchase desired purples directly. This MAY result in a reduction of supply of less-desirable purples currently randomly generated, and a RELATIVE increase in their price (effectively reducing the price spread of purples); other effects may swamp this
* More people playing new alts and new players, somewhat less farming overall
* One fascinating unanswered question is whether all purples will cost the same amount of xAMs. Interesting implications either way; how quickly will those effects push out to the market?
* Will there immediately or eventually be other things you can buy with xAMs? Presumably the value of "slice of a purple" will be high enough you'd not want to spend it on trivial things even if available, but with the new Incarnate stuff looming, we may have other stuff to worry about. (Are the devs making purples more available because there will be new rarer stuff to worry about? This certainly wouldn't be the first MMO to have "Green is the new Purple" situations; and there's precedent even in CoH (when was the last time you tried for a Crystal-O?))

Also, are you using dual monitors? There have been some known memory leakage problems with dual-monitor setups recently.

I'd suggest also checking your cables and connectors on both ends carefully. If the video cable is loose, has a bent or damaged pin, or has been damaged and has an intermittent short somewhere, that can cause problems similar to what you are seeing; in fact these days having one or several colors go out is more commonly a cable problem than a card problem. Be sure to check all 4 connectors (system, system end of cable, monitor end of cable, monitor), although the cable connectors are more likely to have problems.

What it probably means is that you shouldn't go back to Cim with pets out until the bug fix that's on Test (1800.201005212331.1T) makes it to Live. In the short term, you may need to wait until the shard's been restarted to go back at all.

"Midnighter Club – Fixed a bug that sometimes caused characters to get stuck in the ground when moving between the Midnighter Club and Cimerora. The previous fix stopped working as soon as a player took a pet through the portal, and would not start working again until the shard was restarted."

(Your issue isn't exactly the same, but it's likely related.)

You might want to take a look at Hugin, the open-source panorama stitching and HDR software. I've not had a chance to play with the 2010 version yet, but the earlier builds I used for a previous project were already quite powerful, and it's got a good mix of stable versions and continuing active development (regular support from Google's Summer of Code helps significantly).

In addition to what others have mentioned, there's a chance that the file on your hard drive that contains the effect for Shiver is bad. Probably a good idea to do a forced verify of CoH, and possibly tell your HD to check for errors (can be time-consuming).

The obvious test would be to try seeing if it's specific to your character, their AT, your system, etc. Ask a friend with Shiver to fire it off near you several times, or perhaps more productively try and arrange an AE test mission. See if the L8 Dom version has the same problems as your L28 Blaster version, whether firing them off in conjunction with other ice powers or onto targets affected with other ice powers makes it worse, etc.

If you can manage to arrange a situation (probably through AE) that causes problems even semi-regularly, someone else can try and duplicate the situation / run the arc on their computer, and see if it's actually a problem with the power or something specifically weird to your computer, like a partly-bad bit on your hard drive.

That's kind of the point; best guess is that there are at least a ridiculously large number of alternate dimensions, quite probably an infinite number of them, and perhaps even a higher-order infinity. Even if dimensions that have a useful non-evil you are one in a million, that's still an infinite number of them out there. There's even a reasonable likelihood that there's an infinite number of them that have successfully conquered their challenges at home, are bored with retirement, and are able to help our world significantly without noticeably leaving their world shorthanded. Of course, the same logic applies to one's foes...

Once you open the multidimensional can of infinite worms, it causes awkward questions from there on out unless you're very careful about the implications and limitations of your setting, which in general CoH has not been.

Some variant of this seems the closest to me. By default, the portal doesn't transit you in space by much if at all, but merely shifts dimensional frequencies. Some SF has the concept that in the general case, the more divergent the alternate universe, the greater the energy required.

As for the naming / coordinate system and structure generally, it seems influenced by a mix of Stargate and Heinlein's The Number of the Beast.

An interesting and important question is whether the universes are quantized, and therefore probably either finite or a low-aleph infinity (with some possible bias toward "countably infinite" as likely); or whether there are real-number-like parameters involved and a much higher order of infinities, although we may not be technically capable of reaching or describing them yet.

One simplistic assumption might be that since the structure of the universes is quantized, if time is also quantized and spacetime is ultimately closed one could arrive at a very large but technically finite number of possible universes; one for each possible value of each possible quanta at each possible time. The difficulty here is that ending up with destination universes that are interestingly different for story purposes, yet on a broader scale astoundingly similar enough to permit travel and survival (let alone alternate selves) is pretty tricky. One handwave might be that the power to go to significantly different universes is considerable, so that the reachable space with current technology is a very near neighborhood; yet that because the equipment used is (comparatively) crude given the forces involved, sufficiently fine adjustments to travel to Heinlein's "World without a J" or even to the one where one of your hydrogen atoms is spinning the other way is not (yet) practical.

Another approach which could be argued is closer to Heinlein's literary perspective, where it's less about the motion of atoms and more about choices. One could handwave that most of the universes we can reach are similar because most of the divergence is recent, with the advent and subsequent spread of sapient creatures that make more explicit choices. Some "river of time" arguments would normally go here, so that "world where I chose to have the blueberry breakfast bar out of the box in my drawer this morning instead of the strawberry one" merges back pretty smoothly into the general "river" of this current narrative universe unless it causes some major change later. The portal technology as we currently know it can then be said to only pick up major "rivers", leading to the desired observed result.

It's not clear whether the devs themselves, whether individually, collectively, or through the device of the story bible, have done much serious thinking about the fundamental structure of the multiverse; it's even less clear whether any such thinking that has occurred is/was either well-informed or well-documented.

Short answer: You do not NEED more than 2 GB of RAM, but you probably WANT more than 2 GB of ram. As games, including CoH, demand more, the number of programs that will run better if they've got 2 GB to themselves will only increase. If you've got older graphics drivers, dual monitors, or do base editing, there are known problems with workarounds you may want to read the other threads on; but in general if you can get your system to 3 GB things are likely to run much better.

More details: Any 32-bit program under relevant versions of Windows can use up to 2 GB of total RAM to improve its performance. Until recently, very few programs a home user would have would actually benefit from this much, but this has been an issue for engineering users and the like for quite a while. (Many programs run increasingly slow and then crash once they get close to the limit, so the practical limit is frequently more like 1.8 GB or so.)

The CoH requirements are woefully out of date. They don't allow for Rikti raids (zone or mothership), zombie invasions, anniversary events, or even being in Atlas while a costume contest is happening. They don't take into account that characters these days have access to many faster ways to switch back and forth through several zones. They don't mean you'll have high frame rates, fast load times, or the like. They are nearly worthless in terms of the sort of experience a player would like to have; they merely indicate that you've got an OK chance of starting the game in the first place.

As features are added to the game, memory size goes up. My minimal testing seems to show that seeing even a simple costume uses at least 3K, and can easily be rather more than that. While the game doesn't *have* to keep everything in RAM, anything it *can* keep around in RAM loads far faster than something it has to go off and search your hard disk for.

Regular use of the /unloadgfx command (after things have stabilized after zoning, to clean out data from the previous zone) can help somewhat, and there are some known bugs with base editing, older graphics drivers, and dual monitors with workarounds.

Ultimately, more and more programs are going to be able to use the full 2 GB for themselves, and there are (somewhat hackish) ways to allow certain programs to use 3 GB. I consider the recommended system for CoH to be one that has 2 GB of free memory for CoH after the operating system and anything else you usually run (web browser, music player, etc.); this usually means that you want at least 3 GB of total RAM, although there may well be substantial improvements in even going to 2.5 GB.

Also note that to actually take full advantage of multiple cores, you need to think in terms of memory-per-core. A quad-core system with only 4 GB of RAM may not actually perform that much better, because while it can run multiple programs, they will be starved of memory and fighting each other for the scraps.