Twenty Questions with the LOTRO Dev Team - July 2013

[Ixinix]

The RNG is fine how it is and doesn't need to be changed, I agree, but that doesn't mean there shouldn't be a system ADDED to help those of us that are unlucky but very dedicated towards optimizing our gear slots.

I'm sure many people have suggested things like barter items or high seal values. Possibly a barter instance so it generates based on your level. Doing nothing about it is really going to drive people away. It's not going to be the only factor but it is one factor that can be alleviated. I am hoping the "tweaks" is related to the suggestions I mentioned.

Totally agree with this.
Turbine, your answer to this topic totally misses the point. Please understand that being unlucky with RNG frustrates people and why would you want to have frustrated people? Putting awesome items in the game is great but never getting it after doing hundreds of the same instance for it is not. What happens is that people give up on the golden items and since they have already run most instances many times what can they do? People will give up on the game because of getting little to no rewards for the work they put in. Please imagine this.... someone does an instance on his main character 100/150+ times and never got the gold item that drops there. Here comes someone that just levelled the same character (which is his 5th alt) to 85 and gets the gold item in his second run in the instance. Can you imagine this makes the person with the main char and countless instance runs want to /uninstall lotro?

Random world drops are one of the worst things that happened to lotro. That and having end game raids without locks.

[Fredelas]

Totally agree with this.
Turbine, your answer to this topic totally misses the point. Please understand that being unlucky with RNG frustrates people and why would you want to have frustrated people? Putting awesome items in the game is great but never getting it after doing hundreds of the same instance for it is not. What happens is that people give up on the golden items and since they have already run most instances many times what can they do? People will give up on the game because of getting little to no rewards for the work they put in. Please imagine this.... someone does an instance on his main character 100/150+ times and never got the gold item that drops there. Here comes someone that just levelled the same character (which is his 5th alt) to 85 and gets the gold item in his second run in the instance. Can you imagine this makes the person with the main char and countless instance runs want to /uninstall lotro?

Random world drops are one of the worst things that happened to lotro. That and having end game raids without locks.

If a player gives up on the game after running an instance 100/150 times and not getting the item he or she wants, what would that player do after getting the item guaranteed in 25 runs or less? Keep playing?

[Witch0King]

I swear to god if Turbine create another Bugud, already had 3 months with nothing from him. I haven't done him for many months. No rewards but 41 marks and a banner..

[phobossion]

The problem with the "old" system or instance loot drops was that it was geared towards kinship-based runs. In a kin, you are running instances with cca the same set of people each time and you can ensure that you'll get all your members geared eventually - simply because the boss will ALWAYS drop the token. PUG and semi-PUG players had the same problem as we have now - they might run the instance 100's of times but because of loosing rolls they never got their rewards. Raid alliances solved this by introducing a DKP system that made sure you will eventually get what you want if you participate. Even if you'd just helped others through the first half of the instance (from which you already had all the gear), you'd still get few dozens of DKP to use when you get to the point of looting the boss dropping what you need. Maybe devs could learn something from that...

[Grimdi]

A2: We have no plans to update the random number generator (RNG). It’s normal for individuals to receive runs of luck (good or bad) in the short term. The random element of the game, may appear to be broken, but the RNG is running for the entire server instance, not for one player at a time. - Nathan

Thanks for addressing this issue, even though it's just the standard deflection answer. I'm still convinced there is a (design?) bug somewhere in the RNG system though, but as we don't get the actual RNG numbers in most places it's hard to argue from this side other than with anecdotal observations.

A13: There is more than one cause for lag today, and as time goes on and the game (and the internet) change, new causes of lag are introduced. We have folks who work on this all the time, and if you are not satisfied with the results, well, know that we aren’t either. We have some brand-new performance monitoring technologies that we have recently put on the live servers, which have started to provide important clues. We are hoping for some significant improvements with the expansion pack, especially for some of the Mounted Combat cases. - Nathan

This actually concerns me a bit. To me it would make much more sense to decouple performance improvements from content and feature updates to verify their effect. Or is the class revamp supposed to be the performance improvement?

A15: For instance or skirmish content:
Lots and lots of playtests and iterations.
Difficulty is intended to increase as both tier and group size increase.

We attempt to tune tier 1 to be beatable by our Palantir players, tier 2 by bullroarer, and then challenge for live. The reason for this isn’t because players on live are better than Palantir or Bullroarer players, it comes down to the groups being better because they’ve been playing together as the same group for years. There’s a familiarity with each other and a gelling over time as kinships that allow for different levels of coordination and collectively overcoming presented problems and situations, that more randomly put together groups of players simply do not have. If we beat tier 2 in house, then it’s not hard enough and gets balanced as such. Many challenges haven’t been beaten by anyone when they launch. We know they’re functioning and the numbers are where we project is doable, but we don’t have the coordination to pull it off ourselves.
Small Fellowship we tune to need a tank or a healer, but try to avoid requiring a trinity, even within challenge level difficulty.

Wait, does that mean you don't actually verify T2 and challenge mechanics are working correctly before Bullroarer testing? That said, are planning to look into rebalancing some of the scaled content for Helms Deep? As it stands there is a vast disparity between certain instances (e.g. Sambrog T2C vs. Lost Temple T2C) in both challenge and time required that isn't reflected in the rewards.

One of the biggest tricks is to be aware of the numbers when it comes to equipment and gear but to not be overly bound by it. Introducing chaos, and requiring communication and coordination by increasing the number of dynamic elements involved creates a far more challenging and satisfying fight than simply raising the health and damage. - JWB

Nice to see some design goals stated, this helps understanding certain instance designs like flight (think you've overtuned the chaos part in that for t2c though).

A19: We are looking at several tweaks to the loot system at the present time. Some may come to light for Helm’s Deep and others will be bigger improvements going into next year. I can’t really give a ton of detail on these changes as they are really in flux right now. - Zimmitti

Again thanks for finally addressing this issue. While I understand that you can't provide information on the actual changes yet, how about giving some info on the design goals?

[Thrabath]

Sapience, can you tell is if the license is extended? Because to be honest the question/answer isn't very clear about that.

You guys want to support/develop LotRO for many years to come and by saying YES, we use the extending-option, you can do it till 2017, but as long as extending isn't officially signed/agreed upon, it's still a mystery. I have the feeling this hasn't been done yet because otherwise you would probably have said "Yes, the license is extended"

[Tecro]

So there is no signature under an agreement to extend the license. If there was Turbine would say so. This is very disappointing.

ETA: of course this doesn't mean it won't get signed before it runs out and if it runs out that doesn't mean that the game has to go away instantly. But there's no contract signed.

I have read a couple of stories concerning this and Turbine does have the rights to extend the current license agreement for 3 more years if the story was accurate so we would at least get LOTRO until 2017 which would be plenty of time to finish the story imo.

[Danchir]

Quite some good news on the 2nd episode. I definitely like it more than the 1st.
Looking forward to the expected performance/technical improvements lagwise and otherwise.

I didn't see any development on the instances matter. Is it decided whether there'll be new and/or revamped instances with HD (not necessarily on release), or is it too early to ask?

The initially planned 'terrorist - antiterrorist' idea for PvP looks quite nice. Too bad it didn't work out.

[Forums still logging you out. Looking forward to taking care of that too.]

[BiteMarks]

Q8: What is the future looking like for pvp? Is it basically a done deal for Turbine with no future? Is it considered a burden that they just want removed from the game? Or is it something they would like to revive and try to make better? Are the couple PVP only council members basically wasting their time with trying to work on pvp stuff?

A8: We don't have any serious big features planned for PvMP in the near term. We've got time booked to rebalance things to accommodate the Freep class changes and we've got a few ideas that might be interesting... or really may not happen.
The systems changes we are putting big effort into right now on the freep side are focused and challenging. If I were to shake up PvP while trying to get everything else balanced, I assure you everything would suffer. – HoarseDev.

Not sure what this can truely mean. Saying that you do not have anyting big planned yet you say you have freep class changes which ALWAYS means creeps lose out and are forgotten about.

Q9: What about small PvMP features? fixes? little mechanic changes? Did you know that the bridge from TR rez to the outpost has 2 bridges rendered?

A9: In the short term, we're going to be focused on the effects of the class changes to PvMP balance. These changes have the potential to change a ton of dynamics, even though they are not changes directly to Creeps. Time and feedback will be best spent there. I've got the threads in the suggestion forum to chew on and there is good stuff in there. For now, I'm not going to open those worm cans just yet. - HoarseDev

Yes, adding new skills/changes for freeps is fine as they have about 95% of the entire game to play in yet all those changes do have a dramtic effect in the Moors. Always had and always will and for months after, we creeps have to suffer the pain and neglect for your poor service.

You keep on about balance and have said so in the past and how you are going to address it, yet here once agian you are pushing more updates to freeps. Devastating and crtis of anyting above 6k can kill a 1v1 fight and freeps have these pretty regularly yet creeps are lucky to get anyting above 1k crit/dev.

In the past you have said creeps will be looked after and when we get anything, it soon gets either removed or nurfed back as the freeps can't take a proper fight and if we get even a little increase in skills etc, all hell breaks out on forums with complaints from them. How about looking after us creeps a lot more

Do creeps get anything in the upcomming update? Are we forgotten once more and yet we pay the same price as freeps yet have less maps to roam around on, less skills, less of the entire game which actually means we PAY MORE for the game then freeps do. Full time creeps should be better rewarded or actually make it so you can just pay for creep side on its own.

Do we get anything even remotely worthwhile? Not even a simple mail box or shared vault

I am now one of many friends that have simply refused to pay/renew their subscriptions for the lack of proper service we full time creeps have had or are going to get in the future and feel that Turbine/WB are losing out on a very lucrative market.

Thank god for f2p

[Ixinix]

If a player gives up on the game after running an instance 100/150 times and not getting the item he or she wants, what would that player do after getting the item guaranteed in 25 runs or less? Keep playing?

Thanks for pointing this out.

I have a total of 7 max level chars. What happens with the current system is that I feel that I HAVE to do instances on my main for a chance of that awesome item. I always liked to experience the different aspects of a dungeon, healing today, tanking tomorrow and DPS-ing the next day. This is what kept instances interesting. I have run Sammath Gul on-level hundreds of times but with all different characters so it never got that boring. With the RNG drops I'm scared that I win a gold item on my 7th alt that I almost never raid with where I would really like to have the gold item from that instance on my main raiding char.

Now, back to your question...
I remember in the past there was a point where I could say my first (main) char was in maintenance mode. For me this meant I basically got all the gear I wanted on my main but could still do instances for a particular jewel and ofcourse use my maxxed out char for raiding. When one character was "done" I started working on the next and before a new update I would have 4/5 out of 7 characters fully geared. I tend to get bored a bit if I play the same character for a long time and this is exactly what happened to me. When I can not get the best items available for my character even when putting a lot of effort in it it demotivates me to play. I barely touched my alts after levelling them to 85 because I always feel the need to go on my main.

So to answer your question what a player would do after getting an item guaranteed after 25 runs...
Play other instances 25 times until you have all the items you want for your main and then go to your next character and do all instances 25 times on that until you are happy with the gear you have. I can make running the same instance over and over again fun by playing with kinnies or friends, trying funny set-ups (5brg, 1 cap SG runs anyone?) and knowing there is a light at the end of the tunnel. There is no light now!!

[Laire]

I agreed that there need to be a barter system beside the RNG. Just like the treasure even reward. You can get the goat from RNG site, but also, with every site you dug, a little closer to the bartered goat. Some people said about "fair", but how it could be fair if someone just run an instance 1-2 times and got the best in slot item, while other ran it few hundreds times and nothing return.

Agreed. A kinmate of mine, on an alt he rarely plays, won a coveted hunter class item... his first run through on that character. It took me 316 runs. One of my kinmates is over 500 runs without it dropping. I know someone who ended up outright quitting the game after their 1000th run of Dungeons with no champion boot drop.

How on earth is that fair? A barter system should be introduced alongside RNG. Worked wonderfully for first agers-- and the amount of tokens one gets increases significantly depending on the difficulty of the instance, to get people to actually do hard t2 stuff.

And some of the best recipes in the game gotten most efficiently from botting (which is what a couple Elendilmir kins did) instead of raiding is outright silly and a dang shame. These recipes should have been in instances only, or at least ALSO in instances.

Also, what Inix said rings perfectly on why we replay it. People have lots and lots of alts. And also, a good deal less discouragement from people who group.

[Argenaut]

Having worked on licensing issues for quite a few years (and software development contracts, etc. as well), there are a number of reasons why Sapience cannot be more specific.

First and foremost, there is undoubtedly rather restrictive confidentiality language. Many if not most contracts of this type do. Only some aspects of the deal can be made public. Anything else is strictly forbidden. Undoubtedly, Sapience has received clear instruction from Turbine Legal about what he can and cannot comment on specifically.

Second, as pointed out before, many contracts have options to extend, or discreet lengths of time (terms) that are covered in detail under the contract and are triggered by certain events, notices, and decisions. Related to this, it would not make good business sense for either side to commit to, say a twenty year license, where if WB/Turbine cannot make the game financially viable, they would still be financially obligated to pay twenty years in license fees. So, it is broken down into segments. Similarly, MEE/Saentz would be foolish to commit to an open ended license with a company that might not do the intellectual property justice. Nobody who owns a wonderful intellectual property as Tolkien's works, and who has affection if not a love for them, would entrust it into the hands of a company that might change over time and start diluting or corrupting it.

You have to police your intellectual property when you license it out!

Sapience has probably answered the question about the license length or extension as best as he can. And for many years to come is a pretty good answer in what is otherwise an unknowable future.

[ColorSpecs]

A15: For instance or skirmish content:
Lots and lots of playtests and iterations.
Difficulty is intended to increase as both tier and group size increase.

We attempt to tune tier 1 to be beatable by our Palantir players, tier 2 by bullroarer, and then challenge for live. The reason for this isn’t because players on live are better than Palantir or Bullroarer players, it comes down to the groups being better because they’ve been playing together as the same group for years. There’s a familiarity with each other and a gelling over time as kinships that allow for different levels of coordination and collectively overcoming presented problems and situations, that more randomly put together groups of players simply do not have. {...} JWB

Isn't it ironic then, that the other systems Turbine refuses to update - Loot distribution, PVMP (audacity) and the RNG - are the very things that are driving the players that group together for years away from the game you're creating challenge mode for?

[ColorSpecs]

If a player gives up on the game after running an instance 100/150 times and not getting the item he or she wants, what would that player do after getting the item guaranteed in 25 runs or less? Keep playing?

Yes.

The most popular instances in the game did not require luck to get things to drop. Just to win the drop. Going all the way back to Carn Dum and Uru, the class items always dropped. In the Rift, the gems always dropped. In Moria, the tokens for radiance always dropped. In SG the challenge boss always dropped something good.

Those instances where things sometimes dropped, like Barad Gulleran or pre-scaled Helegrod for example, were never as popular as the ones that had a guaranteed drop.

[Namesse]

If a player gives up on the game after running an instance 100/150 times and not getting the item he or she wants, what would that player do after getting the item guaranteed in 25 runs or less? Keep playing?

I'm with others here, keep playing.

Back in MoM when the different instances always dropped gear tokens, I ran the instances for my gear, my alt (really only 1 at that time), my kinmates' main and alts, friends' main and alts; we'd roll from one instance to the next to get the required radiance gear and had a lot of fun. After everyone had the minimum gear we'd still run the instances for deeds and for the good jewelry drops - even if the desired loot didn't reliably drop, it dropped more frequently than 1 out of 100 or 1 out of 500, even if I didn't win the roll. Just seeing it drop on an achievable basis made me and others believe that it was possible to see it drop and win the roll.

In short, it was easy to be helpful to kin and friends to help them get guaranteed gear compared to running an instance multiple times where no one gets anything other than a bounty and a relic.

[Nindolvoth]

Add a barter system where you trade loot pieces you don't want or need for currency. Once you get enough currency you could buy the item.

[Turgos]

Please go ahead with developing these small (perhaps closed?) PvMP zones. Please Please Please Please!

And don't forget to put enough PvE content in those zones... NOT!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!

[Frisco]

Having worked on licensing issues for quite a few years (and software development contracts, etc. as well), there are a number of reasons why Sapience cannot be more specific.

First and foremost, there is undoubtedly rather restrictive confidentiality language. Many if not most contracts of this type do. Only some aspects of the deal can be made public. Anything else is strictly forbidden. Undoubtedly, Sapience has received clear instruction from Turbine Legal about what he can and cannot comment on specifically.

Second, as pointed out before, many contracts have options to extend, or discreet lengths of time (terms) that are covered in detail under the contract and are triggered by certain events, notices, and decisions. Related to this, it would not make good business sense for either side to commit to, say a twenty year license, where if WB/Turbine cannot make the game financially viable, they would still be financially obligated to pay twenty years in license fees. So, it is broken down into segments. Similarly, MEE/Saentz would be foolish to commit to an open ended license with a company that might not do the intellectual property justice. Nobody who owns a wonderful intellectual property as Tolkien's works, and who has affection if not a love for them, would entrust it into the hands of a company that might change over time and start diluting or corrupting it.

You have to police your intellectual property when you license it out!

Sapience has probably answered the question about the license length or extension as best as he can. And for many years to come is a pretty good answer in what is otherwise an unknowable future.

The problem is, the same sort of vague non-answers such policies would require are the same exact non-answers that would be used to placate a distressed playerbase by a spokesperson who isn't allowed to tell the entire truth. The answer to question 16 *is* vague, and there's no way we'll know the reason why until either official notice of the extension or notice of shutdown is announced.

I would've much rather seen a straight-out honest answer like "we can't comment on a license extension until it is finalized" instead of what we got. And the worst are the condescending replies from Turbine insisting that it's not vague, and the allusions to tinfoil hats.

All that it takes to see the vagueness is to imagine what the PR comment would be if the game WAS ending and the extension wasn't signed. Pretty much exactly what was said in the Q&A. A non-binding answer followed by a non-sequitur. The relationship with MEE isn't relevant, is it? Turbine holds the option, so the deciding factor in continuing the game would be whether or not WB wants to continue to develop it, not MEE, right?

So either something *actually* got signed/agreed upon and Turbine WILL be developing the game for 3 more years, a brand new contract was signed that actually gives the rights for "many" years, or nothing was signed, and the "plan" is to develop the game for "many more" years. But we still don't know. It's not conspiracy theory, it's just that one shouldn't accept even strong hints (wink wink nudge nudge) as fact.

[Frisco]

If a player gives up on the game after running an instance 100/150 times and not getting the item he or she wants, what would that player do after getting the item guaranteed in 25 runs or less? Keep playing?

In my experience, some people keep playing after they get their stuff, some don't. Can't be helped either way. Some people will quit after trying really hard to get something and failing, some people will hear the odds and never bother trying.

What I do think, though, is that the drop rates of the gold class items are very short-sighted. Sure, more people may be willing to run the same instances 100+ times to get the best gear in the game, but since it's pretty clear that scaled content is going to have a big place in the future of this game (and rightfully so--no reason to waste all that hard work when a few tweaks can make the old stuff enjoyable again [note: I do not believe they did enough tweaking to most of the scaled instances to make them enjoyable, but there's still potential]), why burn people out with 100 Ost Elendil runs at 85 when you are banking on the fact that they will see Ost Elendil as endgame content again at L95?

[SNy-lotrolinux-EU]

Q2: Are there any plans to change the RNG? I know a lot of people are beyond frustrated at it, as it doesn’t seem as random for some as it does for others.

A2: We have no plans to update the random number generator (RNG). It’s normal for individuals to receive runs of luck (good or bad) in the short term. The random element of the game, may appear to be broken, but the RNG is running for the entire server instance, not for one player at a time. - Nathan

Well, in general, this type of Q/A on a somewhat regular basis is appreciated, but I would like to voice my concern over this one here, like some other posters already did.

I think that, in fact, the question is a poor one indeed.
See, this topic has been discussed to some extent in other threads, and while maybe not a general consensus, the soundness of the pRNG results as such has, I think, been found to be OK-ish. As in: the function that returns pseudo-random numbers in some given range is assumed to work within reasonable parameters (some experiments with crafting etc.).

HOWEVER.
The soundness in the distribution of raw pRNG results does not bear any direct relation to the soundness of specific results in its application to certain problems (especially, but not limited to, the individual loot distribution system).

I want to stress that part.

While the pRNG itself might work perfectly fine, there is a possibility that its application to the loot distribution is, in fact, flawed and the system therefore not working as intended. I am not saying it is, I am saying that from a somewhat educated standpoint, the possibility cannot be dismissed lightly without actually taking a look at the code.

Anyway, threads and postings to the extent that things seem to misbehave have been in the forum for months, so far without any reaction I am aware of (other than Q2, of course, with the flaw as outlined).
Also, I want to bring your attention again to a former faux-pas in said application of the pRNG, otherwise known as the Wi-Flag.

So, please, consider this revised question:
Has the application of the pRNG to the system of the new, individual loot distribution been scrutinized and verified to work within the expected parameters?

In closing, I sincerely hope that this is not going the way of recent discussions (viz: accusations of 'not liking' or outright ignoring answers given) and appreciate any worthwhile discussion on the issue of pRNG application.

HAND,
SNy

[Annariel]

I agree. People in general don't understand randomness or probability. If they did then casinos would not be a viable business. So what I think might be acceptable is a system that is NOT random but which feels random. Ie, if the chance to crit is 25%, then it would artificially make sure that you can never get a string of 6 losses in a row. Not at all random but it feels like what the average person thinks is random.

The other solution is to discard the randomness or preception of randomness completely. No more crit vs non-crit for example. A lot of games do this. But that's a much bigger change overall, upending what a lot of people expect here, tossing out the current (broken) economy), etc.

(My apologies if something like this has already been experimented with - I've never had a level-capped character.)

There's also the 'randomness' of crafting nodes like ore or wood. The name indicates whether it's going to give 1-3 or 4-6 chunks (or whatever), but whether it's one, two or three (or four, five or six) appears to be random.

The nice part is that this gives a finite number of possibilities for both the number of chunks you'll get by mining X nodes and the number of nodes you will need to mine to gather Y chunks.

I think this could be used in combination with a barter system to ensure that some players get their item sooner than others, but in the end, if you run it enough times, you're guaranteed to get it.

It would be kind of similar to the treasure hunt system, except there you can also find items (or a combination of items and tokens) - here, the reward would always be tokens, but in various amount.

[Annariel]

Wait, does that mean you don't actually verify T2 and challenge mechanics are working correctly before Bullroarer testing?

That's what I thought when I first read it, but then I realized that if I had to test content like this, I'd use some kind of 'cheat' - maybe give the whole fellowship Inspired Greatness, or some other trick like that. That way, I could verify all the mechanics reasonably quickly.

[cdq1958]

(My apologies if something like this has already been experimented with - I've never had a level-capped character.)

There's also the 'randomness' of crafting nodes like ore or wood. The name indicates whether it's going to give 1-3 or 4-6 chunks (or whatever), but whether it's one, two or three (or four, five or six) appears to be random.

The nice part is that this gives a finite number of possibilities for both the number of chunks you'll get by mining X nodes and the number of nodes you will need to mine to gather Y chunks.

I think this could be used in combination with a barter system to ensure that some players get their item sooner than others, but in the end, if you run it enough times, you're guaranteed to get it.

It would be kind of similar to the treasure hunt system, except there you can also find items (or a combination of items and tokens) - here, the reward would always be tokens, but in various amount.

That is the way that loot works now, except that the upper and lower bounds are looser. I forget the names applied, but basically you have a look-up table. For example, a number from 0 to 8000 gives you common loot. 8001 to 9000 gives you better loot. 9001 to 9500 gives you even better loot. 9501 to 9750 gives you the next set. 9999 gives you the best loot.

That said, having a barter option based upon Marks, Medallions and Seals would help.

Random means that when an event happens is not predictable, not that it can't happen as long as said event is possible.

[cdq1958]

Snip


Wait, does that mean you don't actually verify T2 and challenge mechanics are working correctly before Bullroarer testing? That said, are planning to look into rebalancing some of the scaled content for Helms Deep? As it stands there is a vast disparity between certain instances (e.g. Sambrog T2C vs. Lost Temple T2C) in both challenge and time required that isn't reflected in the rewards.

Snip

It means that the limits of the Palantir and Bullroarer programs preclude exhaustive testing. Pharma example. Med X gets developed by chemists. It undergoes safety testing on animals first. Once past this, a group of 10 to 20 young, healthy individuals are picked for initial safety and pharmacological (dose-response) tests. This group is inherently not representative of the general population. Once past this, Phase 2 trials begin. These are larger groups but still are unrepresentative of the general population. After acquiring enough initial (and unrepresentative) response data, then the Phase 3 trials begin. These are 'double blind' and increase in time, expense and size. Here is where the rubber meets the road. If good enough, the FDA approves the drug. Now, though, the samples tested are still unrepresentative of the general population. Now you market it to millions/billions. Now you get real world safety and efficacy data. Some products will be taken off the market, some will have their indications limited, some will be great for some subgroups and bad for others.

[lele0511sd]

I would like them to put into practice something to improve the instance finder, I see in games like neverwinter and WOW, if you do not arrange for the fellow instance and activate it so you own the instance finder seeks players for you so close to the group, this and very interesting!