FOH Misinformation

[Ratt]

This thread is in reference to this:

http://www.fohguild.org/forums/showthread.php?threadid=2252

Once again, FOH is disseminating misinformation. I've never responded to these types of threads before that contain so much bogus information is hilarious. However, I felt the need to speak up. I sometimes wonder if FoH is somehow in cahoots with Verant to spread weird rumors. I'm not one to subscribe to conspiracy theories, but some of the posts these guys make are just so far off the deep end and so far skewed from the TRUTH that it makes me wonder.

Anyway, the fact is, there's no frigging "soft cap" on mana, at least not one that I've found. I have tested far past the "4250" cap they quoted, and casting a 100 mana spell results in mana that is appropriate in reduction. Using the client UI with %% as testing is like using a yard stick to measure millimeters. Why the hell would you do something like that?

I guess if a yard stick is all you got, then you gotta measure with it... but I _KNOW_ the FOH crew uses SEQ, so why the false information? A simple, 5 minute test with SEQ would reveal the truth of the issue.

As I stated a LONG FRIGGIN TIME AGO, the +mana scale for 200+ int is a SLIDING scale that I was NEVER able to pin down the exact formula for. It is not, nor has it ever been a +6 per int over 200. I thought I had pinned it down to 6.5 per int, but that turned out to not fit the curve either.

The entire Verant mana system is fairly complex in regards to INT/Mana items, and anyone using the UI as a measuring stick is either just plain ignorant or wants attention. To add insult to injury, a couple of the people posting seem to have forgotten that MANA UPDATES only come when a spell is cast. Your client estimates how much mana you *might* have, and a mana cap at 4250 IN THE CLIENT may exist but it does not appear appear to exist ON THE SERVER... as evidenced by CASTING SPELLS (duh) and looking at the resulting mana.

Now, in closing, I'm not saying they are 100% wrong. What I am saying is that everything they've provided for evidence so far is either so horribly skewed by bad data, or outright propaganda that taking anything said in that post as of this writing is like buying Enron stock. Maybe Frozboz works for Arthur Anderson? :)

Anyway, please do more research with more valid tools than the VERANT PROVIDED UI before making grand pronouncements like this.

Thanks

[Ratt]

Frozboz posted a clarification:


By softcap I am implying you can still go over 4250 with KEI, familiar, or other mana buffs, but items cannot take you past that number.

Addressing this issue, I believe this to be true now clarified.

The reason behind this is not a "soft cap" per sey, as much as it is from the fact that you can only have a certain %% of +mana gear over your "base mana" from INT.

This has been a known "cap" for several years now, and I think what he's seeing is someone who is up against the +mana-gear : +mana-int ratio. This was/is a known thing, and it's nothing new... been like that forever. Why they chose to make a big fuss about it now is beyond me. Verant even acknowledged the "cap" to be in there (a couple years ago).

Oh.. and BTW - the "soft cap" on INT is 252, not 255. You get zero mana for INT above 252.

[Cryonic]

Just read through the posts. Looks like they are claiming a cap of +1400 in +mana items. Is it possible that there is a +mana item cap? Wish I could test this, but don't have a int or wis caster above 50, hell I never got any of my chars to 50.

Guess the best test is something like this:

Start with just your +wis/+int gear on. Med till you are full. Test by casting an effect from an item (jboots, symbol of Innoruuk, etc...) till SEQ stops showing your mana go up. Add a +mana item, med to full, cast item again. Keep doing this till you surpass the +1400 mana item "cap". Does SEQ still show your max mana going up?

That seems like a valid test for this scenario.

[fgay trader]

They also completely ignored the fact that putting on +MANA items increases not only the mana cap, but also the actual mana. The percentage will never change that way - both current pool and the cap grow together.

Putting on +INT (or +WIS depending on class) items, on the other hand, only increases the mana cap, not the current mana pool. That's why the percentage drops when you put on WIS/INT gear - the Max mana grows while current pool stays the same.

Someone pointed out that their tests were invalid somewhere on 3rd page or so, but everyone seemed to ignore that post.

[Aniar]

FGay said: "They also completely ignored the fact that putting on +MANA items increases not only the mana cap, but also the actual mana. The percentage will never change that way - both current pool and the cap grow together."

No, I'm not a ShowEQ user, I'm not smart enough to be one, and I couldn't find an answer in a search...

If this were true, why can't I remove +mana items, then re-add them and cast another nuke?

[domesticbeer]

See Ratt's post

You get a mana pool update only when you cast a spell. So theoretically it would take 2 casts to update the mana pool to reflect that you have the extra mana.

Cast 400mana dd
remove mana item
cast jboots till mana update is recieved
add mana item
cast 400mana dd

Seems like a lot of work.

[fgay trader]

Originally posted by Aniar
...why can't I remove +mana items, then re-add them and cast another nuke?

Do you mean when you're OOM? They fixed that exploit a while ago. I believe when you take off MANA items while OOM your mana count actually drops below zero.

It looks like they made it work just like taking off +HP items when you're very low on health. I've seen my HP go into negatives just before the "LOADING PLEASE WAIT" message when I'd pull HP rings to escape exp deaths on Tallon.

[Aniar]

Makes sense, thanks FGay

[Gullork]

I dunno. It may be an established cap and all for the last several years, but Verant continues to put out increasingly more uber items every expansion. The longer it goes, the more the cap makes a difference as the better the gear gets that could easily outpace it. I rather think the 50% or whatever it is cap on mana vs. int/wis is a fair idea... until you get to level 60.

At level 60, Verant should just remove the darn cap. Otherwise why continue to put out even better equipment?

Kind of an interesting read FOH provided us though lol. Thx for info Ratt. :)

[throx]

Can someone explain why the people testing found that taking off a pure +mana item and putting it back on again when at full mana caused them to appear to lose mana? I've always been under the impression that +mana items didn't add to your current total when you put them on. The negative mana idea is interesting, but I doubt it is based in fact. Try it sometime - cast until oom. Time how long it takes before can cast spell X. Cast until oom again and then take off a +100mana item. Time how long before you can cast spell X. If "negative mana" exists it will be longer. If not, the times will be the same.

I've been known to be more often wrong than right. Just shooting ideas out here.

As for the "measuring with a yardstick" comment - as long as your yardstick is accurate you can measure things MUCH smaller than a yard with it. It's a throwaway comment that sounds good on first impressions but doesn't hold water when examined closely.

Assuming the client uses the same mana formlae as the server (big assumption, I know), looking for the 100-99 transition and measuring exactly how much mana you have to take off to hit it can be a very useful tool.

[Cryonic]

1% of 4000 mana is 40 mana. That is the resolution of the clients display. So it seems stupid to think that they can tell if +3 or +5 mana is pushing them over the limit.

[throx]

Each percentage point is around 40 mana, true. This doesn't mean that this is the finest resolution you can measure.

To take an analogy, I can easily determine the thickness of a sheet of paper using a yardstick. I just have to stack the paper to the height of a yard and divide by the number of sheets in the pile. Similarly in EQ, if you adjust your +mana items to slowly increase the amount of mana you are adding and removing by a small number then you can determine the cutoff which causes the percentage marker to move (like they did on the FoH boards). You can also cast low mana spells many times and measure the % change in mana to get an accurate indication (outside of SEQ) of how much mana each percentage point represents.

[Ratt]

Originally posted by throx
Can someone explain why the people testing found that taking off a pure +mana item and putting it back on again when at full mana caused them to appear to lose mana? I've always been under the impression that +mana items didn't add to your current total when you put them on. The negative mana idea is interesting, but I doubt it is based in fact. Try it sometime - cast until oom. Time how long it takes before can cast spell X. Cast until oom again and then take off a +100mana item. Time how long before you can cast spell X. If "negative mana" exists it will be longer. If not, the times will be the same.

I don't even know why this is a question. The answer is fairly obvious. When you put on a +mana item, the top end of your mana pool increases.

If you have 2000 mana and put on a +200 mana item, you still have 2000 mana, but you can now med to 2200, instead of 2000.


As for the "measuring with a yardstick" comment - as long as your yardstick is accurate you can measure things MUCH smaller than a yard with it. It's a throwaway comment that sounds good on first impressions but doesn't hold water when examined closely.

Yea, and you can whack off with a weed eater too, but that doesn't mean it's the best way to go about it. As long as the weed eater feels good... you can whack off with it. *Most* people would find that kinda strange... just like *most* people would find measuring something at millimeter values kind of odd using a yard stick. Your garden variety yard stick is much less accurate at those levels than your garden variety millimeter ruler.

Using the client interface has two drawbacks:

1. It's supplied by Verant and no external testing or verification can be done on it's accuracy. This applies to both unintentional data skew and intentional skew, which I don't believe anyone here would argue Verant is guilty of both.

2. 100 ticks on a measure of 4000+ possible points on a line is NOT a very valid statistical gradient when we are trying to verify points to 1 degree of accuracy. (IE - is it 1388, 1389, 1421, etc...) We need SINGLE point accuracy, the client does not provide this information to the user.



Assuming the client uses the same mana formlae as the server (big assumption, I know), looking for the 100-99 transition and measuring exactly how much mana you have to take off to hit it can be a very useful tool.

As stated here is totally false. Using 1 mana will drop you from 100 to 99, so it's totally invalid. Do you really only have 100 mana? No... going with larger values, coupled with mana regen, and a huge yardstick as a measuring tool, you can not get accurate data that means anything.


Each percentage point is around 40 mana, true. This doesn't mean that this is the finest resolution you can measure.

To take an analogy, I can easily determine the thickness of a sheet of paper using a yardstick. I just have to stack the paper to the height of a yard and divide by the number of sheets in the pile. Similarly in EQ, if you adjust your +mana items to slowly increase the amount of mana you are adding and removing by a small number then you can determine the cutoff which causes the percentage marker to move (like they did on the FoH boards). You can also cast low mana spells many times and measure the % change in mana to get an accurate indication (outside of SEQ) of how much mana each percentage point represents.

Again, I disagree. There's too many variables to measure your mana accurately (regen, ft, SCM, etc...) with only 100 data points available over a 4000+ point spread. Hell, even WITH SEQ, it's fairly hard to pin down exact numbers due to all the variables. You are kidding yourself if you think you can use the client to give anything but the most crude estimates available, and even those would be suspect due to the nature of the data gathering mechanism.

[Cryonic]

If you could do it fast enough that Mana regen didn't impact it. As a cleric I would practice conjuration by summoning food/water (5mana/cast). SEQ only showed a drop of 3 mana per cast. Without that information, you end up with a big error in the amount of mana used if you think it is still costing you 5 mana a casting. Granted you could just cast a 100mana spell and reduce the impact of mana regen (because you are gaining mana even as you cast a spell).

[throx]

Relpying to Ratt:


I don't even know why this is a question. The answer is fairly obvious. When you put on a +mana item, the top end of your mana pool increases.

Exactly. FGay said differently, which is why I asked the question. By extension, when you take a +mana item OFF you are reducing the top of your mana pool so quickly taking an +mana item off then on will reduce your pool by the amount of +mana on that item.

If you med to full at 2200 mana (using your example), then take off a +200 mana item and put it back on again then you have 2000 mana. If you take off and on a +1 mana item you have 2199 mana.


just like *most* people would find measuring something at millimeter values kind of odd using a yard stick. Your garden variety yard stick is much less accurate at those levels than your garden variety millimeter ruler.

Yes, but when you only have a yardstick it is still possible to measure to that accuracy. As a matter of fact, physicists, chemists, bioligists and pretty much every brand of science does this sort of thing routinely. You can get an extremely precise measurement using crude equipment if you have to.

As for the whole weedeater thing, whatever takes your fancy. Probably a good description of the client though.


Now, back to the accuracy thing:


We need SINGLE point accuracy

Using 1 mana will drop you from 100 to 99

There's your single point of accuracy. Coupled with what you confirmed previously, if I have a +1 mana item and remove it quickly and put it back on my mana will drop by one point. If I am below the mana cap then this will drop my mana percentage from 100 to 99. If I am above the mana cap then it will NOT drop the percentage from 100 to 99.

You can therefore determine with 1 point accuracy what cap is in +mana items using only the interface and the things you've states are a fact.

This won't give you the measure of your total mana, but will give the EXACT number of +mana which is the cap at your level, assuming you have sufficient variation in the +mana gear to change your total additions by one at a time, exactly as done by several people on a number of different boards.

[Cryonic]

That all depends on where the client rounds off at. If 40 mana represents 1%, but it rounds off after 25 mana, then you still have a fudge factor of 15 mana for that %.

I still think that it should just be tested with SEQ (which truly does have 1 mana point accuracy) and see what we come up with and who cares what the rest say.

[PainNSuffering]

Yea, and you can whack off with a weed eater too, but that doesn't mean it's the best way to go about it. As long as the weed eater feels good... you can whack off with it. *Most* people would find that kinda strange...

don't worry your not alone here ratt.

just take take off the plastic fishing wire it comes with, and attach some large rubber tubing (like they use in a sling shot). oil it up with a good lube and put on at least 2 condoms (to minimized impact). lube up both of the condoms. you get a great feeling slap and wrap effect, then the nice tight pull and it unwraps, oooooooohh! nothing better than fireing up ol betsy for a lonly night.


oh, what was the topic here again?

[domesticbeer]

Just one thing....


hmmmm that was disgusting!!!

[Neuro MT]

The cap on items is 1388. I tested this myself. You complain about the methodology, well, I didn't test it casting spells. I simply removed items and put them back on again to see a change.

The client will display any mana amount under 100% as 99% (Or less). The client does not round up, so even a 1 mana change will register. Test it yourself if you don't believe me. Up until 1388, even a 1 mana increase in mana pool will drop your mana to 99% for one tick. Above 1388, even a 125 mana item will not budge your mana meter.

This cap, I believe, is a direct ratio of your total int-based mana. Someone told me that SEQ showed a 4164 total mana pool unbuffed, even when the pool should be higher. Since 4164 divided evenly into 3, AND the resulting divisor was '1388' I concluded that not only was the 4164 figure accurate, but the 'cap' was actually a function of your total mana pool, basically, 50% of your mana pool is the cap on +mana items. You say this 'limit' was known for a long time, well, not by the general public. I knew there was a cap on +mana under level 20 or so, but I had never heard of any limitation above lvl 20.

Also, for your comment that stats stop at 252, that too is incorrect. I have seen my mana pool increase going from 253 to 255 int, so you get mana gain all the way up to 255.

Try actually testing these statements before you make them.

[Ratt]

Originally posted by Neuro MT
The cap on items is 1388. I tested this myself. You complain about the methodology, well, I didn't test it casting spells. I simply removed items and put them back on again to see a change.

Again, your methodolgy is invalid. You are using a closed source, un-verifiable method to test. I, and many others, do not trust what comes out of Verant. You are crazy to do so.


The client will display any mana amount under 100% as 99% (Or less). The client does not round up, so even a 1 mana change will register. Test it yourself if you don't believe me. Up until 1388, even a 1 mana increase in mana pool will drop your mana to 99% for one tick. Above 1388, even a 125 mana item will not budge your mana meter.

Again, you are using a highly inaccurate method to measure this. I'm not saying this _isn't_ happening, but using the client without verifying it VIA SEQ leaves anything you come up with suspect.


This cap, I believe, is a direct ratio of your total int-based mana. Someone told me that SEQ showed a 4164 total mana pool unbuffed, even when the pool should be higher. Since 4164 divided evenly into 3, AND the resulting divisor was '1388' I concluded that not only was the 4164 figure accurate, but the 'cap' was actually a function of your total mana pool, basically, 50% of your mana pool is the cap on +mana items. You say this 'limit' was known for a long time, well, not by the general public. I knew there was a cap on +mana under level 20 or so, but I had never heard of any limitation above lvl 20.

Yes, this was and is known by the general public. Verant stated this a long time ago, publically. Both on their message boards and also on all the news sites. I don't know how much more public you can get than this. It's not my fault you or others forgot this... I've never forgotten it and a lot of others never have forgotten it. It's been something very near and dear to me for quite a while as I tried to balance my character in mana vs hp.


Also, for your comment that stats stop at 252, that too is incorrect. I have seen my mana pool increase going from 253 to 255 int, so you get mana gain all the way up to 255.

Try actually testing these statements before you make them.

Negative. After 252, there is zero mana increase. Unless something has changed since I last tested this, there is zero change in mana past 252 INT. I'm not barring the possibility of change, but I find it highly dubious that they'd have made a change of this nature. Again, using the client to "test" your theory is flawed. Just because the bar on your client tells you something does not necessarily make it reality.

I will admit, however, that the data I have is fairly old. Back when we were trying to incoroporate some mana stuff into SEQ, I ran a set of comprehensive tests, but this was probably more than a year ago. So things could have changed, and I will be MORE than open to data derived using SEQ that is current... I will not, however, be very open to data derived from the client. I'm not saying that the data from the client is WRONG, just that it is suspect.

I'll say it once again, the client "estimates" how much mana you have. The server is the final arbitrator of how much manner you REALLY have. Taking the clients word for what you do or do not have is sheer insanity. Barring sitting down in Verant's offices and going over the code and/or looking at status indicators on the server in real time, SEQ is really the ONLY viable method to test this. Testing with the client is totally and utterly invalid. I've heard nothing that convinces me otherwise and I know enough about the client to know that it's *usually* wrong. You just don't notice it, because the errors are so small, and they don't affect yoru game play. But we are talking about single point accuracy here, and that is something the client is notoriously bad about.

[Neuro MT]

And how is using SEQ any different? By your own admission, trusting the data stream being sent to the client is JUST as bad as trusting the client itself.

The fact is, the client will act with the same level of vagueness in any instance. The fact that a 1 mana increase will register a change in the client AT ANY POINT below 1388, and NO mana increase will register a change above 1388 is IMHO very indicative of a change in data transmission at this point. Occam's Razor suggests that the client in accurate in this assessment, as there is no reason for Verant to push any false info at this point. And if Verant IS pushing false info, SEQ will be just as vulnerable to the false data.

As for stats pushing above 252, I tested it 2 days ago using the same method. A more indirect method is using Stamina. Since STAMINA pushes HP up past 252, it stands to reason Int does as well. Unless you are saying the client is lying about that?

Until you use SEQ to prove that my methodology is flawed, and that there is no cap at 1388, you have no basis to stand on. I have empirical testing, you have supposition. When evidence and theory come into conflict, theory must change.

[fgay trader]

Originally posted by Neuro MT
When evidence and theory come into conflict, theory must change.

What Ratt is trying to tell you that your evidence is most likely flawed, as the EQ Client has been known to calculate things wrong often in the past.

It's true that both EQ and SEQ receive the same data, but SEQ developers can actually tell you how they interpret this data. Verant will never tell you that especially if their interpretation is missleading or bugged. I mean how long have they been denying the existance of "Hell Levels" until they admitted it as their bug in how exp is being calculated?

All Ratt is saying is that you should not take what EQ Client is telling you as the one and absolute truth and that your "basis to stand on" could be based on wrong facts that Verant feeds you.

[Neuro MT]

My retort, however, was that until someone shows me the client is inaccurate, and my findings are in error, I stand by them. You tell me the client COULD be in error. I agree, it could. But my findings fit with accepted tests and with expected results. Therefore, I believe them to be accurate. I believe 1388 to be the correct cap, and since I arrived at this number using only the client, I tend to believe it.

[fryfrog]

inacurate in general? or just inacurate in THIS specific instance?

for inaccurate in general i suppose we could point out that until you actually CAST a spell, the client is only GUESSING at how much mana you have (and is frequently wrong). thats pretty proven and easily observed. ever cast a spell you KNOW only uses about X amount of mana, you have X+Y but when you are done casting X+Y is gone? er, well not sure if that is a good way of explaining.

[throx]

What Neuro is saying is that he has verified the client has a hard coded mana cap at +1388 in mana items. From there you have to assume one of two things:

i) The client code has good reason to arbitrarily limit +mana at 1388 (at Lv60) because that's what the coded limit on the servers is.

ii) For some reason best known to themselves the coders at Verant put a mana cap into the client but a different one on the server. It's well known that the client's mana numbers are inaccurate when presented with mana recharge or drain effects but this is not one of those cases. For it to work this way the code must deliberately cap mana differently on the client and the server.

I find the notion that the static mana caculation formula being identical on client and server to be the most reasonable hypothesis.

[Neuro MT]

The client still resyncs with the server when you perform an action such as clicking jboots or on the tick. If the cap were only client-side, then putting on a large +mana item over the cap would drop to 97% after the tick or after you click jboots. Sadly, this does not happen. Thus, the cap is not only client side, it is server-side as well.

Just because the client makes mistakes doesn't mean you can't factor those mistakes into your method and compensate for them.

[Morannon]

There is a mana cap... though they dont say how much.

http://eq.castersrealm.com/viewarticle.asp?Article=4457

[Cryonic]

http://eq.crgaming.com/viewarticle.asp?Article=4457

So I guess this problem will now be a moot point.

[mbozio]

an evidence the client is not accurate:

You got hit by a 'fake' AE effect that drop ur mana. What happen ? Your client show ur mana dropping down to zero. What can u do in game to get ur mana back since this is a fake debuff ? hit a right click item. What happen at that time ? The client allow you to cast since u don't need mana, the server then send back to the client the accurate and only valide mana left in ur pool. What happen to you client ? U see ur mana go up to what u have really.

That way, we know that the client mana pool and the server mana pool are DIFFERANT instance. But more important, the SERVER pool is the one veran choose to be accurate in all case since it's value will overide the client one.

When you want to test, the only accurate way to do it is to test the SERVER pool. And the only way u can do it is to read the value u get from server. This is what seq does, it show u what numerical value is sent to ur client.

On a side note, casting a spell to try to get any accurate info in game without knowing the real data gave by seq is imo naive since 2 cast of the same spell might result in differant mana cost.

[Roadkill]

"I sometimes wonder if FoH is somehow in cahoots with Verant to spread weird rumors"

From what I've read in a few places, a lot of them ARE Verant. I've come across references to this a few times in the years that I've played this game, in that they have the largest collection of Verant employees as members than any other guild. And I believe it. One, the people working on this game want to see all it has to offer and be on top of their own game... two, just look at their success and some of the things that they just "happen" to know, sometimes even before the content has hit the live servers (Seru Bane weapons being the best example right there).

[RSB]

Status bars in EQ are not be all end all. Sure they run off 2 sets of numbers. 1 is the client side guess and the other is the server shooting down a new value. But even then they act weird. Look at the code that they put in to make things look smooth in the bars. How many times have you bandaged someone and seen their health jump up to 80% then back down to 20%.

I've noticed that a 415 mana spell moves my mana bar less when it is full then when it is empty. I've verified this with screen shot multiple times to make sure that it isn't just the randomness of saving mana while casting.

Since we know the the stream comes from the server and that SEQ and EQ get the same data if SEQ shows 2000 mana and the the mana bar shows 2 bubbles exactly does do we assume that your total mana is 5000? no....why not? Who knows what their formula for displaying the bar is. The only way to tell full mana is sit your ass down, med to full, wait a few more minutes, cast a spell like Jboots or SoTF and see what SEQ spits out. If it shows 4500 then you know that EQ formula shows MANA of 2000 as 40% but 4500 as 100%.

The number SEQ uses IS the raw number. EQ then alters the display of that number which means that it isn't necissarily correct, loss of percission etc.

[Circles]

using the mana bar is flawed. what to test it?

get a caster, go completely oom, and starve yourself. (pretty easy to grab a caster vendor and test this)

get a bard to sing the mana shot song that directly increases your mana, not the songs with spell icons.

cast a spell.

notice it says you dont have enough mana, even though you have a full bar? THATS BECAUSE THE CODING IS CRAP CLIENT SIDE.

Trust people that work daily with the packets, not some graphical representation.

[wrero]

Another way to know the client is flawed without using showeq (I do not use showeq)....

Most casters, especially now with the new ui and the ability to show % mana, have seen that sometimes the UI will show you have enough mana to cast a spell.... You start to cast, the cast progress shows up, and then "not enough mana to cast this spell" and your mana drops a bit. WTF does it say not enough mana and then uses up some of it????? It's not. The client is wrong, the server said "not enough mana" and then sent the correct figure.

I have also noticed, again easier with the new ui, that occasionally when I cast jboots or the Staff of Temp Flux, my mana drops. WTF does jboots or the staff use mana??? Again, just to reinforce what has been said here, it doesn't. The client is wrong, and it received an update from the server with the correct mana.

I understand why this information isn't always being sent.... people have enough lag as it is. I also understand why the server is the final authority on [almost] everything - prevents cheating. What doesn't make a lot of sence is why they can't get the client and the server in better sync so that these errors go away.

It is SOOOO frustrating to know that I need 9% mana to cast a spell.... I stand, I cast, it starts, then my mana drops down to 7% and says "not enough mana to cast this spell"...... OMG frustrating.