So its either one or the other. You either consume like Ghandi or Garfield, nothing in between. My server, set up a few days ago using Akkas installer, is the latter, and it happens on both UF and ROF2.

Thats confusing, because I tried it with the binaries from the update script, which I assume is the same binaries that one
gets with the installer. So I have no idea why some (using the installer) would experience that and someone like myself
does not, when using, what should be the same code ?

Again, this is a known issue, so it doesn't do any good to ponder why it seems normal for some and not for others.

When we have time to get granular to fix it, we will.

demonstarr has been dissecting the client when he has time but hasn't felt like finishing it off yet because he's had other stuff going on he cares more about

https://github.com/EQEmu/Server/issues/308

I can gather no more details from the client :P I have to go find time to sit down and watch packets on live, which is a bummer, but I will get around to it.

Aye but thats the thing Akka (BTW, much thanks for your installer, I wouldnt have gotten a server up for me and my buds otherwise)

This doesnt seem to me to be a client side issue, as both versions of the client have the issue (anyone still using titanium chime in for confirmation here?). Now I've no working knowledge of the packet system so someone smack me if I've got this wrong but, it seems to me its the server that decides how often an item is consumed and then updates the client accordingly, telling it "that rat sandwich just got eaten, remove it from inventory"

If its a case of the hunger value being out of range I've seen mentioned in another thread, it could throw the calculations way off for the delay between eating. Being more full than possible could cause the delay to go negative, capped to a minimum, making someone eat like Gabriel Iglesias. On the other hand, being hungrier than possible could calculate an eating delay so great that it appears indefinite. People are reporting both. Has anyone running these tests tried stuffing their characters face with a stack of rations before letting it run? Does sitting idle or active playing make a difference? Dying for instance, does it properly reset hunger? What about when the toon is handed off to another zone server? If I'm even close on this, fixing it could be as simple as a basic sanity check on the variable.

If the code and binaries are identical between servers, then you gotta consider the hardware. How do multicores and threading affect the routines that figure timing on consumption?

You're not reading me correctly.

We have ways of looking at client pseudo code to get an idea of what client food consumption rates should look like. Sometimes we have to go on Live and do packet captures to get an idea of how things work correctly.

I'm far aware it's a server issue, I posted the issue 2 years ago.

Again, we'll fix it when someone feels like fixing it, it's not a super complex issue

I have to figure out decay rates basically. I know when the client auto-eats and how often it auto-eats, but it depends on decay rates which are server controlled from what I can tell. Which means I have sit down and record hunger values and when they change and how often they change. I've been working on other projects and playing :P

There was a rule to disable feeding completely, right? Make your char always full?

PS
I was never fan of how SOE implemented the food - its was unneeded burden on a bag slot ;)

Ah, there was my confusion. Now its actually moreso. Why would they let the client decide when to eat? That seems an easy target for an anti-hunger hack on live.

The server still knows your hunger and thirst levels.

So is there a way to disable eating of food altogether? Someone else asked this but there was no answer. So new servers are just dealing with this and people are consuming food super fast? I just used Akkas installer on my laptop and it does the same thing as reported here. ALso, I used the same installer about a year ago (after this was reported in 2014 by Akka) and it didn't have this problem.

[EDIT: Nevermind I see it was caused by getting rid of blob in the DB] Also, in the meantime while it's being worked on, what revision of the server should we compile to avoid the issue?

ok so some definined variable increases as you eat.. larger it is, longer to wait before hungry, right?

there is a server rule that increases how much eating/drinking adds.. think i set mine to "5000" and i'm not sure if my toon ate a second piece of food in hours of playing. i was concerend about overflow so i didn't pick a ginormous number. have to look into how it's used in the code, how it's defined etc

at the very least you should be able to slow it down to the point almost never occurs.. concerns about stat food? add it to a vendor or something and cheap.. then no biggie if it's eaten.

no longer a wasted inventory slot, too :p

Yeah that's the thing... those settings are not changing it, and this is already a known problem but it doesn't effect all server setups or clients for some reason. So I was more looking for a setting that turns eating off altogether. I don't think messing with the code before compiling will be good because there's other things that mess with eating.

This is happening to my Human and Iksar characters across all kinds of servers here, legit and custom. Emu servers are just unplayable now for me, and started in the recent past.

Using RoF1 and RoF2 clients and standard MQ2 binaries (non cheat)