http://www.filedropper.com/atbalancesheetYou may find the spread sheet with my tested info at that link.
In order to properly inform a change and re-balancing of AT, I have (with the help of a few people) tested out AT in order to see just how much exp/hour can be obtained by each race and in order to compare each level. I will use a standard research format for analyzing the data and will provide the data set as well. Keep in mind this is a basic data set, I could be much more thorough but it would require access to tools I do not have (ie. God).
Methods
3 runs of 4 rooms per rank with an AH earl with maxed 241-251 mod c2, stun, b2. For each run, time, respawn times, exp, luck and other variables. The process involved some statistical tools and calculations. I used a basic excel spreadsheet with some standard calculations and statistical knowledge. For kill speed of the ltgen, gen and fdm area I asked for Supaguard, Nyandere, Ayumi, Botar and Cosruni to help out. This provided data for all current races.
And important thing to note is that all these characters have all relevant skills maxed out and have used a variety of stat boosters such as gpots and dpots. These stats do not reflect the situation of character which are still ranking and may in fact falsify some statistics. For example, the only reason why an AH is able to clear gen/fdm so fast is simply because of a stun c2 b2 stun b2 b2 rotation which cannot be achieved unless you have maxed out stun. Mods lower than 230 will not allow for the second b2 in rotation and add another amount of time to the clear rate. It is important to note that these clear time are by players who have mastered their class and their strengths. Given this caveat, exp/hour rates are inadvertently inflated and would not properly reflect actual gains.
Kill exp was added and corrected in order to be closer to reality but it may be on the low side to be safe. Avg skels per room was used in order to approximate the kill exp per room clear but the skel amount may vary by either room or even by play style. Furthermore, another corrective tool was added which accounted for walking, spelling, click delay, etc. This involved a overall 15% penalty to gains to account for moments of idleness, server backups, spelling self, moving, messaging and all other activities that are not pure killing or grabbing scrolls. It is likely that this penalty should be higher due to randomness or spawns and due to a voluntary calculation that spawns occur every 6minutes for areas under general which is untrue. The spawns occur each 6.5minutes with means that one spawn should be deducted additionally from the count and would reduce rates but I am still unsure if this should be used in the calculations.
Data
Looking at the progression of stats we can see that there is a larger increase in mods in both cols, brig gen and fdm. All others seem to have a pretty linear increase in mods. Beyond that data shows that there is a great variance between minimum and maximum obtained values from scrolls. Depending on the rank, the max value scrolls can vary from ~250k per hour to 600k per hour. Given rng and statistical principles, these fringe rates are incredibly rare but this gives you an idea of how possibly unreliable town can be. Theoretically you could get 34k from scrolls in majors in an hour or get 279k as well.
One other interesting piece of information is the general stats that the skeletons have. It is very interesting to see they have immunity, concentrate and meditate. All 3 stats are generally along the same base/mods as the other general stats and may pose a problem especially around later ranks where melee classes may have limited to no capacity to stun/warstun skeletons which then reduces the kill speed significantly.
Another interesting piece of data is the kill speed of optimal to near optimal builds. It was shown that Seyans earn between 91 and 49% of optimal exp rates, warriors earn between 100 and 38% of optimal exp rates and arch-templars earn between 51 and 20% due to time to kill (TTK). From the data AH, Sorc and Necros can potentially kill within scroll spawn time and have full clears. However, the 3 test players were very skilled in the use of their race and may have used optimal techniques which are not necessarily conventional knowledge. This in turn may have allowed them to reach full clears.
Analysis
The immediate issue that arose in this dataset is the significant decline in exp rates for melee classes such as Seyan, Warrior and ATs. Their TTKs varied between 2m40s – 5m49s for gen and 4m4s – 9m54s for FDM. The TTK for ltgen was similar in all races except for sorc and AT. The sorc speed was related to poor teleport placement which in conjunction with stealth reduced kill speed in test run. ATs take nearly 3 times longer to clear a room in ltgen, between 2 and 6 times longer in gen and between 2 and 10 times longer in FDM. Arch Templars appear to make between 51% and 20% of the optimal exp rate in the last 3 areas of the town. Both seyans and warriors suffer from a penalty to their exp rates yet not as pronounced as for Templars. In no way are these proportions applicable to pents which is the other exp gaining method available at that rank.
Using Time corrected exp rates which include corrected kill exp, ATs make an hourly rate of 90k in fdms which is equivalent to one solve in uw barons (an area they can solo on dpot like the one used for this test). An efficient Templar may reach rates of around 600k per hour in barons and probably higher with spell sets. Warriors and seyans may also reach this exp rate and most likely surpass it. Comparing these rates to town, seyans make at least 3 times more exp in pents at that rank, warriors make 3.5 times that amount and Templars make 6times the amount. Spell casters and necros may make anywhere from 1.5 to 3 times the amount in pents. In all likeliness, both general and fdm areas are in no way competitive forms of exp. To the point where ltgens provide more exp for melees with less difficulty. Despite a clear advantage for spell casters and necros, the town provides much less exp than pents can provide at a much earlier point.
The following issues appear evident.
1.
Exp rates are disproportionate
2.
TTK is extremely disproportionate
3.
Skeletons and scroll skeletons have different spawn rates
4.
General and Fdm have double the spawn rate for scroll skeletons as opposed to other areas
5.
Skeletons have high mods of meditate, concentrate and immunity which effectively reduced TTK especially for races that rely on melee without being able to use c2/wc perpetually.
6.
The degree of variance in scrolls may discourage players as the immediate reward is not high enough.
7.
Exp rates seem to stagnate at around 250-300k for scrolls from brig gen to fdm. While potentials seems highest, even theoretical averages does not seem to differ much.
8.
On a basic level, warriors, seyans and Templars lack the capacity to effectively clear a room of healing npcs by design. Wc requires endurance which would then require greps in order to sustain making it near impossible to sustain. Warriors may not have raised their spells or ss gear to adequately dispatch multiple rooms of high immunity, healing npcs. Seyans have the most capacity of the 3 but still lack in overall kill speed due to the mana cost of both b2, stun and c2.
9.
The door areas are not properly marked as non npc and may result in a single skeleton block the entry which greatly raises the TTK especially for melees.
10.
Entry hallway skels make the portal exp scroll irrelevant as walking back will most likely cost you a much more lucrative scrolls. Therefore players will remain in the areas walking back towards first room.
11.
Single spawn points may in fact limit spawns which then negatively impact the speed of exp gain.
12.
Layout for certain rooms are not beneficial for AOE tactics.
13.
Lower ranks have significantly high degrees of curse which may limit lesser ranked players from participating.
Proposed solutions
1.
Exp rates are disproportionate
Reduce variance which makes scrolls easier to predict which in turns allows for easier and more efficient fine tuning of exp rates.
2.
TTK is extremely disproportionate
Find ways to increase melee TTK without affecting spellcaster TTK. Such ways may include reducing hp of skeletons to above blast2 threshold. Removing concentrate which limits total hp per room to more manageable levels. You will find a table of b2c2 thresholds and suggested hp ranges which allow for lower levels of hp and therefore potentially higher TTK while not affecting spellcaster TTK since they cannot 1hit blast. Keep in mind 251 mod is around the highest achievable mod without SS gear or prison gear.
3.
Skeletons and scroll skeletons have different spawn rates
This has a potential of desyncing spawns which overall reduces exp rates.
4.
General and Fdm have double the spawn rate for scroll skeletons as opposed to other areas
While this appears to be a feature that serves to facilitate melee race’s experience, it is shown that warriors and Templars will have to fight with respawns which makes it very difficult for them to even get the scroll in the first place. Have more time on scroll respawn only allows a second set of mobs to appear. The only thing it does is reduce the exp rate of AH, sorc and Necro. In both AH and sorc cases, it is pretty tedious and extremely challenging to reproduce top killing speed without specific knowledge of appropriate stealth and stunning strategies. Necros seem to have it good as well but test sample explained having received a large amount of crits which may impact speed. Overall, necros would appear to fall behind if spawn rates were similar to lower areas allowing only sorc and hara to gain a potential 340k boost which still puts them in about 30% less exp gains than pents with significantly more tedious gameplay.
5.
Skeletons have high mods of meditate, concentrate and immunity which effectively reduced TTK especially for races that rely on melee without being able to use c2/wc perpetually.
See #2
6.
The degree of variance in scrolls may discourage players as the immediate reward is not high enough.
It appears that players have expressed a general dissatisfaction with low runs which does not push them to return. Gains are indeed minimal in low runs and even avg/the. Avgs are somewhat on the lower side depending on race vs pent gains.
7.
Exp rates seem to stagnate at around 250-300k for scrolls from brig gen to fdm. While potentials seems highest, even theoretical averages does not seem to differ much.
Difficulty continues to raise but the exp rates seem to remain within those bounds when taking only scrolls in consideration. Given that most races cannot complete the highest rank at full clear speeds until at least noble, it makes it very useless. Most players report staying in magen/ltgen and others have expressed a desire to have gen/fdm removed due to inefficiency. All in all, there is no added value to current gen/fdm areas and they are significantly less efficient than ltgen or pents. Overall, unused real estate.
8.
On a basic level, warriors, seyans and Templars lack the capacity to effectively clear a room of healing npcs by design. Wc requires endurance which would then require greps in order to sustain making it near impossible to sustain. Warriors may not have raised their spells or ss gear to adequately dispatch multiple rooms of high immunity, healing npcs. Seyans have the most capacity of the 3 but still lack in overall kill speed due to the mana cost of both b2, stun and c2.
Related to TTK issue. Could increase potency of melee aoe attacks or simply reduce stats of monsters being killed.
9.
The door areas are not properly marked as non npc and may result in a single skeleton block the entry which greatly raises the TTK especially for melees.
This is a build issue and could be easily resolved as long as all doors have at least a 3x5 area around door tile in order to prevent single target forced melee which increases TTK significantly for melees.
10.
Entry hallway skels make the portal exp scroll irrelevant as walking back will most likely cost you a much more lucrative scrolls. Therefore players will remain in the areas walking back towards first room.
Remove hallway skels, put signs and let newbies die if they get to that point. Worst case put teleport point inside the first large room and place skeletons in hallway. So that people do not need to fight through the corridor everytime if they want to use the teleporter. Also increase value of exp scrolls from teleporter.
11.
Single spawn points may in fact limit spawns which then negatively impact the speed of exp gain.
More spawn points?
12.
Layout for certain rooms are not beneficial for AOE tactics.
Tables and such but that is a build issue.
13.
Lower ranks have significantly high degrees of curse which may limit lesser ranked players from participating.
Simple balancing issue. Lower ranks tend to have lower mod res/imm which blocks access to area. Maybe lower curse on major/ltcol rank.
My final suggestion is to simply make random scroll drops the way to go. This will still be random and will still be rng based but would actually make the area more random and potentially fix some issues. Make one scroll drop guaranteed and the rest rng based. Could easily raise exp rates and you could justify lowering max exp from scroll by a bit.
I strongly believe that when areas are not used, that they are not very well done. Why does no one pent in earl icies? Because exp/difficulty/speed are not balanced, much easier and faster to pent uw barons. Why does no one use general/fdm areas? Because they are vastly ineffective when you add difficulty. If they are unused, changes should be made or area should be put to better use in my opinion. I think that while town should not be a 100% pent alternative, it should at least account for 75-80% of rates. If it doesn’t what is the point? Providing an AFK leveling experience? Rules are actually against afk leveling which is basically AT is now. Drop in there check back in 3 minutes, rinse, repeat. I don’t think that’s the envisioned function of the area.
Reduce TTK, improve rewards and balance stats to allow for faster melee TTK. This will greatly change the discrepancy and provide more space for players to play instead of simple maintaining a subpar function which promotes afk playing.
Thank you for your patience in reading this post and checking out my table.
Fri Jan 15, 2016 9:33 pm
(I have to be purple because of champ quest