Paralyze Testing

[Belkin]

So since everybody has been asking for Paralyze test results and I've been curious myself (For merit distribution in light of the Slow II tests).

I have preformed extremely simple tests in a short amount of time to get a general idea of what the "soft cap" is.

Tests were done in East Ronfaure [S] on Too Weak Ladybugs by counting the amount of times a Ladybug was paralyzed out of the total amount of times it attacked.

No MND gear and full MND gear yielded the same result (As expected from a Too Weak mob).

10 DIFFERENT (Killed them after the first Paralyze II wore off to avoid building resistance) Ladybugs had Paralyze II cast on them. The proc rate was as follows:

Test 1-5 were done with no gear at all
Test 6-10 were done with full MND gear

Test 1: 39%
Test 2: 31%
Test 3: 18%
Test 4: 33%
Test 5: 34%
Test 6: 36%
Test 7: 40%
Test 8: 23%
Test 9: 28%
Test 10: 37%

This leads me to believe the proc rate "cap" is around 40%. Obviously you can get some freak of nature tests and get a 50%-60% proc rate but this is highly unlikely.

This got me thinking about what the proc rate for Paralyze I was. So I tested with the same method as above.

Only 5 Tests were done (Mainly because I don't plan on using it and I didn't care that much)

Test 1 & 2 were done with no gear
Test 3-5 were done with full MND gear

Test 1: 17%
Test 2: 18%
Test 3: 17%
Test 4: 19%
Test 5: 19%

Probably get a capped proc rate of 20% with Paralyze I (Again, expect some freaks of nature due to the random nature of Paralyze in general).

I will test later on the highest level Aerns in sea in a couple hours to see what kind of results I can get on tougher mobs.

If anybody can suggest a better testing method, I'm all for it. But I hope this gives some RDMs some peace of mind knowing that Paralyze II is about 15-20% better than Paralyze I.

[Pindle]

What about people with different amounts of Paralyze 2 merited?

[Belkin]

All tests were done with 1 merit in Paralyze II. If it works the same way Slow II does, it shouldn't make a difference as long as you have capped MND and don't get resisted.

[Callisto]

Nice, so I don't feel so bad about my debate on bumping it up 2 more slots, I think I've .decided to drop Phalanx 2 and go 4 Slow/4 Para/2 Blind, but good to see some evidence that Para II is actually worth the merits/MP cost.

What about people with different amounts of Paralyze 2 merited?

Baseless speculation, but I would assume it's similar to Slow 2 in that it caps the same but requires less dMND to hit said cap.

[Snapples]

...what.... 1 merit into the now 5 times meritable spell and only 10 castings on TW mobs doesnt prove jack shit

[Callisto]

...what.... 1 merit into the now 5 times meritable spell and only 10 castings on TW mobs doesnt prove jack shit

Sorry, missed the numbers you posted from your tests.

At any rate, I have level 2 at the moment, if I can get time after Nyzul tonight I'll do something similar and see if my results are about the same.

[Belkin]

Actually, casting on Too Weak mobs is the most reliable way to find the cap of a spell, if you think about it. They never resist and you always get full potency. So I'd say I found some quality information really.

It's not like I am going to find some radical new information if I tested 50 more times on Too Weak mobs. I was getting full potency every time.

[Nekio]

Sorry, missed the numbers you posted from your tests.

He's right. The numbers are statistically useless. How many attack rounds went off for each casting? PROTIP: A percentage (per CENT = per 100) value calculated from a sample less than one hundred is beyond meaningless; an extrapolation unto itself. This point comes up all the time for a reason... if the stats don't show a significant difference, a difference does not exist for all practical purposes, regardless of any trend you want to invent.

If you do the stats, the chances that these test results even show a difference between Para I and Para II is only about 50/50.

[Suiram]

If you still have the information (and if not, can you record in future trials) in addition to the proc rate, the number of times paralyze procced before wearing, and the duration in seconds that it lasted.

[Belkin]

Did not record the durations. The average number of times they hit before it wore was 28.

[Belkin]

Sorry, missed the numbers you posted from your tests.

He's right. The numbers are statistically useless. How many attack rounds went off for each casting? PROTIP: A percentage (per CENT = per 100) value calculated from a sample less than one hundred is beyond meaningless; an extrapolation unto itself.

So what you are saying is that the Paralyze has to last for at least 100 hits? If so, it will be EXTREMELY hard to test, seeing as the longest my Paralyze lasted was 35 hits.

[Nekio]

No, I'm saying that the test doesn't mean or tell us anything useful yet. A better test would be on several mobs of the same level, resulting in data that can be pooled.

[Belkin]

While I will give you that other mobs could be more or less resistant and that further tests still need to be done (As I mentioned I would be doing in the OP), I still think you can't argue the few numbers that were given. Paralyze II proc'd an average of 32% of the time it was on the mob. This is a good amount better than the Paralyze I results, which yielded 18% average.

Of course in order to slim these numbers down to a more exact figure, endless tests are required.

But what I set out to figure out I have accomplished with just these tests. You can see Paralyze II is better than Paralyze I by a good margin.

So maybe we still have no clue about exact numbers, but we do know how much better Paralyze II is than Paralyze I.

[Kirschy]

I had actualli planned a similar test. While I think it's hard to get exact data in this way, I think it's a step toward a closer in understanding how paralyze works. Paralyze appears to have some randomness in duration, and it won't ever last 100 hits. But taking multiple tests and averaging them like you did is the best we can do.

I'm very interested in seeing more tests if you plan on performing them. Crabs in particular are very nice for testing as they have MND in the mid-60s, attack fast, and they're easy to tank.

If it helps. You can use Banish to the test the MND of your target.

Damage = 14 + dMND

If you're rdm/whm, and not using any MAB or staff, adjust your MND until you hit 14 damage /w Banish. That will be dMND = 0. Adding 1 MND should raise the damage by 1 point each time. (When dMND is lower than 0, you need about 1-3 MND to raise the banish damage by 1.)

Just want to say thanks for the testing, and I hope to see more results if you continue with this~

[Belkin]

I definitely plan on continuing, as I know the results are extremely rough right now. I've got some LS events to do before I can continue testing though. And thanks for the hint about the crabs and Banish. I'll definitely look into them as test subjects.

[Snapples]

any word on slow2 and mnd at lvl5? if it is still capped the same as 3 then it just makes it a lot easier to hit the max slow without gear swapping. i need to know!

[Callisto]

If it helps. You can use Banish to the test the MND of your target.

Damage = 14 + dMND

Awesome, getting a MND value was a huge question I had, thanks.

[Sargas]

If it helps. You can use Banish to the test the MND of your target.

Damage = 14 + dMND

If you're rdm/whm, and not using any MAB or staff, adjust your MND until you hit 14 damage /w Banish. That will be dMND = 0. Adding 1 MND should raise the damage by 1 point each time. (When dMND is lower than 0, you need about 1-3 MND to raise the banish damage by 1.)

Wouldn't a RDM need to take their MAB traits into account when doing this? Or is that included here?

[terranova]

Time for me to retire Paralyze I.

[shirak]

All tests were done with 1 merit in Paralyze II. If it works the same way Slow II does, it shouldn't make a difference as long as you have capped MND and don't get resisted.

Except each new merit gives you "increases effect by 1"

Probably thats a straight 1% to cap giving a 45% proc rate, which is substantial. Also that +10macc will give more consistent results towards that cap.