Forums → Games → Gemcraft CS stronger Gems with Pool&Blood

Im gonna add poolbound to my gem testing program and see what i get on high levels.

Guys... There may be some problems with my method...





At grade 51 the multiplier of bloodbound is much higher even with a lesser boundlvl.

Gonna try it with gemweaving, the specialability loss may be compensated.

I cannot read the text in those pics very well, but it looks like the triple gem has a 454 bloodbound multiplier, compared to the quad gem's 270 bloodbound and 139 poolbound multipliers.

This is what I was talking about in my first reply to you : eventually the bloodbound takes off, and the presence of poolbound dilutes it.

I'm actually considering starting an endurance run that uses a tiny bit of white in the base killgem, but attempts to squeeze it out (and will, given what I've learned from attempts to almost-but-not-quite squeeze the red out of such gems). I have a suspicion that starting with some white, but squeezing it out as you go along, may be the best way to make use of it in any supergem (short of just not using it at all).

Hmm. Squeezing the white out, so far, has me bombing wave 23 to 1.6e24 HP. The white is *almost* squeezed out completely.

In the 32 gem initial recipe, what I did was replace the Y1+R1+O1/B1+B1+B2 (Orange or Black depending on mana or kill gem) with Y1+R1+O1/B1+W1+B2. Basically, that second Black grade 1, I used White instead. There is probably a better spot to introduce the White, but OMG that is a lot more powerful (especially early on) than leaving that one G1 gem Black instead of White.

Of course, that DOES require squeezing the White out, which can be something of a pain. My method of squeezing out the red was to make a gem with either a O1 or B1 instead of the R1, and using the dupes of that in the 16 or 64 gem upgrade plan, such that only one of those 16 or 64 gems had the Red. I'd then have to upgrade the gem lacking the red, in order to have something to meld into the thing on the next iteration. I'm doing the same thing now with the White - made a second managem and killgem as per psorek's recipe, and I've been farming extra hits (via traps) with dupes of those.

Since I use the 64 gem upgrade plan, rather than the 16 gem upgrade plan, I know that the single G1 White gem will be squeezed out on the next iteration (going from displayed Grade 35 to displayed Grade 40). EDIT - Hmm. Interesting. Although that one G1 RED went away in the 35-40 upgrade in the past, the G1 WHITE did not. I'll have to see when it goes away. I wonder what was different here.

Since I use Mana Lock, I have a feel for about how much mana I normally get in the first couple of Mana Lock cycles. And the Managem was getting me a tremendously increased amount of mana in these first few cycles so far.

Hm, Astro, what does 'sqeezing' do? You will still have a Quad after, no (including the penalties while upgrading further)? So the goal primarily is not to loose 'space' for black, I assume?

I wonder what would turn out as the 'quickest' way:
(a) the way you do it, but probably with a bit more white in the beginning..
(b) or even start with a Quad and switch to Tricolor B at like G45? Well, probably doesn't make sense as it takes only the first 6-7 waves to get there ...
On the other hand I would still be interested in a comparison of Quad with squeezed out W and a Tricolor B at like G70 ... It should be quite a huge difference, I think.

Basically, regardless of the upgrade plan you use, if you use nothing but dupes of the gem you are trying to upgrade, you will maintain the same ratio of gems that make up the gem.

1R+3W+4O makes a decent managem, yes? If you upgrade it with U, or with one of psorek's upgrade plans, you still maintain that same 14 RWO ratio.

Now, if you want to reduce the red (say, you are not having 1,000 monsters over that gem, or in range of the chain hit, at any time) you might make a 4W+4O gem up combine with it. The resulting gem would have 1R+3W+4O+4W+4O .. or instead of 1:7 Red:Everything else, it would be 1:15 Red:Everything else. The total effect of the red would be reduced.

Eventually, depending upon the gem upgrade plan used, if that 1 Red was the ONLY red in the gem, you would squeeze it out completely, and have nothing but an White/Orange dual gem. Obviously, you do not quite want to squeeze the red out completely.

What I am trying to do here is something similar. Except, I *AM* trying to reduce the White completely. You see, the White that starts in the gem is a lot more powerful than the Black. However, after a while, that Black becomes a lot more powerful than any amount of White. At that point, the White is a hindrance, not a help. By attempting to squeeze it out completely, I'm basically trying to turn a quad gem (WBOR and WBYR) into a triple gem (BOR and BYR).

The issue I've encountered though, is an odd one. When I tried to reduce as much as possible, without eliminating, the red in a triple gem, I found that that 1R would last through to grade 35. During the 64 gem upgrade plan, it was gone during one of the later intermediary combines, and so was not there anymore when the gem was at grade 40. The 1W I introduced however (by changing one 1B into a 1W), has not faded away yet, though currently the gems are only grade 45 (40 for my killgem, did not quite have the mana to bring that to grade 45).

Currently, at grade 40, my killgem (with a Hit level of 27 and a MP level of 47) has a crit multiplier of x6,896,714,344.64. My triple gem (lacking the 1W) that I've been using for the other 63 gems in the 64 gem upgrade plan, is at (hit level 11) x5,036,791,330.08 crit multiplier. A dupe of the killgem (so, its hit level is 1) only has a crit multiplier of x469,041,500.56. The ONLY thing keeping the killgem better than the triple I'm diluting its W with is the hits I've farmed, and the hit level.

I basically have two questions right now.
1 : Why did I get the Red completely out at Grade 40 but the White still remains at Grade 45? When will the White be squeezed out completely, leaving me with a Triple gem?
2 : Is there a better spot to put that Grade 1 White gem in, instead of that particular spot I put it in in the initial 32 gem recipes?

Hm, I am familiar with the concept of 'squeezing', as I did it myself. My main question was: Once the color is squeezed out, you still have a Quad, no? At least the game says so for the NAME of the gem ... You suggest, that once a color is squeezed out, the game treats this gem as a Tricolor? That would imply that at the point were a color disappears the reamaining specials should increase drastically ....

What you observed for R and W is strange. Maybe that is different because you were dealing with a Tricolor while now you are dealing with a Quad? So it would not be about the absolute percentage of a color but rather a relation ... ?

I rather expected it to be the same for the W as it was for the R, simply due to the 1:everything else ratio getting so dang BIG.

The only thing I can figure is that the W is tied to my MP level, while the R has no such boost. I am only guessing, but perhaps the color disappears when its power is reduced to near zero by the squeezing? I do not know. I'm going to try to see when (if) the W gets squeezed out completely. If it does not, though, then I'm going to have to farm a bunch of hits on the triple gem dupes in order to make the triple a stronger gem than the quad, by hit level. At an equal hit level, it would already be stronger (as evidenced by the crit multipliers in my previous post).

This whole run is an experiment, really. So far though, I'm rather disappointed. I had hoped the W would have been squeezed out completely by now, giving me the starting advantage of the W (and you should compare the killgems, psorek's 32 gem and the mod I made of it) but the ending advantage of &quoture" B.

Finally! Though I should have made a dupe of it or something for comparison purposes ... at Grade 50 the W went away. Unfortunately I did not write down the Grade 45 mana leech or anything else, but I will record such things for comparison purposes with the kill gem next mana lock cycle.

I would expect though, that a second W in the initial gem would make it even harder to remove (higher than Grade 50, aka 60e).

Bah .. too long since that last post to edit this in.

Grade 45 (after being moved to gem box)
418,346,870,279,626-7,414,412,233,730,244 damage
x97,059,322,252.32 crit damage
1,860.8 chain
x1,028.68 bloodbound (HL 24)
x15.71 poolbound (ML 51)

Now, I WAS farming hits, but I'm going to avoid using those gems (I'll use them in the triple's upgrade, and merge it next iteration).

Grade 50 (still in gem box)
9,200,373,158,334,008-200,315,953,876,417,920 damage
x1,304,049,231,706.35 crit damage
2,987.08 chain
x1,829.11 bloodbound (HL 24)

Yeah, squeezing the W out worked well. Or did it? Fortunately, the Triple I'd been feeding it to squeeze the W out got exactly the same HL when I combined those gems I'd been farming hits on ...

Grade 50 (in gem box)
10,782,400,780,929,254-232,732,343,535,028,540 damage
x1,915,467,208,119.84 crit damage
3,471.91 chain
x2,248.26 bloodbound (HL 24)

Looking at the damage amounts of the two gems, the Quad turned into a Triple, and the Triple I was feeding into it, the loss of that 1B hurt me a LOT.

If I was to try this again, I would look at replacing a Y/O rather than the B in the base gem. The initial gain was nice, but I'm losing a lot of my total damage here. 261,221,865,750,997,321,047,370,571,520 damage from the killgem vs 445,791,172,310,033,116,594,784,716,260 ... that is roughly 50% of the damage rating on the gem lost! Now I REALLY wish I'd looked at the before and after's on the managem. The crit special is almost 50% larger on the triple than on the quad-turned-triple, I would hazard the guess that the same is probably true for the managem, if I had compared those.

Ok, just noticed based on cost that somehow, I'd not gotten five of the gems in the control upgrade.

A HL0 dupe of the current killgem (the one that was Quad but trned Triple when I squeezed the W out) is :
Grade 50
388,169,271,925,430-8,451,450,461,096,004 damage
x55,018,620,671.56 crit damage
331 chain hit
x77.17 bloodbound
damage rating : 464,987,147,038,789,086,989,898,684 (4.6e26)

A fresh control (triple gem only)
490,846,252,277,475-10,546,715,297,151,698 damage
x85,743,800,437.84 crit damage
388.95 chain
x96.21 bloodbound
damage rating : 904,315,451,704,830,347,827,692,026 (9.0e26)

As you can see, its still a loss of about half the damage rating on this killgem.

And for what its worth, I just Dupe'd the manafarm gem. Sitting in the gembox, it is :
Grade 50
372,556,779,328,954-6,935,866,640,578,032 damage
320.49 chain
95,266,640,699.26 mana/hit
x73.43 bloodbound (HL1)

A triple gem for comparison purposes:
Grade 50
462,879,895,478,108-8,653,456,520,250,462 damage
375.85 chain
148,448,709,803.7 mana/hit
x91.54 bloodbound

The managem and killgem are still classified as Quad Gems. That means that, despite losing the W as a modifier, they are still subject to the same mods that Quad gems are : +220% damage, 120% specials power. Triple gems, on the other hand, are +220% damage, 150% specials power.

Doing the math on the Bloodbound special, that is almost exactly the difference between the two gems. Factoring in the bloodbound multiplier, that is almost exactly the difference between the two gems' mana leech as well.

Anyone think that W is still a viable thing to use in their manafarming gems? Because I don't. Sure, it started off real strong compared to nothing but B .. but even after it got stripped from the effects list, it still hampers the gem by quite a lot. It may be worthwhile starting off with the lone W, but if you do, around grade 30, replace that gem with one that never had the W to begin with. Start farming the hits so your Hit Level is good enough that the replacement gem is an automatic upgrade.

Hm, I still don't completely understand the goal of your experiment:

It primarily shows what could be expected? A Quad stays a Quad, even if one color has been squeezed out. So its not a big surprise that the specials are at about 50% of a Tricolor (at G50, at higher grades the difference should be even bigger)?

For the 'sqeezing' I found out how it works:
If the base value of a special drops below 1, the corresponding color is squeezed out. For poolbound then it is: If PoolBound_Multiplier - (ManapoolLevel * gainPerManapoolLevel) < 1 ...

Edit: hadn't read your last post. No I never relly thought W could make sense there. I just thought It might make sense to start with ROBW and switch to ROB at about G45. But that doesn't make sense either as reaching G45 doesn't take too long anyway...

Basically .. I sought to validate, or invalidate, what others were describing with theory, by going and actually doing it. I was initially amazed at how much stronger that 1W made the starting gems.

But in the end, that 1W made the gems a lot weaker than they would have been without it, even with the W squeezed out. Confirmation of what we expected, which is a good thing.

Also, now Saiyaku (or however it is spelled, the thread creator) knows what's what with adding W to to a B-inspired gem.

As for the squeezing thing .. I'd noticed that the overall value of the PB multiplier going up. This was due to the MP level going up. I'd have a gem with, say, 10.06 multiplier. I'd do the 64 gem upgrade, and have the 5 grades higher resulting gem have a 10.04 PB multiplier. A manalock cycle or two later (when I went to upgrade it) due to the MP level going up, it was no longer 10.04; instead, it was at 11.67 or something. Each gem I upgraded had a higher PB multiplier than the resulting gem of the previous upgrade iteration, again due to MP level going up. Each gem upgrade iteration brought the MP multiplier down by 0.02 from wherever it was before starting the upgrade.

The long and short of it is I was expecting the PB multiplier to go from 15.57 down to 15.55 when I went from grade 45 to grade 50 gem, I did not expect it to go away entirely at that point. I expected it to take a lot longer because I suspected the total multiplier is what was looked at, and I figured that it would start to stay shrinking when the MP stopped increasing in level so often.

NO I think this thread is not finished yet >=(

While you tryin to squeeze the **** out of white =D and use the 64 method, i did a gemweaving run.(duplicate and combine with 2 gems of a specific color and same grade to purify the gem)
Tricolor: Manaleech&bloodbound
Quadcolor: Pool&blood

These are my results so far:

Grade 40 Tricolor in Trap and Amp(G40 Pure orange):

G40 Quad in Trap and Amp(G40 Pure orange):

G50 Tri no Trap no Amp:

G50 Quad no Trap no Amp:

G50 Tri in Trap with Amp(G50 Pure orange):

G50Quad in Trap with Amp(G50 Pure orange):

G54 Tri no Trap no Amp:

G54 Quad no Trap no Amp:

G54 Tri in Trap with Amp(G54 Pure orange):

G54 Quad in Trap with Amp(G54 Pure orange):

I was not intending to eliminate poolbound in firstplace.
The Gems are sourronded by 5 not 6 Amps. Kinda missed that due the map... The results may differ on a 6 Amp comparison.

So my theory complains that
1. Pool and Blood multiplier, which are added before multiplication, could be useful as they CAN compensate some loss of the quad modifier.
2. the special ability loss of Quad CAN also be compensated by Ampbonus as their growth is not effected by the Quadmodifier and is added before multiplication too

My testrun confirms the CAN but it seems like even with gemweaving method the tricolor could catch up to Quadcolor at grade 56 or 58 with the boundmultiplier.

If u compare the 54 Grades without trap without amp u can see the difference of the manaleech which is absolutly laughable of the Quad but with the ampbonus it is comparable to the Tricolor. (May differ with a 6. Amp-.-... thats my fault.)

But still the added boundmultplier of the Quad is higher then the Tricolor yet.

I m not into the 64 upg plan yet but i may take a look the next days. Their may some other methods to exploit the Quad even more.

What kinda bothers me now is that the Quad seems to have a higher rate of fire... I dunno if i did some mistakes during gemweaving but i m quite sure i did not...

Gonna take a closer look next time.

This method may not be the best for endgame run with G70 or G80 or higher, but for our newcomers it could be.

So pls be kind to me even if i annoy you with this post a lot >.< .

I don'tunderstand that you're so desperately trying to prove something here.

Comparing these two gems just doesn't make sense. You could have upgraded the TRI with B only instead of O, then its multiplier would have easily surpassed the QUAD.

In your case the multiplier of TRI is a little lower, but it leeches more mana.
Why do you start talking about the multiplier ('could catch up at G58 ...' then after having chosen a way of upgrading that negatively affects this multiplier?

You are adjusting the starting conditions and the conclusions of your experiments in a way, so that what you want to be true can be proven. Yeah, that's a little bit annoying my intelligence.