L F' U2 B' //EO 4/4
R , U2 F2 D' {R F2 D2 L} //2c3 2e 8/12
(D' R2 B2 D'^ B2 D B2 L2) //almost a double slice but has parity 8-1/19
Luckily we know we can fix party issue by doing either 
{} = R B2 U2 R or {} = L U2 F2 L
the latter gives 2e2e+slice directly so we do that.
And then 2e2e solves nicely with
(L2 B2 L2 B2 L2 ^^ B2) //slice 6-3 2/22
^=w
^^=Uw' U //+2/24

Almost got bailed out from not finding any good DRs (this and 4c4 in 10 were the best ones) but unlucky slices give 24 in the end. Kind of deserved tbh...

Horrible that I had to use 12 move DR but at least it had 4 variations so it was still propably going to give 23-24. It is 23 opt but not trivial since there were variationsit needs 1 move longer LS that I had and it happens to be +0.

L F2 R //EO 3/3
D' B , D2 F U' F' //2c3 4e 6/9
B2 U^ (D^ B2 U2 F2 D') //HTR +2 ext 7/16
L2 B2 L2 D2 B2 U2 //slice 6-1/21
^=w //+0/21

Had several 4c3 in 10 DRs but used most of my time on this one so I tried all 1-2 move reductions to 4b2.

B' D' F' L' //EO 4/4
B' , R2 D' F' U' F' //4c3 8e 6/10 
(F2 U) U2 F2 U L2 F2 U //HTR directly to pure blockyfish :D 8/18
B2 D2 L2 //solved 3/21

4c3 8e solved in 11 moves :D

There were some annoying jzps I found that turned out to be not so good so I wasted a lot of time on those. At least:
U B (D' F) , R2 D //jzp (and 4c4e on one side) in 6
R2 U(or U') B' D F //jzp (and 4c4e on one side) in 5

D B2 F' U //jzp 4/4
B2 D2 F2 R' B2 L //2c4 +1 extension 6/10
U2 F R2 F'^ D2 F^ D2 F //HTR 8/18
L2 D2 L2 B2 //slice 4/22
^=w //+0

Had an early backup 23:
D B2 F' U' //jzp 4/4
R2 B2 L' F2 R //2c4 5/9
(F) U2 F R2 D2 F U2 F //HTR 8/17
U2 F2 D2 L2 U2 R2 //solved 6/23
I also had the L2 F2 variation but missed the B2 D2 variation that gives a reasonably easy 22.
Luckily the extension saved the 22 anyway.

(L D2 B' D) //EO+rzp 4c4e 4/22
U2 D2 B2 R B' //4a1 4e 5/9
D2 F2 D2 {F2 L'} //HTR 5/14
F2 [L2 F2 D2] //double slice 4-1/17
[]= U' D F2 R2 U D //slice +2/19
{}= R2 L2 B2 L R2 //+3/22

I also tried solving wrong slice first. You can get the normal slice in +1 but the wrong slice becomes horrible +4 :D
D2 F2 D2 F2 [L' //HTR 5/14
F2 L2] F2 D2 //double slice 4-1/17
[]=L R2 B2 R2 //wrong slice +1/18 but it really sucks

Double slice is 21 opt BUT you have to solve wrong slice last AND you have to solve normal slice unoptimally with +1 instead of +0 so didnt find that:
D2 F2 D2 F2 L' //HTR 5/14
F2 [L2 F2] D2} //double slice 4-1/17
[]=R2 B2 L2 {R2 //wrong slice +2/19
{}=D U' B2 U D //+2/21

(B L' F) //EO+rzp 4c6e 3/3
(R2 L2 D U2 R F2 L) //2c3 6e 7/10
(U) R2 U ^ R2 U' //HTR 5/15
R' L F2 R L B2 //slice 6/21
^=E2 //+1/22

Had so many 10 move DRs but all except this one were 4-5qt.
Also the 8 move DR but it was 4c4 and seemed hopeless. And NISSY confirms that it is 22 opt and all optimals seem unfindable (either 6qt or some DR breaking nonsense...)

(R2 F' L D) //EO 4/4
(B) (F2 L' D2 R B) //4b2 4e 6/10
L D2 L^ F2 L'^ U2 B2 D2 R //HTR 9/19 nice path 4b2 4e -> 4a1 2e -> 0c0 2e
F2 B2 L2 F2 B2 //slice 5-1/23
^=w //+0/23

DR is 22 optimal by switching L D2 L. I was so close to finding that but there were so many HTR variations that I didn't quite get there in time..

edit: damn I didnt NISS the rzp since I didnt realise it was 21 ARM... stupid mistake again.

L2 U2 R' F' //EO 4/4
L' (R2 U' R2 L2 D' L) //4a2 6e 7/11
U (U2 R2 B2 U') //HTR 5/16
(L2 U2 L2 F2) //wrong slice 4/20
(F2 L2 F2 U2 B2 R2 B2 U2) //solved 8-5/23

Other DRs I looked at:
D2 R' B' , L' , D' B2 L2 U D2 L //2c4 in 10. There is a 23 but needs 10 move HTR so nope.
D2 R' B (D') (L F2 R U' R' D) //and variation 2c4 in 10 but doomed. 24 opt.

edit: WOW I missed a set of linear 4 move EOs... Dont remember when this has happened last time. R* D2 R' B

U B' D' F //EO+rzp 4c4e 4/4
(R' D B2 R2 U2 R' D') //nice vibe DR solution for 4a1 4e 7/11
(F2 {L2 B2 L'}) //HTR +1 extension for 4/15
(D2 [B2 D2 R2]) //double slice 4/19
[]= B' F' R2 U2 B' F //slice +1/20
{}= R2 F2 L R2 //+1/21

Unique optimal from DR!

Happy with this since I found the non-trivial (at least for me) DR and then unique optimal from that via double slice.

Also ties cubelib with my settings.

L' D F' D //EO+rzp 4/4
(L2 B R2 F' D2 F' R') //2c3 2e 7/11
(D2 F D2 F' R2 U2 F') //HTR 7/18
(U2 B2 R2 L2 U2 L2) //solved 6/24

Really bad performance this week... 

R' B' F' R //EO+rzp 4/4
F2 U2 D' B //4c3 4/8
U2 B2 D L2 U2 F2 U' R2 U //HTR 9/17
R2 F2 D2 R2 F2 R2 F2 //solved 7/24

Somehow just couldnt find a good finish...

(D F' L B) //EO 4/4
(D) L R2 D F2 R2 D //2c3 4e with variations 7/11
B2 L'^ (L2 D2 B2 L'^ B2 L) //HTR 8/19
(D2 B2 D2 B2) //slice 4/23 
^=w //+0/23

Only DR set that I had so had lots of time to NISS all 2 move reductions to 4b2 2e from all 4 variations.

Also had this:
. //E slice EO on U/D axis
L' //rzp 4c4e
D2 F B2 R //2c4-2s in 5

But shortest slice fixes are 5 moves and only one of them gives a promising subset of 4b2 2e:
U D' R' B2 R F B' //4b2 2e in 12 but the HTRs on the other side are not good so I was quite quick to discard this DR-Xs...

B2 D B //EO 3/3
L D' (B2 L' U2 R' U) //4a2 2e 7/10
(R^) R D2 R'^ F2 D2 F2 R' //HTR 8/18
U2 B2 U2 R2^ F2 //slice 5/23 
^=w //+0/23

Non-trivial optimal from DR. Had to use 4qt 4a2 -> 4a1 2e -> 0c0 2e path.

There were some 22s but I didnt have time to exhaustively check all 4 move EOs fro rzp in 5 so kind of okay to miss those.

(U F' U) //EO 3/3
(R') B R //2c5 3/6 obviously DR in 6 will be 5qt.....
(B L2 F' D2 R2 B' U2 B U2 B) //HTR 10/16
(R2 B2 D2 [R2 F2]) //wrong slice 5/21
[]=R L D2 L' R //+3 but -2 with normal so +1/22

Also had some other 16 movers from this DR but unfortunately nothing better. It is 14 opt but needs HT before reduction so quite hard.

(B D R D2 L) //EO 5/5
(U2 F) //rzp 2c3e 2/7
D2 F2 U B' D B' //DR 2c3 2e 6/13
U' L2 U' L2 U //HTR 5/18
R2 B F' R2 B F //solved 6-2/22

Most Kalindu solution I have had in a long time :D
Rzp in 7, DR in 13 and 9 move finish...
I was looking for 6 move rzps since scramble sucks but happened to see the 7 to 3c2e.
Then after switching I just solved the shortest DR anyway since I already switched. And it just happened to be straight in the 5 move trigger. So I just tried the 2 possible 5 move HTRs and instantly found this :D

Rough scramble.

Had these DR-Xs but they both suck...

L , D2 F , R2 D2 F' D' F //4b2-2s in 8 but no short slice fixes
F, D , L R U2 R U //2c4-2s but 2c4 and -2s so not promising

R' F' D //EO 3/3
(L B) U2 L2 F2 L B' //2c4 7/10
R (R^ U2 R^ F2 R') //HTR 6/16
(D2 F2 * B2) //2e+slice 3/19
*= F2 D2 B2 L2 U2 B2 D2 R2^ //slice 8-4/23
^=w //+0/23

L U2 D B //EO+rzp 4/4
U2 L' B2 U' L2 U' //2c3 4e 6/10
(R' U2 R) F2 U2 F2 R //HTR 7/17
U2 F2 R2 D2 //slice 4/21
+1/22

(R F U') //EO 3/3
(R' B') (L' R' F2 R' U2 B) //4b2 4e 8/11
B2 L2 B2 L' B2 U2 R' //HTR 7/18
U2 F2 L2 D2 B2 //solved 5-1/22

(R) F2 R //EO 3/3
B' , D B2 D2 F U F' //4a2 4e 7/10 you can also do B instead of B'
[U B2 R2 U2 R2 U] //6/16 HTR
F2 L2 F2 B2 //slice 4/20
[]= Uw B2 R2 U2 R2 Uw' U2] //+1/21
Optimal from these 2 DRs. There is 19 to slice with NISS HTR that I had but slice sucks.

Overwhelming scramble. So many 10 move DRs to check that I couldn't check them all.

Missed the 9 move DR that gives 20:
(B R2 U')             // eoud       (3/3)
(F)                   // rzplr-eoud (1/4)
R D2 R2 D2 F          // drlr-eoud  (5/9)
This 20 I should have found...

There is also a 19 but from DR in 11 and no way I'm checking DRs in 11 on this scramble.

U' L' B  R' D //EO 5/5
B (R' B' R B') //4c3 5/10
(D2 L'^ F2 L'^) D2 L2 B2 L //HTR via shortest possible reduction to 4a1 4e 8/18
D2 B2 R2^ B2 //slice 4/22
^=w //+0

Other:
+1 extensions from final DR:
U' L' B R' D, B (R' D2 B' R B)
U' L' B R' D, B (R' U2 F' L F) //2c3 4e in 11

(L' U' B' L2 D) F' U2 F2 R' B2 R //4c3 in 11
(L' U' B' L2 D) L' R' F2 B' L' B //4a2 6e in 11

L U2 B //EO+rzp 3/3
(R2 U' B2 D' U2 L2 U L') //4b2 2e 8/11 with 4 variations
R2 U2 L2 U^ B2 D^ //shortest HTR on the other side so "whack the variations" was easy... 6/17
(U2^ F2 L2 B2 L2 D2) //slice 6-1/22
^= +0/22

Also had:
L' U2 B (U L D B2 L U2 D2 R) //2c3 4e in 11
and the flipped variation:
L' U2 B F2 U F2 U R2 D (U L D) //0c0 4e in 12

R (R2 D2 R') //EO 4/4
(U') B L2 D'F U' //2c3 4e 6/10
L2 F (B' U2 R2 F') //HTR 6/16
(D2 L2 U2 L2 B2) //double slice 5/21 but really 20 since F (B2) 

Now solving this "normally" by doing wrong slice first was hopeless +3 and +2. 
But since this was the only DR I had I tried the other way around.
After DR we have:
(B' U2 R2 [F' D2 L2 U2 L2 B2 F'] L2)
[]= F {B2 U2 R2 D2} R2 F //wrong slice +0/21
{}= U2 D2 F2 U2 L2 U2 R2 //+2/23

I think this was the first time I found this kind of slice solution!

I had:
U R F L U'
B2 R2 F U2 F2 R //4b2 4e and variation in 11 but those absolutely suck...

Also had:
(D L2 U' F2 U' D B) //DR-2s in 7 but it is 4b5 so no hope there.

(U' F' D R L') //EO+rzp 5/5
D' F2 D2 R2 L2 U' F' //4a1 6e 7/12
I really wanted to get U R L type DR to try HTR-1s / HTR-2s stuff :D
Of course I also checked the 2 mover but DR in 11, from R' B (L) rzp, was the best I could do. 
And I argue that U R L DR in 12 is better than "normal" DR in 11.
Especially since my DR in 11 was 2c4. But here that 2c4 happened to have a 10 move finish :DDD

"Unfortunately" here optimal was not of the HTR-2s form but rather "normal" HTR-1s form:
D R2 L2 U L2 U' //HTR 6/18
B2 D2 //"FR" for the slice we try to solve here 2/20
L2 B2 R2 ^ F2 ^^//M slice 24/24 but we will most likely cancel moves slicing since we have R L'
^=M' //+1/25
^=M  //-3/22 R L' from inverse cancels and the D cancels because last F2 becomes D2

Not super cool because no HTR-2s but still quite cool!

U' (F D' R D) //EO+rzp 7c8e 5/5
(R2 D2 L F' L B) //4b2 2e with tons of variations 6/11
L2 F2 U2 L^ D2 R^ //HTR 6/17
F2 U2 [F2 B2 L2 D2 L2] //slice 7/24
but [] = L2^ D2 L2^ F2 B2 //slice -1/23
^=w //+0/23

So many variations and lot of them are not even HTR equivalent...
There are 23s from several of them and even a 22 from at least one of them but 4qt.

D2 F B R' //EO 4/4
U' B' (D' B' U' B') //4c4 6/10
(U B2 U) //I switched all 3 move reductions to 4b2. 3/13
L2 F2 U R2 F2 U //HTR 6/19 This was already in trigger but HTRs in 17 were not as good.
F2 D2 R2 //slice 3/22
+0 with blockyfish

I only had this and the 2c4 in 11 linearly from same rzp.
+12 finish from 4c4 :D

(L2 D B' R) //EO+rzp 8c8e 4/4
(F B U D F' U2 B) //DR bt bad subset 7/11
L2 U^ R2 U'^ R2 L2 D //HTR 7/18
F2 B2 R2 F2 D2 // slice 5/23
^=w //+0

Found this in the last 5 minutes since I spent most of my time on these:
(F U' B') (D) L' R' D' R D //2c3 4e in 9
and
D L2 B2 L' B2 U' F D B //2c4 in 9
but those are ultimate bait :D
2c3 is 22 opt but all opt and opt+1 need 9 or 10 move HTR :D
2c4 is also 22 opt but with 6qt. And all opt+1 again needs 9 move HTR as well...

Luckily the finish from 4a1 6e was linear and easy to find quickly.

L R' B' U2 R //EO+rzp 5/5
(D U' F2 L2 U' B) //4b2 4e with few variations 6/11
(U^) L2 D2 F2 R2 F2 U'^ //HTR 7/18
B' F' L2 F B' //slice 5/23
^=w //+0

Only decent set of DRs I had. 
This was the kind of scramble where my lack of 5 move EO knowledge usually causes issues. 
But this time luckily one of the EOs I did find was the "correct" one.
Also there are no sliceEO skips and only one 1 move sliceEO so nothing there either...
Only thing better than this that cubelib gives is 22 from 2+3 NISS EO so I'm not finding that without getting extremely lucky.

D2 B R2 F //EO 4/4
L (F2 U' L') //4a2 4e 4/8
B2 U (B2 L2 B2 R2 U) //HTR 7/15
(B2 D2 ^ R2 F2 B2 ^) //slice 5/20 
^= E2 //+1/21

DR opt and also best cubelib gives with standard rzp=5 and DR=11.

D L' F2 L //EO+rzp 4/4
F2 L2 F2 U' R2 D2 F' //4b2 2e 7/11 tons of variations!
L2 B2 U' F2 D //HTR 5/16
F2 B2 R2 U2 F2 //solved 5/21

(F' D R B) //EO 4/4
(D) L2 U2 R D' //4b2 2e 5/9
F2 L2 D2 L' D2 R //HTR 6/15
D2 R2 //FR 2/17
B2 F2 U2 F2 D2 //direct solve 5-1/21 

Boring solve but good result

had 2c3 4e in 10, 2c3 2e in 11 and 4c3 in 10 but nothing wanted to finish...

R' U F B //EO +rzp 4c4e 4/4
(F2 D' B2 R B2 L') //oh look the other axis is 6 to DR :D 2c4 6/10
Unfortunately it looks kind of doomed. Shortest linear HTRs are 9 moves and other side is stacked corners.
Now I left out the F B and did HTR-1s //-2/8 
(F2 D' B2 L2 {U R2 U' R2 U) //HTR-1s 9/17
(L2 D2} [B2 L2 F B']) //double slice 6/23
[]= F B U2 //slice -2/21
{}= Uw R2 Uw' R2 Uw2 U' L2 Dw2 //+1/22

Only "decent" DR I found anad just got lucky with it. Everything was bad subset.

Also had these DR-1s:
U' D B2 U D2 R' D2 R //2c4-1s in 8 and variations. But shortest edge fixes dont improve the subset so doomed.
(R D2 L' U2 L2 B' R2 B) //4b5-1s in 8 but horrible subset and shortest edge fixes dont improve the subset so also hopeless.

F (D' L' F) //EO 4/4
(R) (D B2 U2 D2 L') //4b2 2e 6/10
(R2 F2 B2 U'^ R2 U'^) //HTR 6/16
(R2 U2^ R2 U2 F2) //slice 5-1/20
^=w //+0/20
DR opt.
Also had:
(R2 D R' L) (D') (F' B' D2 B U) //2c3 2e in 10. 21opt.

And since there were not that many sub5 EOs and some slice EO skips I again also checked for DR-fe. There were 2 DR-2s in 6:
. //E slice EO on R/L (all flipped)
U' B' D' B2 U F //2c4-2s in 6
and
. //E slice EO on U/D (all flipped) + 7c8e rzp
B2 U' B R' B R //2c3-2s in 6

I only had time to look at the first one but the best I could find in time was:
U' B' D' B2 U F //DR-2s 6/6
U D' //DR-1s 2/8
(R2 # D) //2c3-1s 4e with edges setup for slicing 2/10
D2 [F R2 U2 B'] R2 F //HTR-1s 7-1/16
U2 R L //2e2e 3/19
#= R2 D2 L2 U2 L2 D2 //slice 6-3/22
[]=Fw' F2 R2 U2 Bw' //+1/23
There is a much easier to see 22 that I somehow missed...
After HTR-1s just do: F2 L2 F2 U2 R L U2 //solved in 22

THIS IS THE COOLEST SOLVE OF MY LIFE!

If you are active on FMC discord you have probably heard of the development of new DR-fe method.
So here it is in action!

First we notice that all E-slice edges are already oriented on both normal and inverse.

(.) //E slice edges oriented
(R') //DR-fe rzp 3c2e for U/D 1/1
(D2 L U L) //DR-2s meaning DR-some flipped edges that need 2 slice moves to fix. 4/5
(L' R') //DR-1s 2-2/5
F2 D' B2 [U R L] //DR 4b2 2e 6/11 with lots of variations 6/11 you can also do it without F2 but it gives 2c4.
(B2 U2 R2 {B' D2 B}) //shortest HTR on inverse so we can "whack the variations" 6/17
[]= U' F2 R' L' //+1/18 just go through the DR line and try to build blocks. LOTS of +1 variations availible
U2 B2 //leave S slice 2-1/19
{}= B2 F L2 F' //+1/20

I'm SO HAPPY with this! 5 to DR-1s is very rare and I knew the expected optimal is like 14-15 moves from that. So I spent all of my time on this since even getting opt+2 would mean like 22 that is a good result for me.

This is NOT optimal from DR-1s but I dont mind. I didnt realise that you can also try to solve the slice in the DR line. Here doing
F2 U' L2 D' F2 L' R' instead of my DR line gives +0 slice for 19 that is optimal for this DR-1s.

This was like the perfect scramble to do DR-fe since there were not that many sub5 EOs but one axis was only 2 flipped edges meaning we will have several slices/axis where slice EO is already done.

L (R2 U' L) //EO 4/4
(B) (R2 B2 D R2 D' F') //2c3 4e 7/11
(U^) R2 D F2 D^ //HTR 5/16
L2 D2 * L2 //2e+slice 3/19
*= D2 F2 D2 F2 D2^ F2 //slice 6-2/23
^=w //+0

Optimal from DR.
Also had:
(L') D' L //EO
F' D (F' R2 B D2 B' D) //4c3 in 11 that looks promising. Also 23 opt.

23 is the best cubelib gives without NISS 5 movers or rzp in 6. Buut there is this 19 with +3 rzp from 3 move EO that some people will probably find since people are so good nowadays:
D' L' (L')         // eolr               (3/3)
(F B D')           // rzpfb-eolr         (3/6)
(R2 D2 R2 F' D)    // drfb-eolr [4b2 4e] (5/11)
B R2 F B L2 D2 F'  // htr-drfb           (7/18)
L2                 // fin                (1/19)

(U F L2 B') //EO+rzp 4/4
U D2 R2 L2 U' L' //4c3 6/10
U^ (D B2 U2^ F2 R2 D^) //HTR 7/17
(R2 F2 R2 U2 B2) //slice 5/22
^=w //+1