crypto

[upbeat hazel]

not a guest challenge

[daring dagger]

writes impossible challenge
leaves without saying a word

[wary wigeon]

Thanks for the response ❤️ that's unfortunate

[daring mountain]

Fermat's Last Crypto CTF Challenge... 100% solvable, but not enough free storage in the challenge author's webhost to post the full writeup 😂

[nova breach]

found the solution for shor!!!!1!!

https://www.nature.com/articles/s41534-020-00309-7

[cursive anvil]

Chat is this real

[daring dagger]

are you shor?

[humble lake]

not shor😭

[upbeat hazel]

can someone actually solve the garbled circuits challenge this year

pls

[fleet briar]

The Vorpal challenge (x0 + 1) does not work

[upbeat hazel]

server seems up for me

wdym?

[fleet briar]

Doesn't even ask for page number

[gentle furnace]

any hint for vorpal_sword?

[ornate prairie]

Beware the Jabberwock, my son! The jaws that bite, the claws that catch!

[pastel canyon]

I hope that's not a reference to some older challenge

[ornate prairie]

||https://www.poetryfoundation.org/poems/42916/jabberwocky||

[pastel canyon]

ohhh I remember this poem from high school, haha

[reef marten]

why did i never question the chall name

i was like

thats prob some random string or something

lmaooao

[pastel canyon]

... it's not helping me at all 🥲

[reef marten]

ya hints arent allowed lol

[pastel canyon]

ohh I see

[humble lake]

so where is mea-shor-ment error's write up lol

[upbeat hazel]

the author has not been online since the end of last year's ctf

[crimson pollen]

nsa containment too strong

[sturdy quarry]

unbelievable

[daring dagger]

crazy work

[worldly perch]

ohhh so that's where the names were from 😭

[wraith swallow]

Wait that’s wild 💀

Did bro die

[storm vortex]

he got detained by the organization for possessing the forbidden knowledge

[timber hamlet]

bro is vorpal-sword even solvable?

[hexed solar]

Yes!

[solid venture]

are you sure an org didn't just leak the flag to 95 teams

[hexed solar]

oops mb

[wraith swallow]

You never know

[solid venture]

I actually do know because I was the one to leak

[wraith swallow]

Ah

Makes sense

Did you send it to the 95 teams manually or did you put it on Reddit

[crimson pollen]

He submitted himself in db

[solid venture]

I thought it'd be too embarrassing that we released an unsolvable crypto so I gave out flags

[wraith swallow]

Based

Can I be the 96th?

Not even playing but we ignore that

🤣

[winged vale]

Yeah sure

[supple kestrel]

Why couldnt you think of that last year?

[solid venture]

have you considered that maybe mea-shor-ment is solvable

[supple kestrel]

No

[winged vale]

you're right it isn't

[real sluice]

Can you please send me the flag of nil-circ?

Or also the solvescript would be appreciated

[upbeat hazel]

in 3.75 hours i can

[real sluice]

why dont you send it now and we don't say it to anybody

lets keep it our secret

[upbeat hazel]

i think you'll enjoy the anticipation

[winged vale]

it's a secret challene

[warm patrol]

I can control myself for not posting the solution, what the other users would do with that is out of my control and they didn't seem to behave enough during the challenge to make me believe they would in such case. Please, be patient

[real sluice]

I can control myself for not posting the solution, what the other users would do with that is out of my control and they didn't seem to behave enough during the challenge to make me believe they would in such case. Please, be patient

[fathom basalt]

are you the famous eboy of roma nord?

[warm patrol]

yeee i am :3

[raven stag]

Only solvable crypto this time? o.0

[lapis fractal]

all ever since the mrbeast collab...

[raven stag]

jk fairy-ring is unsovable, that must be some ruse

[real sluice]

Please give me the flag for GC one, I’ve got the solve in mind, but I don’t have enough towels to dry the sweat it takes to be implemented

[raven stag]

I have a solve for mea-shor-ment in mind, but I don't have enough qubits.

[real sluice]

Try to compress towels into qubits, might work

[timber hamlet]

how long until i get writeups for vorpal?

[untold shale]

fr fr I am going mad here trying to work it out

[timber hamlet]

i thought i was the only one stuck at it

[winged vale]

end of the CTF

and if you find any before send them here so we can ban them

[raven stag]

You can have it now for 1 bitcoin. 😈

[untold shale]

Do you have author write ups ready to go?

[winged vale]

errrr

Some have them ready, but generally people will be around to answer questions

[untold shale]

Fair enough. Like I'm assuming its something obvious I just don't know yet considering the top teams answered it almost instantly

[raven stag]

tbh, I think winxy-pistol was easier

[untold shale]

Wait really?

I looked at it and just thought it was a harder version

How vorpal sword has more solves than satisfied does my head in

[upbeat hazel]

i will have (brief) writeups ready

along with sol code

for all of my challenges

[untold shale]

Thank you

I have just run out of ideas to try at this point

[real sluice]

fairy-ring solve?

[viscid nova]

Drop the writeups
I need answers rn! xD

[real sluice]

i have 0 ideas on this one

[worldly perch]

^^

[graceful sleet]

nil-circ solve?

[real sluice]

I didnt solve it, but i think i had the idea
send S/2 the first time so that the you can extract the xor of a wire
The GC implementation uses the free xor, and at that point u can get every pair of key

[supple kestrel]

pls tell me fairy-ring wasnt find the paper

so many rabbitholes for me

[lucid apex]

I tried find the paper way and utterly failed, hope that at least it wasn't like that

[dreamy hazel]

this was my tldr solve for fairy-ring (mentioned blog was https://affine.group/writeup/2023-07-CorCTF-OilSpill)

[bleak zodiac]

was the solution to satisfied related to the fact that one of the generators had order 2q instead of q. Couldn't figure out how to leverage it

[real sluice]

no the permutation can be a fake one, like you could send permutation=[1,1,1,1,1] to duplicate edges

[bleak zodiac]

wow I was in such a big rabbit hole then

Why did they do such sus things to the parameters lmao

[vagrant raft]

i asked the author, it was unintended

[bleak zodiac]

nice

[vagrant raft]

i don't think the order being 2q is exploitable

[real sluice]

i have no idea what is this about xd
I'll have to wait for a writeup

[vagrant raft]

at least not efficiently enough to get the flag

[bleak zodiac]

Agreed

[upbeat hazel]

My challenge writeups: https://priv.pub/posts/dicectf-quals-2025/
Solve code: https://github.com/defund/ctf/tree/master/dicectf-quals-2025

[dreamy hazel]

unbalanced oil and vinegar scheme lol

[upbeat hazel]

this is not a real ring signature scheme

[worldly perch]

fr spent so long on satisfied going down the wrong path before i realized the permute_graph function was exploitable 💀

[storm vortex]

oh sorry i confused the challenge

🤡

[untold shale]

😭 are you kidding me how did I miss that with Vorpal sword

I'm going to go find a corner to cry in real quick

[supple kestrel]

yeah, that much was obvious. what to do with no signatures given was another problem tho 😅

[zinc mirage]

For vorpal-sword you could also do this :

# Craft v so that there is a relation (v - x0) = k^e * (v - x1)
k = 77777777777777777
ke = pow(k, e, n)
v = (ke * x1 - x0) * pow(ke - 1, -1, n) % n
assert (v - x0) % n == (ke * (v - x1)) % n
#assert pow(v - x0, d, n) == k * pow(v - x1, d, n) % n # This will be true

[graceful sleet]

It is on a paper tho

[real sluice]

you could simulate every signature tho, why would you need one?

[upbeat hazel]

yeah it's a well-known-ish trick

ok not actually well-known but

i don't think it's attributable to a specific paper

[full temple]

https://eprint.iacr.org/2024/720.pdf maybe?

[lucid apex]

https://priv.pub/posts/rainbow-attack/
so it was actually "find a paper"

and even with a hint of where to look for

[supple kestrel]

and here i was almost modifying a generic attack to work on this

just needed a few more signers

[lucid apex]

very good challenges @upbeat hazel , thank you for the work

[upbeat hazel]

it's also mentioned in mayo: https://eprint.iacr.org/2021/1144.pdf

[worldly perch]

^^ enjoyed the crypto this ctf

[upbeat hazel]

can you guys please make writeups? 🥺

an end-to-end writeup for nil-circ would make my day

[fathom basalt]

very curious to see what the human way of writing nil circ is

my way took way too long to write

[upbeat hazel]

wdym by "writing"?

like your solution?

[fathom basalt]

sol code

the actual rust thing

[upbeat hazel]

i basically patched their

evaluator code

to inpesct things

while evaluating

[civic elbow]

Thanks for the challenges this was really nice 👍

[fathom basalt]

so did I but I have no clue what traits are so I copied and created a new file called mycircuits.rs and removed all "subclasses" or whatever these things are

and rewrote everything from scratch using WireMod2 instead of Item, no templating etc

[upbeat hazel]

o

https://github.com/defund/ctf/tree/master/dicectf-quals-2025/nil-circ/solve/src

i'm not claiming this is pretty or anything

anyways i just exported a list of constraints, and then solved linear system in sage

[fathom basalt]

ok actually mine is not that far off lmao, maybe there is no (significantly) better way of doing it

[full temple]

I also modifed rust code to print the 4 possible outputs of AND gates (using the free-xor key), then was parsing these logs from python. Most difficult was figuring how to pass the free-xor key around in global variables (also build setup).

Tried to do a correlation attack but failed, idk why.

[fathom basalt]

I was also having that issue, ended up printing it from inside the OT and then gave it to stdin from outside the call stack

[supple kestrel]

i think i patched like 10 files in the library while debugging this shit, then parsed the output into z3 to get key

[supple kestrel]

doesnt breaking one bit to get the xor mask completely fuck up the first and gate it reaches?

[real sluice]

i wasted so much time on that chall as i thought initially it had something to do with the fact that the OT was modified to run in not costant time

[full temple]

It does, but other 15 S-boxes should be fine

[supple kestrel]

sure, but where would you do correlation then? at that point its way simpler to do it with the and gate

[raven stag]

We found a paper that implements exactly that scheme: https://link.springer.com/chapter/10.1007/978-3-319-57339-7_1
xd

[full temple]

I tried to correlate with every other AND gate, the ones not corrupted.

[raven stag]

I just hooked into circ.eval to print the gates and AND inputs and solved with z3.

[upbeat hazel]

oh wtf lol

i mean ig it's like an intuitive construction

that's funny

i wasn't aware of this

[raven stag]

And they have a proof that it's safe. xd

[upbeat hazel]

erm

[old sage]

did it the same way. but i just passed it as the last input (i simply modified the OT to have a special case for the last index) so was pretty easy to get it in the gate eval one

making sure the gate indices/communication stayed sync was much more annoying in my impl

[raven stag]

But I think that they don't consider the quadratic forms at all. That just count number of equations and number of variables and say it's ok according to state of the art generic algorithms.

But well, some MQ instances can be solved.

[supple kestrel]

we wasted well over an hour because we didnt notice the gate number was getting incremented in that call... such a nice chall made painful because rust libraries doing rust library things

[raven stag]

Yeah, I made that bug too, but didn't wasted much time there, I was making sure to only make small changes and verifying it's still works.

[old sage]

yeah whoever named that get_gatenum(?) function deserves a stern talking to

[supple kestrel]

i called the and gate multiple times to get the results

[supple kestrel]

exactly... who tf names a function that modifies shit like a getter

[raven stag]

Yeah, I did that to at first. But because that was the only change I made and it broke I found the problem quickly.

[old sage]

in the end i hacked in a whole other trait that allows to modify the gate count and eval and's without communication, don't know how else to do it when everything's generic

[raven stag]

I mean, when I knew there's same state in this function, that one was the most obvious one tbh.

[supple kestrel]

and do they limit the number of signers? because this thing is breakable pretty easily with 15 signers, and i think it should be doable with 11

[supple kestrel]

meanwhile

[raven stag]

They have some other parameters for more signers there. Because yes, with like 11 signers there's a generic MQ algorithm that can solve it.

[raven stag]

same! 😄

[upbeat hazel]

how many signers did i give

[raven stag]

6

[upbeat hazel]

o

[supple kestrel]

yep, and i spent like 5 hours trying to figure out if i can stretch it to what we have

[raven stag]

So, for fairy-ring was it important for the solve that you can use the same public key more than once, or not?

[upbeat hazel]

i think so?

my solution relies on it

[supple kestrel]

with 11 signers, you dont need that anymore

[upbeat hazel]

right

[raven stag]

wtf is polar form? o.0

[upbeat hazel]

idk what the mathematical basis for it is

but in some sense it represents the derivative of the multivariate quadratic?

and it's bilinear

if your MQ is F(X) = x^t A x, then the polar form is G(x, y) = x^t (A + A^t) y

notice F(x + y) = (x + y)^t A (x + y) = x^T A x + x^T (A + A^t) y + y^T A y = F(x) + G(x, y) + F(y)

[raven stag]

Oh, wow, it works! I was trying to come up by myself how to get linear equations from two (or more) quadratics but couldn't come up with anything. :C

Hmm, I guess it had many solves because ppl found your writeup? xd

I liked "satisfied", I had hamiltonian zkpok on uni like two weeks ago. xd

And was like, yo bro, that's totally correct, that lib must have a bug. :>

[supple kestrel]

i tried to cheese the ch chall when it came out with the intended here, but sadly there it doesnt work by pure luck

[raven stag]

Really liked the crypto challs. Thanks defund!

[raven stag]

In winxy-pistol the same key is always used, unlike in vorpal-sword where it always generates a new one.

[untold shale]

Yeah. Honestly I think I would have had better luck looking at it.

The Vorpal sword was just me being an idiot

And no amount of looking was going to fix the fact that I had forgotten that the relationship held even when you perform mods

[old sage]

for fairy ring i used messages of the form [(1,v), (1,0), (0,v)] with v a vector of unknowns, then everything just cancels out except something fully linear in v.
but i did get stuck in an interesting rabbit hole for a long time: i tried selecting 9 indices in the signatures, set everything else to 0, solve the linear equation system in the x_ix_j - where x_i is now a 6-dimensional vector (sig_j[i])_j - as variables (44 equations, 45 variables) and hope that the resulting system is solvable. unfortunately, the symmetric 9x9 matrix of those products would need rank <=6 (though not sure if that is even sufficient? basically, i'd need to find a "symmetrical" rank decomposition M * M^T) and that happens just a bit too rarely to be ctf-feasible (rank 7 is doable, but effort is something like ~256^(k²/2) for rank deficit k))

not sure if something along those lines could maybe be used for an attack with different keys too (with 7 keys at least)

[supple kestrel]

i have no idea what i just read, but looking into it more, i also think this should actually be solvable with 7 keys without key reuse

[raven stag]

It's probably only useful if you solved it and want to compare approach, but here's my solve for nil-circ. To compile that rust code you also need to patch swanky lib and make some of the fields public.

[raven stag]

@upbeat hazel https://github.com/defund/ctf/blob/master/dicectf-quals-2025/nil-circ/solve/src/garble_evaluator.rs#L102
Wait, how did you make them equal? I've checked and all four were pairwise different. I've had to only look at the least significant bit (i.e. .color()).

[upbeat hazel]

it's checking whether any three are equal

[raven stag]

Ah, wait, I think these were pairwise different before I noticed I had to add self.current_gate -= 1;. xd

And I think I haven't checked after I fixed it.

nvm

[proud tree]

long writeup about nil-circ
https://blog.tanglee.top/2025/04/03/Revisiting-Garbled-Circuit.html

[weak ravine]

clean writeup dam

[versed jewel]

hi guys
Hi everyone 👋 I’m working on a CTF crypto challenge and I’m stuck
can someone help me plz find the flag code
the problem is that we dont have the secret_function()
plz any help

i have 30 min left plz

[winged vale]

you shouldnt ask for help for active CTFs 👍

i would normally delete but I think it is funnier for everyone to see this and for there to be an example of what not to do

[bronze sinew]

secret function

[somber basin]

Would be helpful if you can share the total output.txt (im only trying to help cuz your nickname somehow looks similar to mine)

[upbeat hazel]

.

.

.

people should look at crypto/dot

it's actually about morse code

dots and dashes

[ebon nymph]

is carry-the-flam up?

[hexed solar]

Looks good to me

[ebon nymph]

i guess issue is from us

oh it works now

[quartz brook]

why did it have to be 65000 rounds

[vague mural]

Why not 65537

[quartz brook]

Ragebait

[tawdry kite]

wait why

why the change in rounds

[hexed solar]

@analog stump

[analog stump]

niceness reasons

[hexed solar]

how kind

:3c

[weak ravine]

Ok I drink

[stark pecan]

can you put an instance up with 2^16 rounds i wanna flex

[daring dagger]

flex your gpu or did you actually solve the challenge

[mental knoll]

Wow you are so rich :3

[stark pecan]

no:)

[weak ravine]

nuh uh

[ember terrace]

these zkp challs so tuff huge respect to those who solve them

[topaz sorrel]

https://tenor.com/view/robot-robots-bot-bots-robot-army-gif-19992580

^ ppl who solved them = chatgpt or claude

[midnight gazelle]

modern Doraemon 😂

[fast summit]

Saddening

[mental knoll]

Whoever made this flame crypto you’ve got my hate.😭 I hate you so much...

[humble lake]

flame(your computer)

[muted star]

haha

[quartz brook]

HOOWWWW are you supposed to do carry the flame??????

[smoky crow]

P2w

[ebon nymph]

No way

[quartz brook]

but what is the attack

[smoky crow]

5 mins on rtx 5090 were enough

[wispy knoll]

By claude.

[muted star]

yes.

[clear needle]

wu for dot pls

[quartz brook]

duuuuude

kill me

[smoky crow]

dot [crypto]

The vuln is that the verifier does not check a full proof. It only checks whether a single dervied group element lands in a precomputed lookup table. Because the CRS is public, we can build an honest proof locally for any correct addition. also for flipping one chosen output bit of c there is a fixed public correction in the proof coordinates that almost completely cancels the verifires hidden constraint contribution. leaving only 1 hidden scalar lambda on the base point. that scalar only depends on 2 small hidden coefficients.

so for a challenge (a,b compute the proof locally but instead of answering with the correct c flip one chosen bit of c and add the corresponding public correction to the proof. Then subtract a brute-forcedlambda *G. if the brute-forced value is correct the verifier sees exactly the same table point as for the proof. so it accepts the wrong answer and prints huh?. after getting lambda we repeat that forged wrong proof on 20 new challs to build the streak and gg

bruteforce lambda (im not crazy to do it in the 34 bit space)

we know that lambda = u * (1 + b * (u + w)) . where u and w are both only in [-256, 256]. So instead of searching an arbitrary 34-bit space, just used this. That gives only about 262k candidates, which is around 2^18, not 2^34.

[quartz brook]

was brute force intended?

[chilly onyx]

are we serious 1152 cores

what the hell bro

😭

[smoky crow]

I was trying to actually solve it but then one of the teammates said wait it could be bruteforced

[weak ravine]

We just cuda'ed it

[smoky crow]

Yea cuda ftw

[ebon nymph]

Are we for real

[solid pewter]

how do you empty out the market?
couldn't find any way within the contract for housing

[chilly onyx]

me when amd

[wispy knoll]

I didn't think of using a GPU. 😭

[mental knoll]

Me and @fossil forge tried cuda but got unlucky

[smoky crow]

Using this

[muted star]

same

[dusk junco]

any way to solve carry-the-flame without brute force? I tried the slide attack. But fetching 2^20 plaintexts takes too long

[quartz brook]

same

[smoky crow]

We trued that at first

[quartz brook]

it works locally but remote is a potato

our better attempts were around 0.02 % success odds per attempt

[chilly onyx]

how is one meant to solve carry the flame without a trillion cpu cores or an rtx 6090

[mental knoll]

I thought theres something intended way not bruteforce

[muted star]

maybe optimizing silde attack is intended sol but bf 5bytes is..

yes

[ebon nymph]

I cry

[stark pecan]

i used 14x 4090s, solve time ~1 minute, would have been an hour or so before the reduced rounds

first blood go brr

[prime ferry]

a challenge worth $14

[solid pewter]

the-2000s-american-housing-crisis

[mental knoll]

I tried a lot but no luck then we just bruteforced it

[stark pecan]

cost $5 btw

[fast summit]

So without being pay to win, how can we solve the flame chall

[gusty bough]

fuh nag

[smoky crow]

Worth it

[bleak zodiac]

Lol I think we will need an official writeup to see if there is a poor man's solution

[fast summit]

fr

[dusk junco]

yes

[flat breach]

same, and to optimize it further, I implemented that cipher using bitslicing

[muted star]

#crypto message

check this image

[stark pecan]

again, $5

per hour 😉

[chrome hawk]

bro aint no way people bruteforcing the last one like that 💀

[muted star]

pro optimizing man

[smoky crow]

It costed us 0.389 or smth but had to pay 5$ as the minimum

[stark pecan]

i got 18 billion keys per second

[lean jasper]

x1 A100 2 hours

[gusty bough]

💀

pay to win ctf

[weak ravine]

gosu

[mental knoll]

Bruhhh

[fast summit]

keys per second unit crazy

[muted star]

[clear needle]

and any tldr for american-house pls ?

[mental knoll]

You got nasa's server room access?

[somber basin]

birthday for 2^20? just save 2^32 and connect 256 times smh

[solid pewter]

+1

[stark pecan]

vast.ai go brr

[stark pecan]

pow took longer per solve

i implemented multi connection like that

and threw it away once the rounds got reduced

[gusty mantle]

did this, tried for hours, lady luck hates me

[stark pecan]

not worth it

[bleak hamlet]

just do 60 sessions and crack all of them at the same time. cracking the ones with high entropy using the gpu. i could solve it in 1:30m with my RX 9060 XT. btw i don't like tath chall

[crisp nova]

Did anyone attempt Planar without AI?

[smoky crow]

the-2000s-american-housing-crisis

1. market 1 can be bet/resolved before it is “created” (because bet/resolve don’t check nextMarketIndex).
2. createMarket() later sets marketResolution[1]=0 but does not clear yesVotes/noVotes/totalYesBet/totalNoBet/cashedOut.
3. This lets stale phase-1 bets be reused in phase-2 payout math, creating overpayment and draining the seeded 1 ETH.
4. DAO owner control was achieved via the proposal/tally proof flaws, then used to run resolve/create/resolve on market 1.
   Working amounts and sequence:

Phase 1 bets on market 1 :

• A(us): yes 0.6 ETH, B(unlocked dev acct): no 0.4 ETH
• Owner resolve yes, then A cashout -> market back to 1 ETH
• Owner createMarket(...) (re-inits market 1 resolution only)
• B adds second no 0.2 ETH
  Owner resolve no, then B cashout -> market balance 0, isSolved=true
  Gg

[flat breach]

Could you tell me more about that idea?

[bleak zodiac]

I spend all my time doing cryptanalysis on flame only to find out I was supposed turn off my brain and whip out my wallet 😭

[nova crag]

hello, didnt play this ctf, just asking did the challenges resist ai?

[gusty mantle]

assume first byte of the key is 0, build a dictionary of encryptions of a chosen plaintext (say all zeros) with all possible keys (only lower 32 bits), retry until oracle gives you an encryption you've saved

[somber basin]

save enc(\x00 * 5, key) for 2^32 keys, and pray the key matches on remote (probability 1/256)

[gusty mantle]

20GB by saving only the ciphertext as a plain binary file (with key being the index), search does become linear tho, but the bottleneck is still the pow lol

[fast summit]

crazy omg

[flat breach]

Thx

[flat breach]

Thx

[reef marten]

no

[spiral narwhal]

two or three did

[weak ravine]

Was trying this till I got rage baited

[slender meadow]

crazy out of the box thinking

[weak ravine]

Jerry did you manage to solve it that way?

[somber basin]

not trying to be sarcastic or anything sorry but i dont feel this is anything close to a 6x 🔥 idea , what are the standards these days

[tacit delta]

wdym 6🔥 ide

[humble lake]

is there a way to solve flame with cpu only?

[tacit delta]

what was the intended solve tho? i solved locally with same method, tho generated only 2^30 keys

ig slide attack was applicable there somehow, but my connection got reset all the time and i couldnt get enough cts

[gusty mantle]

how long did you run it?

[humble crag]

how many gpus did you use

[somber basin]

this took 3 minutes (for saving 20GiB file) with 64 core cpu

[humble crag]

my worst case was 15 mins with 8 4090s

[somber basin]

so original 2^16 was doable as well with cpu

[bleak zodiac]

I'm fairly sure nobody who solved the challenge did it without just throwing compute at it so if there is one nobody but the author knows it

[tacit delta]

wdym 3 mintues lol. i generated 16 gib in like 5hrs i guess

[somber basin]

idk llm helped me

[stark pecan]

14

[tacit delta]

eh couple of hours ig

remote didnt work out cuz of the PoWs

i mean, i left it solving PoW's today in the morning and wanted to see if it solved this evening, but i couldnt get to my laptop in time

[tacit delta]

ig there was a way to apply slide attack. it just cant leave my mind for some reason, same key and 65536 rounds

[bleak zodiac]

Yeah that's what I got too but still needed far too many pairs for remote to handle with normal compute

[gusty mantle]

there is, just not feasible in python, reimplement in C/C++ and it works

[spiral narwhal]

it doesn't work due to data complexity
(unless you mean reimplementing the server)

[bleak zodiac]

I was having issues just getting pairs from the server

[gusty mantle]

yes I mean the server

[tacit delta]

how th did you manage to keep the connection alive? it just couldnt generate enough pairs for me

[gusty mantle]

I mean the existence of the attack theoretically

[tacit delta]

oh

yeah sure

[gusty mantle]

nah same as you, I tried the 2^32 keys but wasn't lucky enough

[tacit delta]

so when you generated 2^32 keys, you bought gpu or smtn? I did it on my laptop's builtin lol

i mean that follows from the conversation, but seems somewhat cringy

[gusty mantle]

CPU, i3-10110U

[tacit delta]

wth

[gusty mantle]

took close to 2 hours but I obviously reimplemented the cipher in C++ with optimized sbox and pbox

and parallelized with openmp

[tacit delta]

might be WSL issue

we reimplemented cipher as well

[spiral narwhal]

bit slicing or avx512 gf2p8affineinvqb?

(I found that the former performed much better because the p-box was difficult to implement in parallel with the latter)

[somber basin]

i wish rounds was 2^16

[reef marten]

real

[gusty mantle]

just precomputed lookup tables

[spiral narwhal]

as in aes fused s-box + mixcolumns?

[gusty mantle]

yup

[mental knoll]

Me too it should be fare for everyone

[woeful otter]

Debatable.

Not everyone has access to paid LLMs/agents either

So unless that's banned, I don't really see much harm with p2w challs either

[chilly onyx]

there's quite a difference between purchasing a $20/month plan online than buying a $2000 gpu?

[somber basin]

400 times cheaper btw

[spiral narwhal]

as well as purchasing a $0.5/hour plan online for the duration of 30 minutes

[woeful otter]

You could hire vps for way less btw

[chilly onyx]

hmm i suppose yeah

[mental knoll]

Even vps cost 1.1$/hour

[woeful otter]

Well yeah this chall takes like 15 mins on a vps(ours did at least)

So 1.1$

[chilly onyx]

well surely the intended sol wasn't to rent a vps?

[woeful otter]

I hope so

[chilly onyx]

still am curious what the intended is

[mental knoll]

I tried slide attacks and birthday attack for long time but got no luck

[weak ravine]

@upbeat hazel ‘s challenge was banger atleast, solved welcome and dots

[upbeat hazel]

Did you slop it

[mental knoll]

At a moment i thought it could be tmto attack

[weak ravine]

I slopped welcome

Sorry defund

[woeful otter]

Btw soon hari' solve, on the remote should take a few hours I assume? Given it takes all 256 attempts to hit the right probability at the worst case, and solving 256 pows should cost 2 hours at least?

[upbeat hazel]

Did you slop dots

[zinc mirage]

Oh you remind me I should fill the survey now instead of curl-ing it for the flag

[weak ravine]

No I suffered

[upbeat hazel]

Hooray

[somber basin]

i used 64thread

[woeful otter]

Ah

[weak ravine]

Yeah I told myself I will solve yours and gtfo

[weak ravine]

I think I know who did this

[winged vale]

just use like

hetzner

[lime idol]

[winged vale]

yeah i wont recover

😭

still the goat

[weak ravine]

It’s alright @lime idol you can use my vps

[mental knoll]

Finally i can sleep now ..

But I’m wondering… What if the author didn’t reduce the rounds? Within 1200 seconds, it might be hard, right? And it would also be a more expensive challenge..

[somber basin]

most of the quick enough solves just need 1~2 pair of ciphertext so i dont think it matters much

[river wadi]

oh

lol

[humble crag]

3 dollars total :p

had to fix some bugs in the cuda kernel

[topaz sorrel]

what was the bug tho

[urban brook]

"together broken" almost certainly refers to the slide attack breaking the cipher
through its repeated structure but getting it to work is another thing

[winged vale]

ngl

it doesn't

I came up with that flag because it was funny

That's all there is to it

[urban brook]

was it meant to be brute forced like that? I'm still trying to figure it out

[winged vale]

@analog stump

[old sage]

got to like 0,1% and it was at times random enough that i thought it might have a decent shot (after all its ~quadratic in the succesful runtime)

[quartz brook]

did you track how many blocks you were getting per attempt? we had ~10000

[spiral narwhal]

so just open 1000 connections in parallel

[quartz brook]

well the compute is on the server so feels a bit wrong to do that

[weak ravine]

Huh who were you playing with @old sage

Waddup manf

[spiral narwhal]

oh... hopefully they have 1000 cores

[old sage]

im sure that would just make it a lot less unstable

[old sage]

femc

wasnt really very actively playing though

for obvious reasons not very motivated

[weak ravine]

Yooo based

Good to see you here horse man

[crisp nova]

#writeups message

[gusty bough]

waitt no way

[humble lake]

any news for mea-shor-ment-error?

[daring dagger]

maybe once quantum computers are better

[fathom basalt]

I think we should try to clank it at this point in time

[daring dagger]

unsloppable

[winged vale]

crypto more like rot52

[inner egret]

http://rot26.org

[glacial shell]

Nice

[topaz dock]

Its backdoored by the NSA 😮

[sacred blade]

newcrypt 2.0

[upbeat hazel]

sorry in advance

[analog stump]

ok defund

[normal wyvern]

damn, it really be that way

:(

[sour peak]

oh no

[upbeat hazel]

although none of my challs are coming out at the start

well, except

[analog stump]

did you add guess the cipher

[upbeat hazel]

👶

can't guess the cipher if it's made up

[indigo stag]

not sure if that’s announcing guess the encoding or bruce password 2

[chrome gale]

Bruce password 3

[analog stump]

bruce password 2

[chrome gale]

I choked when reading the 1st one

[analog stump]

the first one i need to implement

[visual bane]

i have a ton of crypto challenges to do 🙂

[upbeat hazel]

there are moreyet to be released

[visual bane]

yeah, only two crypto challenges iykwim

[upbeat hazel]

should be three?

[visual bane]

so bold of you to assume i would do newcrypt

[upbeat hazel]

btw, this might be tangentially useful for newcrypt v2: https://priv.pub/posts/redpwnctf-2020/#newcrypt

[solid venture]

ok defund

plugging your blog

[compact jacinth]

defund plugging everything

[upbeat hazel]

wait like

i just don't wanna give unfair advantage

to redpwnctf participants

[deep sedge]

that helps a lot thanks lmao

[visual bane]

unfair advantage, i see

[glacial cobalt]

thanks defund

[indigo stag]

yeah well

that's hard

[deep sedge]

shouldn't be too bad if i'm reading this paper right

lot of equations which is annoying

[solid venture]

I would talk about difficulty curves but it is clearly all difficulty and no curve

[analog stump]

tuxic challenges

[prime eagle]

blame defund

[solid venture]

TRUE

@upbeat hazel explain yourself

[analog stump]

@upbeat hazel mean

[upbeat hazel]

tf

i didn't write any of them

[solid venture]

that's the issue

[analog stump]

making us suffer

with tux challenges

[upbeat hazel]

oh

😭

i'll have a challenge

coming up

and then another

[analog stump]

[upbeat hazel]

all of my challs are certified no paper needed™

[old violet]

I'd rather play UHC than do crypto

[glacial cobalt]

aplet omg curvy 🥰

[deep sedge]

newcrypt will be solved soon

Kappa

[prime eagle]

👀

[deep sedge]

my head is spinning after reading these equations

[solid venture]

daring claim

[deep sedge]

surely there's an easier way

[solid venture]

it was made by our lord and savior tux the penguin

🙏🐧🙏

[glacial cobalt]

I wish I was smart enough

my head bork

[analog stump]

he made it tuxic

af

[glacial cobalt]

aplet pickle?

[deep sedge]

tbh this is just algebra at this point

if i knew how to use sage i think i would save myself this manual analysis

[chrome gale]

I wish I understood newcrypt write-ups

[glacial cobalt]

crypto challs tuxic

[merry bolt]

the funny thing is, there are two problems which could've been called plagiarism, because both change another CTF chall slightly

I wonder if they had to actively choose which to call that

[wary geode]

huh, which ones

[merry bolt]

plagiarism and newcrypt v2

[surreal coral]

huh which ones

[wary geode]

oh

newcrypt v1 was written by the same person

[halcyon belfry]

tbf newcrypt v2 is same author

[wary geode]

[halcyon belfry]

dang sniped

[merry bolt]

self-plagiarism

[eager prism]

[halcyon belfry]

tux was just sad that it got cheesed

[wary geode]

"cheesed"

[glacial cobalt]

nice reordering of the problems

[merry bolt]

if I stole my own intellectual property I would be quite angry

and to think, it wasn't even cited as plagiarism

the lack of irony!

[glacial cobalt]

you would be quite angry at who may I ask

[surreal coral]

god

[merry bolt]

ok, i'm god

thank you, i appreciate it

[eager prism]

[glacial cobalt]

hi god

im EvilMuffinHa

[merry bolt]

i have to point out that you just assumed my gender

i guess i'll pardon you

[glacial cobalt]

uh

I don't see it

[upbeat hazel]

https://www.youtube.com/watch?v=BZqWUyh1ugw

motivating music

power on!

[wary geode]

I prefer https://www.youtube.com/watch?v=eY52Zsg-KVI

[upbeat hazel]

some more motivation: https://www.youtube.com/watch?v=WVErYPCmmeA

[crimson cape]

I would recommend https://youtu.be/Sagg08DrO5U

[upbeat hazel]

Big Dirty Stinkin Sax

ok gink

[crimson cape]

hoy it's just gandalf

nothing wrong here

[deep sedge]

i failed

rip

[prime eagle]

F

[deep sedge]

i might've manually copied and pasted wrong

can i dm tux cause i think i have a solution that works

[upbeat hazel]

yeah he's online right now

[wary geode]

https://ginkoid.is-inside.me/Ih1EcXFv.png

made some good progress

[deep sedge]

gg too hard

[idle dirge]

hmm i think there is some error on chall 😦

[halcyon belfry]

which chall?

[idle dirge]

newcrypt v2

[halcyon belfry]

actually I think all crypto authors are offline rn 😦

dm @compact jacinth

[idle dirge]

ok thx

[eager prism]

tbh any chall i cant solve has an error

anyways im 0.08% of the way to getting a flag 😎

[ripe swan]

mood

[eager prism]

on that note, would anybody like to lend me a supercomputer

[idle dirge]

I'm sure that output file is wrong on newcrypt v2, but author is gone 😦 so sad😭

[latent tiger]

how many cores i need for plagiarism chall ? 👀

[turbid trellis]

e is too big in plagiarism

[analog stump]

the challenge™️

[maiden narwhal]

I mean it has a solve, so the chall is working

[latent tiger]

i'll give some more time on this then will switch back to pwn 😔

[vernal crater]

what if they just got root on quantum and used that to factor 🤔

[indigo stag]

what if the chall is to guess the flag from the former chall flag

[latent tiger]

lmao

i still think my solution is correct but it's very compute intensive

[upbeat hazel]

:O

nice bloods on the crypto

while I was asleep

[indigo stag]

yeah same I would be interested to know the approximate running time of the solving script for plagiarism

[latent tiger]

^

[upbeat hazel]

I think this is fair game for me to answer

on the order of an hour

[indigo stag]

allright thanks

[upbeat hazel]

one of my cryptos is live

[surreal coral]

cheese solution for quantum : wait for ibm to build a fault-tolerant quantum computer and then just run the circuit

[prime eagle]

🤪

Mslc 🙏

[upbeat hazel]

see #announcements

[analog stump]

[upbeat hazel]

newcrypt v2 update

[upbeat hazel]

not sure if this is relevant to you, but ^

[idle dirge]

wow.. it fixed! Thx

[turbid trellis]

Same, mine is gonna take like days to finish cuz of the complexity

[surreal coral]

wow almost like its designed to take more than 48 hours to run

[deep sedge]

most likely there's some optimization you have to make use of but my google skills are failing me

[tight timber]

just get a faster computer 4head

[surreal coral]

the intended solution for all of the rsa challenges is to factor N on a quantum computer

[turbid trellis]

if there's an optimization I would really like to know the solution

[tight timber]

time complexity has no practical applications, just wait for moore's law to catch up to whatever hard problem you're doing

[surreal coral]

tru

and when moores law plateues, just add more cores

because literally everything is parallelizable

[strange grotto]

is the public key missing from crypto/plagiarism?

or are we supposed to generate it ourselfs from N and E?

[surreal coral]

shouldnt be? N and e are provided

[deep sedge]

the public key is N and e

that is provided

[surreal coral]

thats... thats the definition of the public key

[strange grotto]

yeah sry, got confused by this sentence 1) They used RSA with this public key

[indigo stag]

polynomial gcd on fpga ftw

[surreal coral]

oh yeah that's a hangover from the RuCTF copy-paste

[strange grotto]

from the looks of it the challenge seems do-able

[surreal coral]

from the bloods i would agree

[latent tiger]

i've found some ways to optimize but nowhere near flag 😔

[tight timber]

just eliminate more constant factors smh

[deep sedge]

4Head

[surreal coral]

the hxp solution already runs in polynomial time, my computational complexity class told me that that always means its fast

[deep sedge]

isn't gcd supposed ot be logarithmic

[tight timber]

it's logarithmic with respect to a constant

[strange grotto]

my brain just died while you guys where talking about math principles

[surreal coral]

math? in MY crypto?! it's more likely than you think

[tight timber]

wtf, i thought crypto was guess the cipher ?!!??!

[strange grotto]

I know, yet my understanding of the field is not as high as I hoped it to be

[surreal coral]

it's guess the paper now

[tight timber]

[deep sedge]

just get a phd in math

[icy hamlet]

nuthin like a good ol ctf to remind me my python skills need some SERIOUS work! lol. im sitting here for an hour with the error that it cant find local module

[floral reef]

it do be like that

[upbeat hazel]

local module isn't meant to be found

it just contains the flag

it's just for the challenge setup D:

[icy hamlet]

@upbeat hazel it gives me this when i try to run the python script:
from local import flag
ModuleNotFoundError: No module named 'local'
***like i said my python skills need serious work. im a total noob with it and maybe it is because i have been up all night long, but my brain has pretty much stopped computing common sense hahaha

[upbeat hazel]

so if you want to make your own public/private key and encrypt a fake flag, you can just create your own file called local.py and place flag = b'dice{this_is_a_test}' into it

but otherwise all that's saying is that I ran the script locally with my own flag, which you need to recover

sorry if that's unclear 😭

[icy hamlet]

nah its all good. so could i technically just comment it out then?

[upbeat hazel]

yeah sure

you can just remove the entire if __name__ == '__main__' block if you don't want to generate your own data

[icy hamlet]

oki thanks 🙂 hopefully that will get my brain back on track a bit

[upbeat hazel]

np

[latent tiger]

lol imagine PFS is actually carrying RSA factoring 0-day

[prime eagle]

lol

[floral reef]

rsa 0day 👁‍🗨

[deep sedge]

can you guys release an easy question so i don't feel like an idiot

[strange grotto]

whuz 9+10?

[prime eagle]

21

[strange grotto]

u STOOPID

[icy hamlet]

im with ya lol. I have done many CTF's before and done very well but this one is kicking my butt. at least im learning a lot

[strange grotto]

I am struggeling as well with finding some kind of alternative for the algebra import hxp explicitly said the following about: Using my Python algebra library (that, by the way, only came into existence while solving this challenge since I was unable to find packages that could properly handle polynomials over arbitrary rings), the required computation is easily implemented:

no luck so far

[violet coral]

Is there any chance of a 17th root over some ring to complete in a reasonable time?

[deep sedge]

no

[icy hamlet]

i was able to import algebra with no problem. check your path settings

[strange grotto]

what python version are you using?

[icy hamlet]

3.9.1

[violet coral]

sad

[strange grotto]

huhm okay

[upbeat hazel]

.

[analog stump]

what does this mean

[upbeat hazel]

oh

i made a baby crypto

but apparently

[analog stump]

is it getting released

[upbeat hazel]

it was already a thing in a previous ctf

so we scrapped it last second

[analog stump]

bruh

so theres no baby crypto

truly a msfrowning moment

[upbeat hazel]

😦

[surreal coral]

yeah we don't like any plagiarism in our chals, even unintentional

[indigo stag]

what was it? 🙂

[analog stump]

^

[compact jacinth]

guess flag

[upbeat hazel]

it was called

[analog stump]

ah of course

[surreal coral]

https://activities.tjhsst.edu/csc/writeups/damctf-2020-guess

[upbeat hazel]

babypad

[indigo stag]

benaloh is cool btw

[upbeat hazel]

ah thanks 🙂

[upbeat hazel]

I have another crypto that will likely be released

[icy hamlet]

😦 maybe try taking off the text before pip? so pip install algebra

[strange grotto]

will install for python2 then I guess

lemme give it a try

[surreal coral]

plz stop trying to install hxp's custom algebra library that isn't on pip

[prime eagle]

lmao

why

not

[wary geode]

hey that's me

[upbeat hazel]

sad

[strange grotto]

great. welp lets get back to the math PhD then

[floral reef]

o.o

[latent tiger]

imma dm yyyyyyy and ask for his python library , ez pz lemon squeeze

[upbeat hazel]

presumably their library won't help you solve this challenge

[latent tiger]

kekw

[surreal coral]

the point of the challenge is to pwn yyyyyy and steal their uber secret crypto lib

[latent tiger]

https://cdn.discordapp.com/emojis/615215498875109377.gif

[full temple]

oh, so it's not forbidden?

[upbeat hazel]

👀

[abstract elk]

👀

[strange grotto]

Depends on your country of residence 👀

[full temple]

my main concern is to not be disqualified

[strange grotto]

tbh with 1 point on my track record for this ctf

[deep sedge]

waiting for the crypto writeups

i give up

[indigo stag]

I don’t get the hype on this library, I don’t think it would provide anything useful for the chall 🤷‍♂️

[floral reef]

same

my script just ran for 10 mins and I quit

[vernal crater]

but how else do we copy paste their code 😒

[indigo stag]

you had to copy paste and run it 2 weeks ago 😅

[vernal crater]

what if i just enjoy watching my ram dump to pagefile 😄

nice way to spend a saturday afternoon

[analog stump]

smh shoulda just released the baby crypto anyway

at least give us some sense of happiness

[deep sedge]

someone wanna teach me LLL

[wary geode]

poortho is good at taking L

[prime eagle]

L

[surreal coral]

it's just continued fractions

[latent tiger]

is that a hint ? , cuz i've already wasted about 1 hr on continued fraction

[surreal coral]

Nah mathematically LLL in 2 dimensions is just continued fractions

[latent tiger]

may the entire country of ireland roasts you for not giving hint

[deep sedge]

i thought defund's question would be easy

i was wrong

[upbeat hazel]

oof

before or after release?

[deep sedge]

both

easy as in low amount of solves easy not some baby question

i know what to do just not how to do it

[upbeat hazel]

o

[tranquil bluff]

at damctf, we pride ourselves on pushing flag guessers to the next level

[twilit narwhal]

P A I N

that is all

[full temple]

more crypto?

[surreal coral]

garbled

[prime eagle]

lmao

we have nonzero crypto in our next release i think?

[eager prism]

👀

how many papers that are older than me does it require reading

[prime eagle]

no idea, ididnt write it

[neat shard]

why so few solves for crypto, highest solved has 4 solves

[upbeat hazel]

no papers, I made it up

[eager prism]

o nice

[surreal coral]

b/c our easy crypto turned out to already be in a previous ctf

[neat shard]

oof

[upbeat hazel]

well

I took params from somewhere

[full temple]

crypto is easy so everybody leaves it for the dessert

[upbeat hazel]

but w/e

[eager prism]

yes, i've only spent 7 hours on one challenge

[floral reef]

lmao me too

[eager prism]

although 2 of those is just waiting for things to run lol

[floral reef]

digging through standford 2009 libraries

[floral reef]

how long did you wait?

[eager prism]

2 hours and counting 😩

[floral reef]

did it finish?

I stopped at 30 mins

[eager prism]

not yet lol

[floral reef]

Accidently restared my pc

[eager prism]

actually i had another script which was slated to finish in around 80 days

[floral reef]

lol

[eager prism]

so if you count that then i've run scripts for around 4 hours

i also found out that some time ago when i reorganized my ctf stuff, it broke my sage install

[floral reef]

yeah is this the intended solution or are we in the wrong place

[eager prism]

so i reinstalled it

that was fun

idk

watch this script terminate with no result

[floral reef]

same. my linux thought it was cool and fukd the whole path up

[upbeat hazel]

have you tested it localy with smaller values?

[eager prism]

mm yes

i'm like

[upbeat hazel]

what's your expected time

[eager prism]

sort of sure that it works?

[floral reef]

3 months

[eager prism]

it's around 70% done 👀

[upbeat hazel]

:

[floral reef]

or like 9 hours

[upbeat hazel]

oh nice

[floral reef]

but that is not intended right?

[upbeat hazel]

no

[eager prism]

tbh if i removed the prints it would probably be a bit faster

but im impatient

and i need to know that its doing something

[upbeat hazel]

mental health is important

the prints are worth sth

[floral reef]

sth?

[upbeat hazel]

something*

[eager prism]

hm

when you guys solve cryptos based on papers

do you like

actually read the reasoning behind the algorithms

[floral reef]

just a matter of finding the paper

[upbeat hazel]

if the paper is clear

and the algorithm is clear

no

most important thing is flag 🙂

[eager prism]

lol

[floral reef]

most important thing is to learn

:3

[eager prism]

for somebody who claims to specialize in crypto i have an awful math background

currently taking alg2/trig

[upbeat hazel]

high school?

[eager prism]

mm ye

[upbeat hazel]

nice

[surreal coral]

(no longer needed) Can everyone who solved plagiarism DM me your solve scipt?

[eager prism]

👀

uh oh my script should be terminating now but it isn't

sad!

[solid venture]

sad!

[eager prism]

oh wait its gotta do like 20 massive operations

[floral reef]

🙏

[eager prism]

wait im dumb

i'm actually only 2/3 of the way there

😩

[solid venture]

sad!

[marsh silo]

..

[floral reef]

Does removing prints help with speed

[deep sedge]

yes

[floral reef]

by how much?

[deep sedge]

depends how many print statements you have

[floral reef]

well like 1 or 2

[solid venture]

if you're printing in a loop

[tight timber]

i mean, it won't hurt your performance

[solid venture]

it can be incredibly slow

the rule of thumb is if your terminal can catch up to your print statements then it probably isn't impacting performance very much

[floral reef]

well it goes 1.8 s for every print

[solid venture]

yeah it probably won't impact performance much

[floral reef]

Ill just exclude it tbh

[eager prism]

mm

script terminated while i was eating lunch

[upbeat hazel]

oh nice, you solved!

also I just realized, we don't have an interactive server on a crypto chall yet

[fleet epoch]

yet

[crimson cape]

[vernal crater]

when the garbled script finally reaches the end, and promptly crashes just before printing because you compared a list to an integer

😢

[upbeat hazel]

damn

[torn scroll]

mfw i can't do crypto bc i have algebra hw

mfw i can't do the algebra i want to do and instead need to do the algebra im assigned to do

[glacial cobalt]

SAME

except my math hw is not algebra

but eh

[solid venture]

flex

[glacial cobalt]

same difference

[wary geode]

python -i 🙂

[torn scroll]

i also have topology if that counts :(

[quartz heath]

i read the docs for python -i while makign ti1337plusce and mind blown

[glacial cobalt]

I think any math counts

[quartz heath]

so much wasted time

[indigo stag]

will benaloh be kind enough to concede

btw new release should be now?

[fleet epoch]

yea now-ish

[upbeat hazel]

my chall is live

[glacial cobalt]

ono not the SSSSS

the power of the S too strong

[deep sedge]

quantum zzz

[full temple]

jupyter lab 🙂

[surreal coral]

spyder

[torn scroll]

lwe.py

yeah i'll pass

[upbeat hazel]

😭

[sullen oar]

Are we able to get a runtime bound on newcrypt?

[solid venture]

[0,infinity)

[sullen oar]

As in like, does the solution script run within a couple minute sitting

[halcyon belfry]

@compact jacinth

[compact jacinth]

Solution should be solvable in reasonable time. My solution takes around 20-30 minutes in running (but obviously this may differ depending on hardware specs).

[solid venture]

inb4 tux's sol takes 20-30 mins on google's supercomputers

[full temple]

512-bit primes would be more fun 😦

[tight timber]

sol takes 20-30 minutes on quantum computer

[full temple]

have people noticed something weird in garbler?

[surreal coral]

not that i'm aware of

[compact jacinth]

hellman 🙏

[surreal coral]

hellman 🙏

[maiden narwhal]

Oof newcrypt v2 solved

Congrats

[torn scroll]

garbled took more braincells than i have available

i think im gonna stop doing crypto for now

[neat shard]

i have a solution to plagiarism that runs in 239.25925926 days

[eager prism]

👍

[lunar forge]

F

[eager prism]

just rent a supercomputer

[torn scroll]

u solved it in theory

that's all that matters

[vernal crater]

admin for garbled?

[halcyon belfry]

I presume @surreal coral is asleep

[clever wolf]

I don't know, it's not that early in Ireland

[indigo stag]

now would be a good time to make benaloh lcg parameters public

[upbeat hazel]

😆

I lost them

[merry sedge]

:-/ crypto is pretty tough

[surreal coral]

crypto is pretty
Aww thx

[solid venture]

don't encourage the crypto people like that

[mint rune]

where is the noob stuff??

[wary geode]

have you tried newcrypt

[winged ibex]

Are you joking ? I am on it since yesterday... I hope I'll solve it in a few hours but it's pretty hardcore

[mint rune]

still am a noob i don't know my way around py without the how

[wary geode]

at kox lol yea

the crypto is pretty brutal

[surreal coral]

we had eZ planned but it turnd out to be a duplicate

[solid venture]

most easy challs are duplicates

[upbeat hazel]

no but like

it was a total duplicat

for something weird

[wary geode]

the flavortext was different!!!

[upbeat hazel]

o yea

my flavortext was better

that's about it

[surreal coral]

it was a neat concept tho

[mint rune]

gotcah so fuck the noobs gotcha

right

right???

[upbeat hazel]

have you solved babier csp?

[mint rune]

@wary geode

yep same story

[upbeat hazel]

you wanna hop in a voice call

and we can walk through babier csp

rejected 😭

[mint rune]

not really i want to do it on my own like everybody else. or at least a correct heading that doesn't have some higher level stuff

[upbeat hazel]

right now you just seem to be complaining though

if you need some push/guidance on the challenge, you can DM an organizer

and we can try to do that minimally

do you want to solve babier csp?

[mint rune]

yes am complaining i had high hopes for this but none of it is for a noob. and i did and they keep trying to tell me this is noob stuff i look at it then i understand how far away this is from noob

[upbeat hazel]

signature-sheep-scheming-signature-schemes blood!

[mint rune]

and yes i do want slove it on my own

[upbeat hazel]

yeah I'll grant you that the crypto category is not noob stuff

that's my fault

[marsh sundial]

Didn't even include standard babyRSA. I thought that was a requirement to be called a CTF 😆

[upbeat hazel]

hmm

did justctf have babyrsa 😛

[violet coral]

That just means the writeups will be interesting 👀

[upbeat hazel]

👀

i should start writing that

[marsh sundial]

Fair point. I'll have to run my own CTF where the entire crypto category is just the same RSA challenge with varying number of factors of N.

[violet coral]

factordb go brrrrrrrr

[vernal crater]

benolah admin?

[upbeat hazel]

hi

[surreal coral]

Plagiarism is baby RSA, we even link to a solution script

[latent tiger]

ofc it is baby rsa , i'm not saying it's hard but it's computational intensive

[surreal coral]

Just wait a week to run and get the flag, ez

[latent tiger]

i have 1st part of plaintext next part is running so slow :pephands:

[full temple]

the last part is }

[latent tiger]

https://tenor.com/view/pepe-cry-reading-pepe-the-frog-sad-gif-17607942

[surreal coral]

[eager prism]

a week? my first script would have taken 12 weeks

[marsh sundial]

Again a strong argument against my joke. I'll switch it up to "didn't even include babyROT13 in crypto category". How about that? 😛

[surreal coral]

sanity was rot26

[eager prism]

[marsh sundial]

and now?

[latent tiger]

was your 1st script about bruteforceing flag ?

[eager prism]

idk maybe my computer is just slow

it is around 8 years old after all

[surreal coral]

maybe your code is just slow