Is it a wrong value for mnt6_final_exponent_last_chunk_abs_of_w0 in mnt6_init.cpp?

[3for]

I'm following the paper to learn the mnt4 and mnt6 curves, it's said that t6=2-t4, the t4 value is 689871209842287392837045615510547309923794945，so the t6 value should be -689871209842287392837045615510547309923794943 instead of -689871209842287392837045615510547309923794944. I'm wondering if the value for mnt6_final_exponent_last_chunk_abs_of_w0 in mnt6_init.cpp is wrong.

[ValarDragon]

/cc @howardwu @Pratyush

Is there an analog of this we can check in Zexe?

[Pratyush]

https://github.com/scipr-lab/zexe/blob/38b6f6f2c6b9a6a1b5e8eb90b1287c98b06520d1/algebra/src/mnt6_298/curves/mod.rs#L45 ?

[ValarDragon]

Thanks! The mnt6_final_exponent_last_chunk_abs_of_w0 value matches whats in Zexe's mnt 6 then.

Additionally the mnt4 exponents match whats in Zexe. I'm not familiar enough with the MNT cycle to know if there is a bug here, in the paper, or if there is some optimization / alternate curve being used.

[3for]

Maybe I misunderstood the relation between t4 and t6？

sage: hex(689871209842287392837045615510547309923794944)
'1eef5546609756bec2a33f0dc9a1b671660000'

sage: hex(689871209842287392837045615510547309923794943)
'1eef5546609756bec2a33f0dc9a1b67165ffff'