Understanding performance of masks: test?a^6 vs. ?a^6test vs. ?a^6

[fsdafsadfsdsdaf]

Hey there,

on the same hardware

?a?a?a?a?a?a --- takes ~ 10 Sek at > 70 GH/s
?a?a?a?a?a?atest --- takes ~ 10 Sek at > 70 GH/s
test?a?a?a?a?a?a --- takes about 25 Min at only ~ 500 MH/s

why is it so slow if a prefix is known? It is more common to know a prefix than a suffix, isn't it? - at least that is what I observed.
-S does not improve the performance (it even slows it down quite a lot)

Thanks

[penguinkeeper]

It's a known problem with the way that Hashcat distributes work, it uses the first 4 characters to parallelise, so when the first 4 characters has a large keyspace, it can distribute work across the thousands of GPU cores, but when it's only a single candidate - "test", Hashcat really struggles to fully make use of the whole GPU, causing very low speeds as you're seeing there. It's a known, architectural problem with Hashcat that unfortunately can't (and won't) be solved

[Therear]

hi there
I don't know whether my suggestion was too late, but I still replyfall guys online
GPU optimization: Hashcat uses GPU-optimized kernels that work best when the variable part of the mask is at the end, not the beginning.
The constant part at the end (like in ?a?a?a?a?a?atest) lets Hashcat re-use precomputed values efficiently.
The constant part at the beginning (like in test?a?a?a?a?a?a) means Hashcat has to recompute more per candidate, breaking optimization.