This is great. I’m actually trying to unlock the 80G in a CMP 170HX the software way for the last 2 weeks (still no luck).

I just don’t know if I’m brave enough (and knowledgeable enough) to go this blown fuses way. Seems a lot more risky.

P.S. I now see this is for the 8GB model, which is insane. The 10G one is the one that until now was proven to have 80G, but never fully accessible through CUDA (it was accessible using CPU memory copy though - much slower).

▲

roger_ 21 hours ago | parent [-]

Was the 80GB actually proven on the 10GB model? I thought it was a mirror of the first 40GB.

	▲	notenchange 32 minutes ago \| parent [-]
		According to this paper, yes: https://www.researchgate.net/publication/408132536_A_Canary_... But I can't see any independent confirmation and having personally tried to contacted the paper author I received no response. So, I'm not sure the paper findings actually hold. I'm now on the process of investigating this and, if I achieve any conclusive result, either for or against, submit a new paper on the subject.