Getting others to do the work on exponents I like (was: Trial Factoring Progress)

[tuckerkao]

Originally Posted by Uncwilly

Remember this post?
Real random things can be quite lumpy.
Patience grasshopper.

Only 109 days remain until AMD Threadripper 5970X launches, so I'll patiently wait for that, then I'll have the computing powers I need at latest by mid-Jan 2022.


If these are the highly composite grounds,

Z48,622,147 <- M168753223
Z48,62Ӿ,127 <- M168767023
Z48,65Ӿ,897 <- M168830323
Z48,673,Ɛ8Ɛ <- M168860123
Z48,6Ӿ6,20Ɛ <- M168926123

then the highly primitive ground should be at Z48,3Ɛ7,237 <- M168202123

I'll try do it all myself, see if I'm correct this time.

[Viliam Furik]

Originally Posted by tuckerkao

Only 109 days remain until AMD Threadripper 5970X launches...

Probably that many, maybe more, maybe less...

[drkirkby]

Originally Posted by tuckerkao

Only 109 days remain until AMD Threadripper 5970X launches

I'm reliably informed that AMD are also working on the development of another CPU which will launch next year. That will have 128 MB of level 4 cache using write only memory. I can't remember the part number of that though.

[tuckerkao]

Originally Posted by drkirkby

I'm reliably informed that AMD are also working on the development of another CPU which will launch next year. That will have 128 MB of level 4 cache using write only memory. I can't remember the part number of that though.

AMD Zen 3 V-Cache, mid-2022, but that's even further away. AMD Threadripper 5970X is faster for the M168M exponents if I want to test them using Prime95 at the same time with another PRP using Radeon RX 6900 XT on Gpuowl.

Zen 4 Ryzen 6950X and RDNA 3 Radeon RX 7900 XT, late-2022, Zen 4 Threadripper 6970X mid=2023 or late-2023.

[Viliam Furik]

Originally Posted by drkirkby

I'm reliably informed that AMD are also working on the development of another CPU which will launch next year. That will have 128 MB of level 4 cache using write only memory. I can't remember the part number of that though.

I think you have made a mistake here.

How on earth could write-only memory be useful for cache? That's like archiving documents that aren't supposed to be accessed, such as documents regarding a mother giving away her child right after birth.

Once a pair of law students came to my high school to give a talk about something, probably related to the law, but I almost only remember asking - "Why are those documents archived if they are to never be seen again? And why are they even created then?". I didn't get a good answer...

This type of archival reminds me of pirate chests, filled with immense treasure, but cursed or something. Why would you want to store your treasure, only to never see it again, and don't let anyone have a single piece of it? Unless you are obsessed by greed and/or enough maliciousness to let the people think they can get it, only to get cursed; there is no point to this. You are better off spending it or simply giving it away and not caring anymore.


That's was a bit too long a detour, but the point is, unless I am wrong, a write-only cache is a useless and stupid idea. Thus I think you have made a mistake.

[Uncwilly]

If the cache is fast write only, and then slow transfer to main RAM, it could free the CPU to do other things. But, this all is a long thread and this is off topic to the main subject.

[tuckerkao]

Originally Posted by Uncwilly

If the cache is fast write only, and then slow transfer to main RAM, it could free the CPU to do other things. But, this all is a long thread and this is off topic to the main subject.

drkirkby must misidentified this thread as "Expectations of Ryzen Threadripper 6000 Series" in the Hardware section.

Just let UncWilly move the 4 to 5 posts above to that thread.

[Viliam Furik]

Originally Posted by tuckerkao

drkirkby must misidentified this thread as "Expectations of Ryzen Threadripper 6000 Series" in the Hardware section.

Just let UncWilly move the 4 to 5 posts above to that thread.

You are showing signs of being crazy... You are basically telling Uncwilly to let Uncwilly move the posts. Or at least that's what it looks like.

[tuckerkao]

Originally Posted by LaurV

I deleted his post(s) because he was talking totally nonsense about things I posted, like "M11.44M is a double double" (???) like a bot would do (i.e. looking at the patterns on my post and reply things with the same pattern but having nothing to do with the subject semantically).

Normally I would ignore a post like that, or take it as a joke (and I have to admit that it was a "cute" joke, with the double-double").

Thus I'm going to run a PRP test of a M168,***,*23 candidate that will cost 1,144 Ghz days with a dozenal 7 ender, so M168202123

It's year 2021 anyway, so I must finish this exponent.

UncWilly has 1 clear rule for me, if my reply to any other thread including M168M, it must be done here.

Originally Posted by Uncwilly

They are free to not indulge your misguided delusions. Their kit, their choice. Stop complaining that people are not helping you with your numerology. And that shadow that you see might be the hammer.

It's probably the delusion from the dozenal numerology telling me that M168202123 has around 1/144 chance being a prime.

[tuckerkao]

M168202129 doesn't have factors below 2^75 unlike M168377329, thus my current exponent has higher chance surviving the larger P-1 bounds.

I've written my name on around 400 exponents now, ought to keep expanding my coverage - https://www.mersenne.ca/status/tf/0/20210201/4/16800

Originally Posted by Viliam Furik

My work on the 49.6M range, TF 74-75, is stopped for now. I may continue later, but for now, it should be considered free.

Sorry.

Resume the work with range +118.6M if possible.

[kruoli]

Originally Posted by tuckerkao

[...] Perhaps if Ben Delo appears in my dream, he'll tell me the correct exponent to run [...]

This is not an esoteric forum.