mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Data

Reply
 
Thread Tools
Old 2022-12-06, 11:08   #1
aperson1
 
Dec 2022

5 Posts
Smile Getting under 20k unfactored exponents in 229m

Hey, I'm a bit new to the forum but not to the project- Sorry if this isn't the appropriate place to put this (or shouldn't have been a new thread) - but this thread isn't about about any of that!

A few months ago, I'd been going through the trial factoring range with the herd- in the mid-200-millions right about now, and noticed that the recently-unreserved 229m range was pretty close to a suspicious milestone: less than 20,000 unfactored exponents. It seems you folks have a keen interest in this milestone, judging by other threads, but I had no idea of this at the time.

Trial factoring work mostly ended several months ago, too, ending up with just 13 unfactored exponents to go to break the barrier. Now, I'm just a hobbyist using my personal computers, so even just 13 factors seemed like an insurmountable peak when I started trying to understand P-1 testing back in June.

Well, almost half a year later I'm happy to announce that I'm more than halfway to my completely arbitrary milestone! I've managed to factor 8 new numbers (229000159, 229000201, 229000963, 229001413, 229001477, 229001533, 229002379, and 229002493) and have the rest of the 229.002 range reserved to complete within the next couple weeks. Of course, I can't readily guess when I'll hit the milestone, but assuming average luck it should definitely be in the next couple of months.

Interestingly, I've been getting unreasonably good luck finding factors... I've been getting them roughly 15% of the time (8 from 52 tests) compared to a predicted ~6.5% success rate. Now, it's not the most significant deviation from the norm, but apparently the chance of getting this lucky is just ~4%.

I was hoping it would be the first range between the current 110s-million and the 330s-million exponents to get under 20,000 unfactored exponents, but an apparently coordinated effort at 168m got there first, as well as the surprisingly composite 254m from just trial factoring work. Looks like 259m will probably be following soon as well. Still, third place is a bronze medal!

Anyway, just to be clear, I don't need any particular help with factoring this range. This post is more of a declaration of intent + a modest in-progress success story from a modest user. I won't try and stop anyone who really wants go to at it, but I'd appreciate being able to get #20,000 knocked off at least!
aperson1 is offline   Reply With Quote
Old 2022-12-06, 15:02   #2
Uncwilly
6809 > 6502
 
Uncwilly's Avatar
 
"""""""""""""""""""
Aug 2003
101×103 Posts

32×11×109 Posts
Default

Welcome to the forum.

Are you using GPU(s) for this? Do you have plans for when this goal is complete?
Uncwilly is offline   Reply With Quote
Old 2022-12-06, 15:56   #3
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

132×31 Posts
Default

Quote:
Originally Posted by aperson1 View Post
Hey, I'm a bit new to the forum but not to the project.

A few months ago, I'd been going through the trial factoring range with the herd- in the mid-200-millions right about now, and noticed that the recently-unreserved 229m range was pretty close to a suspicious milestone: less than 20,000 unfactored exponents. It seems you folks have a keen interest in this milestone, judging by other threads, but I had no idea of this at the time.

Trial factoring work mostly ended several months ago, too, ending up with just 13 unfactored exponents to go to break the barrier. Now, I'm just a hobbyist using my personal computers, so even just 13 factors seemed like an insurmountable peak when I started trying to understand P-1 testing back in June.

Well, almost half a year later I'm happy to announce that I'm more than halfway to my completely arbitrary milestone! I've managed to factor 8 new numbers ... I can't readily guess when I'll hit the milestone, but assuming average luck it should definitely be in the next couple of months.

Interestingly, I've been getting unreasonably good luck finding factors... I've been getting them roughly 15% of the time (8 from 52 tests) compared to a predicted ~6.5% success rate. Now, it's not the most significant deviation from the norm, but apparently the chance of getting this lucky is just ~4%.

I'd appreciate being able to get #20,000 knocked off at least!
Good to hear from you.
The under 20 Million or 20 Thousand or 2 Thousand, depending on the depth was the pet project I started several years ago.
And as you noticed we got up to 110M with a LOT of help form a lot of interested participants.

As I understand the prescribed level of TF and P-1 will "eventually" get all the remaining 100K ranges under 2,000 unfactored but it will be a while and I know its fun for some like yourself to take on a pet range. Have Fun!!!

You mentioned that TF worked ended a few months ago but in fact that was only for the 74 bit range.
If you look here and notice where the yellow band is you will notice that eventually your range will be factored up to at least 79 bits, probably more because GPUs are SOOOO fast. You can expect to see the TF crew back in your range early in the New Year. Hopefully, you finish before they get there.

If you have a GPU that you can use for TF it will find factors quickly at the current bit levels and beyond.
Otherwise the P-1 you are doing is valuable too.

As for the success rate you are seeing it might be luck, but after many years of P-1 my average success rate is always above the expected too. I think their calculator here might be low.

WELCOME and ENJOY!!!
petrw1 is offline   Reply With Quote
Old 2022-12-07, 02:02   #4
aperson1
 
Dec 2022

510 Posts
Default

Thanks for the welcomes and kind words, everyone.

Quote:
Originally Posted by Uncwilly View Post
Welcome to the forum.

Are you using GPU(s) for this? Do you have plans for when this goal is complete?
I didn't realize that GPU work was quite so efficient until you all mentioned it. I'll definitely have another try figuring out how to set it up! It sounds like my 1060 should work decently enough...

After I finish this, I'm not entirely sure. I might target other regions close to the 2* barriers (either in existing miniprojects or high-level exponents like this), or focus on areas of the big mersenne.ca graph with unusually few factors (like 148.9m) If the speeds quoted for other GPU factoring apply well to mine, then I might have to figure that out quite soon...
aperson1 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Getting <2k unfactored exponents for 108.3M Zhangrc Data 63 2022-10-28 09:31
COMPLETE!!!! Thinking out loud about getting under 20M unfactored exponents petrw1 Data 1414 2022-08-31 09:50
Interesting properties about Mersenne(-related) exponents and Wagstaff(-related) exponents sweety439 sweety439 0 2021-06-24 02:29
Unreserving exponents(these exponents haven't been done) jasong Marin's Mersenne-aries 7 2006-12-22 21:59
Question on unfactored numbers... WraithX GMP-ECM 1 2006-03-19 22:16

All times are UTC. The time now is 13:01.


Fri Dec 9 13:01:13 UTC 2022 up 113 days, 10:29, 0 users, load averages: 1.14, 1.13, 1.10

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.

≠ ± ∓ ÷ × · − √ ‰ ⊗ ⊕ ⊖ ⊘ ⊙ ≤ ≥ ≦ ≧ ≨ ≩ ≺ ≻ ≼ ≽ ⊏ ⊐ ⊑ ⊒ ² ³ °
∠ ∟ ° ≅ ~ ‖ ⟂ ⫛
≡ ≜ ≈ ∝ ∞ ≪ ≫ ⌊⌋ ⌈⌉ ∘ ∏ ∐ ∑ ∧ ∨ ∩ ∪ ⨀ ⊕ ⊗ 𝖕 𝖖 𝖗 ⊲ ⊳
∅ ∖ ∁ ↦ ↣ ∩ ∪ ⊆ ⊂ ⊄ ⊊ ⊇ ⊃ ⊅ ⊋ ⊖ ∈ ∉ ∋ ∌ ℕ ℤ ℚ ℝ ℂ ℵ ℶ ℷ ℸ 𝓟
¬ ∨ ∧ ⊕ → ← ⇒ ⇐ ⇔ ∀ ∃ ∄ ∴ ∵ ⊤ ⊥ ⊢ ⊨ ⫤ ⊣ … ⋯ ⋮ ⋰ ⋱
∫ ∬ ∭ ∮ ∯ ∰ ∇ ∆ δ ∂ ℱ ℒ ℓ
𝛢𝛼 𝛣𝛽 𝛤𝛾 𝛥𝛿 𝛦𝜀𝜖 𝛧𝜁 𝛨𝜂 𝛩𝜃𝜗 𝛪𝜄 𝛫𝜅 𝛬𝜆 𝛭𝜇 𝛮𝜈 𝛯𝜉 𝛰𝜊 𝛱𝜋 𝛲𝜌 𝛴𝜎𝜍 𝛵𝜏 𝛶𝜐 𝛷𝜙𝜑 𝛸𝜒 𝛹𝜓 𝛺𝜔