mersenneforum.org  

Go Back   mersenneforum.org > Search Forums

Showing results 1 to 25 of 1000
Search took 0.35 seconds.
Search: Posts Made By: frmky
Forum: And now for something completely different 2022-11-29, 21:35
Replies: 217
Views: 23,655
Posted By frmky
Thanks. I agree that copying those definitions...

Thanks. I agree that copying those definitions isn't a good idea. I was hoping that it could be done using the documented public GMP interface relatively easily without loss of speed. Too much to ask...
Forum: And now for something completely different 2022-11-29, 20:20
Replies: 217
Views: 23,655
Posted By frmky
Can this be relatively easily rewritten with...

Can this be relatively easily rewritten with mpz_size(), mpz_limbs_modify(), and mpz_limbs_finish() to remove the dependency on gmp-impl.h, or is there something I'm missing in my quick review of the...
Forum: Cunningham Tables 2022-11-27, 18:13
Replies: 66
Views: 23,743
Posted By frmky
Does that work with lasieve5? Last I tried it ran...

Does that work with lasieve5? Last I tried it ran significantly slower and sometimes stalled, but I might have been using lasieve4 at the time. I'll try it again. I also plan to try using 36-bit LPs...
Forum: NFS@Home 2022-11-24, 04:41
Replies: 180
Views: 16,820
Posted By frmky
250__853_29m1 is factored. biMidjXs

250__853_29m1 is factored.

biMidjXs
Forum: Cunningham Tables 2022-11-23, 18:25
Replies: 66
Views: 23,743
Posted By frmky
And it's done. Another surprisingly smooth run. ...

And it's done. Another surprisingly smooth run.

KPbdQgwN
Forum: Cunningham Tables 2022-11-22, 03:02
Replies: 42
Views: 4,335
Posted By frmky
Presuming 2,1097+ goes ok, I'll do 2,2194L&M...

Presuming 2,1097+ goes ok, I'll do 2,2194L&M next. That'll give you some time for ECM.
Forum: GpuOwl 2022-11-21, 20:18
Replies: 2,866
Views: 440,648
Posted By frmky
Likely it will. While the RX 7900 XTX is overall...

Likely it will. While the RX 7900 XTX is overall a much faster card (a MI60 is based on the same architecture as the Radeon VII) the 1/16 FP64/FP32 ratio on the RX 7900 XTX slows it down for gpuOwl.
Forum: GpuOwl 2022-11-21, 02:20
Replies: 2,866
Views: 440,648
Posted By frmky
32GB MI60's are available for under $1000. But...

32GB MI60's are available for under $1000. But they are passively cooled so you'd have to deal with rigging a cooler for it.
Forum: GpuOwl 2022-11-20, 21:17
Replies: 2,866
Views: 440,648
Posted By frmky
But to the user they appear as two distinct GPUs...

But to the user they appear as two distinct GPUs with separate memory spaces. Data transfer between the two over Infinity Fabric is much slower than HBM2, and applications must be coded to support...
Forum: GpuOwl 2022-11-20, 20:15
Replies: 2,866
Views: 440,648
Posted By frmky
A MI250 is really two MI210's in the same...

A MI250 is really two MI210's in the same package. Since gpuOwl doesn't support multiple GPUs, a single run on a MI250 will run on one of them and give the same times as a MI210. You can just run two...
Forum: GpuOwl 2022-11-20, 18:18
Replies: 2,866
Views: 440,648
Posted By frmky
Also the MI250X (and MI250) are really two GPUs...

Also the MI250X (and MI250) are really two GPUs packaged together, like the nVidia K80. AMD hides this a bit in their marketing. So expect only half of that performance in a single gpuOwl run. And...
Forum: NFS@Home 2022-11-20, 17:56
Replies: 137
Views: 11,405
Posted By frmky
12p7_256 is factored. L48pgDbt

12p7_256 is factored.

L48pgDbt
Forum: NFS@Home 2022-11-20, 04:53
Replies: 180
Views: 16,820
Posted By frmky
5p3_418 is done. 1miQ4Dzc

5p3_418 is done.

1miQ4Dzc
Forum: Cunningham Tables 2022-11-19, 22:25
Replies: 66
Views: 23,743
Posted By frmky
2,1091+ is in LA now. linear algebra completed...

2,1091+ is in LA now.
linear algebra completed 226672 of 115553837 dimensions (0.2%, ETA 78h56m)
Forum: GpuOwl 2022-11-18, 07:05
Replies: 2,866
Views: 440,648
Posted By frmky
Here's a couple more: nVidia A40 ./gpuowl...

Here's a couple more:

nVidia A40
./gpuowl -log 10000 -iters 200000 -prp 77936867
20221118 00:43:41 GpuOwl VERSION v7.2-112-gd6ad1e0
20221118 00:43:41 GpuOwl VERSION v7.2-112-gd6ad1e0
20221118...
Forum: GpuOwl 2022-11-18, 04:18
Replies: 2,866
Views: 440,648
Posted By frmky
Generally you want to run binaries compiled using...

Generally you want to run binaries compiled using CUDA 11.8 (or later) toolkit on an RTX 4090 to take full advantage of the architecture. And the FP64 speed on that card is 1/64 FP32, so running LL,...
Forum: Marin's Mersenne-aries 2022-11-15, 01:00
Replies: 31
Views: 81,942
Posted By frmky
Just as a comment, those than I (Greg) did were...

Just as a comment, those than I (Greg) did were run on nVidia Tesla hardware with ECC enabled both on the system and GPU memory. Doesn't make them error-proof but I would be very surprised if there's...
Forum: Cunningham Tables 2022-11-13, 08:19
Replies: 66
Views: 23,743
Posted By frmky
Not yet. With the 14 day deadline I'm still...

Not yet. With the 14 day deadline I'm still waiting for more stragglers to find their way back.
Forum: Factoring 2022-11-13, 04:28
Replies: 147
Views: 21,240
Posted By frmky
The mp library this uses, CGBN...

The mp library this uses, CGBN (https://github.com/NVlabs/CGBN), only supports up to 32K bits.
Forum: FactorDB 2022-11-10, 16:46
Replies: 588
Views: 117,205
Posted By frmky
Perhaps it was reindexed? The new entry says it...

Perhaps it was reindexed? The new entry says it was created a couple of days ago.
http://factordb.com/index.php?id=1100000003950159299
Forum: NFS@Home 2022-11-10, 16:38
Replies: 180
Views: 16,820
Posted By frmky
8p3_882M is factored. agAuvfEq

8p3_882M is factored.

agAuvfEq
Forum: NFS@Home 2022-11-10, 16:32
Replies: 32
Views: 3,975
Posted By frmky
125642407351_23m1 has split. mp12yFB6

125642407351_23m1 has split.

mp12yFB6
Forum: NFS@Home 2022-11-09, 22:49
Replies: 137
Views: 11,405
Posted By frmky
C190_785232_11573 is done. JPpUw7gw

C190_785232_11573 is done.

JPpUw7gw
Forum: NFS@Home 2022-11-09, 00:09
Replies: 1,943
Views: 193,264
Posted By frmky
Thanks for reviewing these pages! Grant number is...

Thanks for reviewing these pages! Grant number is the same. It's been renewed each time since 2008. But ACCESS has replaced XSEDE. I quickly edited the contributions link. Please let me know who I...
Forum: Msieve 2022-11-04, 19:25
Replies: 9
Views: 1,094
Posted By frmky
I've now merged these changes into the...

I've now merged these changes into the msieve-lacuda-nfsathome branch. The only remaining uses of pthreads are a timeout thread created with the -d option and during GPU polynomial selection....
Showing results 1 to 25 of 1000

 
All times are UTC. The time now is 14:30.


Thu Dec 1 14:30:35 UTC 2022 up 105 days, 11:59, 1 user, load averages: 1.03, 1.08, 1.05

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