mersenneforum.org  

Go Back   mersenneforum.org > Search Forums

Showing results 1 to 25 of 1000
Search took 0.39 seconds.
Search: Posts Made By: henryzz
Forum: Linux 2021-09-21, 08:36
Replies: 6
Views: 146
Posted By henryzz
Thank you. This works with very low memory...

Thank you. This works with very low memory usage(not even 1GB).

The only issue is that it outputs in a different format:
[('comp1', count1), ('comp2',count2)]while the previous method outputs:...
Forum: Linux 2021-09-20, 17:36
Replies: 6
Views: 146
Posted By henryzz
I have used those options(and forgot about them...

I have used those options(and forgot about them when writing the post). This reduces the memory usage down to the buffer + the file size. It slows it down significantly though.


The number of...
Forum: Linux 2021-09-20, 15:17
Replies: 6
Views: 146
Posted By henryzz
Low memory line counting in huge files

I have been running a program that outputs a line to a file every time it encounters a composite. Many of these lines are duplicated and I am interested in targeting the composites that are...
Forum: GPU Computing 2021-09-20, 12:19
Replies: 8
Views: 402
Posted By henryzz
The one you linked mentioned FP64 support at...

The one you linked mentioned FP64 support at least in the xe-hp variant. I would imagine that we will see 1/32 speed or similar dp support in the consumer variant as well for code compatibility.
Forum: Msieve 2021-09-19, 08:02
Replies: 52
Sticky: Msieve GPU LA
Views: 2,617
Posted By henryzz
I looked through the code a few years ago and...

I looked through the code a few years ago and found no issues. Lasieve4 is also fine although it is limited to 96 bit mfba/r.
Forum: Software 2021-09-09, 14:19
Replies: 7
Views: 286
Posted By henryzz
This would be a nice feature but far from...

This would be a nice feature but far from necessary as there are workarounds. If I recall correctly the code implementing this sort of thing is a mess in PFGW and rogue has avoided playing with it...
Forum: Factoring 2021-09-09, 14:06
Replies: 965
Views: 112,803
Posted By henryzz
I have finally had time to think further on this...

I have finally had time to think further on this and have a few more questions:

Am I correct in thinking that 5 being a Fermat prime made it easier to determine the optimal way of finding all the...
Forum: Hardware 2021-09-05, 21:00
Replies: 12
Views: 612
Posted By henryzz
I was under the impression that power usage was...

I was under the impression that power usage was O(f * V^2) of cause higher frequency requires higher voltage(perhaps giving O(f^3) in some ranges of f).
Forum: Factoring 2021-09-01, 21:11
Replies: 92
Views: 3,494
Posted By henryzz
My guess is that your gcc version may be too old....

My guess is that your gcc version may be too old. I would try the most recent version you can get your hands on. The easiest way may be to update your OS into a version that isn't end of life.
Forum: Factoring 2021-09-01, 09:28
Replies: 10
Views: 252
Posted By henryzz
That will affect efficiency not cause it to...

That will affect efficiency not cause it to crash.



Two ideas:

Can the ggnfs siever understand lss? At best it will ignore it I think.

We have had issues in the past with compilers...
Forum: Factoring 2021-08-29, 14:29
Replies: 15
Views: 1,015
Posted By henryzz
c169 1/8: 1 = 120118613191929471309914301089279...

c169 1/8: 1 = 120118613191929471309914301089279 (33 digits) * c137

c274 1/8: 2 = 510031915039841133936511192966368943 (36 digits) * c238

c99 1/8: 7 = 3671595612167512594604024710268314567 (37...
Forum: Factoring 2021-08-29, 07:08
Replies: 92
Views: 3,494
Posted By henryzz
Reducing S_BITS_PER_CALL has fixed it for me....

Reducing S_BITS_PER_CALL has fixed it for me. Thank you 😀
Forum: Factoring 2021-08-28, 22:40
Replies: 92
Views: 3,494
Posted By henryzz
The same thing. If I run less curves at...

The same thing.


If I run less curves at once it works. Possibly just that my gpu is pathetic (750 Ti):
echo "(2^499-1)/20959" | ./ecm -gpu -cgbn -sigma 3:1000 20000
GMP-ECM 7.0.5-dev...
Forum: Factoring 2021-08-28, 21:55
Replies: 92
Views: 3,494
Posted By henryzz
With Wraith's patch(with the header fixed) I now...

With Wraith's patch(with the header fixed) I now get:
echo "(2^499-1)/20959" | ./ecm -gpu -cgbn -gpucurves 3584 -sigma 3:1000 2000
GMP-ECM 7.0.5-dev [configured with GMP 6.2.99, --enable-asm-redc,...
Forum: GPU Computing 2021-08-28, 16:51
Replies: 375
Views: 52,431
Posted By henryzz
I have a more recent version of CUDA than you...

I have a more recent version of CUDA than you linked in(11.4 rather than 8). As such I tried to recompile. I got the following error:
/usr/local/cuda/bin/nvcc -g -O3 -I/usr/local/cuda/include -I....
Forum: Factoring 2021-08-28, 16:27
Replies: 92
Views: 3,494
Posted By henryzz
There still seems to be cudaDeviceSynchronize...

There still seems to be cudaDeviceSynchronize issues although I get it at a very different line number.



echo "(2^499-1)/20959" | ./ecm -gpu -cgbn -gpucurves 3584 -sigma 3:1000 20000
GMP-ECM...
Forum: Factoring 2021-08-25, 11:59
Replies: 92
Views: 3,494
Posted By henryzz
This is an impressive speedup. I assume there is...

This is an impressive speedup. I assume there is no chance of stage 2 being ported to run on CUDA? GPU memory was too small back when the old GPU code was written. I suspect that may still be the...
Forum: Conjectures 'R Us 2021-08-16, 22:00
Replies: 404
Views: 70,252
Posted By henryzz
What's the minimum effective n for BOINC for base...

What's the minimum effective n for BOINC for base 3? Presumably, we should be targeting getting it to that point so that future work can be done there. E.g. No point in getting it to 50k if BOINC...
Forum: Conjectures 'R Us 2021-08-15, 11:04
Replies: 979
Views: 118,367
Posted By henryzz
Also how base 1030 when it looks like base 2?

Also how base 1030 when it looks like base 2?
Forum: Prime Gap Searches 2021-08-12, 12:56
Replies: 261
Sticky: Prime Gap News
Views: 63,727
Posted By henryzz
I have just submitted my first 30 merit gap:...

I have just submitted my first 30 merit gap: 90240 30.742 47193529 * 2999#/2303070 -50488 to +39752
This knocks two gaps off...
Forum: Computer Science & Computational Number Theory 2021-08-09, 10:41
Replies: 19
Views: 2,256
Posted By henryzz
Unless I am missing something you could divide it...

Unless I am missing something you could divide it by 2 though.

The following may or may not be optimal(likely not)
106 / 2
53 - 1
52 / 4
13 - 1
12 / 3
4 / 2
Forum: Aliquot Sequences 2021-08-09, 07:13
Replies: 1,351
Views: 137,982
Posted By henryzz
I have taken 720 up to >100 digits and 90 digit...

I have taken 720 up to >100 digits and 90 digit composites for n up to 40.
Forum: Aliquot Sequences 2021-08-06, 17:23
Replies: 1,351
Views: 137,982
Posted By henryzz
I am going to take a shot at 6!=720

I am going to take a shot at 6!=720
Forum: Prime Gap Searches 2021-08-01, 15:12
Replies: 133
Views: 30,185
Posted By henryzz
Are you aware of the blog post for this paper?...

Are you aware of the blog post for this paper? You could try asking a question there(the author might have notifications for new comments)....
Forum: No Prime Left Behind 2021-08-01, 15:06
Replies: 599
Views: 122,829
Posted By henryzz
Have you considered hosting your servers on the...

Have you considered hosting your servers on the cloud? You might find that more reliable although there might be bit of work to transition it.
Showing results 1 to 25 of 1000

 
All times are UTC. The time now is 17:24.


Tue Sep 21 17:24:48 UTC 2021 up 60 days, 11:53, 0 users, load averages: 1.96, 1.91, 1.80

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, 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.