ECM users - version 30.9-30.13 (see post#316) - Page 8

joejoefla · 2022-10-03, 23:32

Quote:

Originally Posted by James Heinrich

I hope you're giving it more than 16MB to work with on a 32GB system.

Yeah, that's the thing. The default was 0.3GB when but the complaint was only happening if I left the memory setting alone.
If you try it you should get the same error.

joejoefla · 2022-10-03, 23:42

Quote:

Originally Posted by kriesel

Or 300MB ~0.3GB. Downloaded defaults are not enough for polynomial-paired P-1, for wavefront DC or first test exponents. joejoefla doesn't say what exponent is being attempted so it's not practical to give guidance for ECM memory requirements, beyond what is in the program's readme.txt, which appears to me to not consider the additional ram needed for polynomial pairing to reach its full potential. It takes lots of Mersenne-number-size buffers to do so, and that takes a lot of ram for stage 2.

George has stated that the program ships with allowed stage 2 ram set to a safely low value, so it does not risk sending new users' systems into page faulting and slowing down their other usage, alienating them early. If there's plenty of free available ram (at least several GB) on the system, it's probably safe to increase prime95 allowed ram by a few GBs.

ECM2=blah,1,2,2660069,-1,50000,5000000,350

That's the exponent I was testing. I didn't think to check the memory resources. I thought the program crashed for reasons other than memory usage until I looked closer at the log.

Prime95 · 2022-10-04, 02:24

Quote:

Originally Posted by joejoefla

Is it because when stage 2 starts there's a bug where memory resources isn't being read correctly?

Not too surprising. It's probably trying to run the old stage 2 code. I'll add the bug to my list of things that need fixing.

joejoefla · 2022-10-04, 02:33

Quote:

Originally Posted by Prime95

Not too surprising. It's probably trying to run the old stage 2 code. I'll add the bug to my list of things that need fixing.

Thanks!

Denial140 · 2022-10-04, 06:47

I had something similar happen when I first ran a curve with build 2. Since i forgot to change the memory settings from the default it crashed at the end of stage 1. I assumed this was because it was stopping because of insufficient memory, and hence trying to write a stage 2 savefile (and crashing as expected) - everything worked fine when there was enough memory allocated to run stage 2.

James Heinrich · 2022-10-04, 13:29

Quote:

Originally Posted by kriesel

George has stated that the program ships with allowed stage 2 ram set to a safely low value

Perhaps on first startup Prime95 could peek at total system RAM and make a better estimate at default configuration than a simple hardcoded 300MB. Maybe something like max(300, totalRAM / 16) or something like that which would still give a very conservative default but allow more than 300MB on systems with lots of RAM.

LaurV · 2022-10-05, 05:53

Got an angry message from 30.9b3 - don't know what it means, but the things work properly here - probably related to "I have to do a lot of code cleaning". Decided to let you know, just in case.

Click image for larger version

Name: angry.PNG
Views: 162
Size: 55.9 KB
ID: 27416

axn · 2022-10-05, 07:44

Quote:

Originally Posted by LaurV

Got an angry message from 30.9b3 - don't know what it means, but the things work properly here - probably related to "I have to do a lot of code cleaning". Decided to let you know, just in case.
Attachment 27416

Looks like a network issue. Something in the middle is rejecting network connection. Are you running from office network? Looks like the kind of error you get from a proxy or something ("Access Denied")

chris2be8 · 2022-10-05, 15:41

That looks like a network glitch truncated the response. So you only received part of the header for the web page but none of the body.

If it worked OK when it retried then don't worry about it.

I've had many similar problems when scraping HTML in my scripts. And written work rounds to retry errors later in most cases.

LaurV · 2022-10-07, 08:52

It repeats every 75 minutes. Ever since. I don't know if it is only cosmetic, and the connections works, or not, the assignments are taken, or not, or the results are reported correctly, or not, because there is no result to report, and it won't be for ~80 hours or so. Maybe it is only cosmetic (cleaning the code as George said).

Older version (renamed exe file in the same folder, using the same prime.txt/local.txt files) seems to work without any error, so it is not a network issue.

axn · 2022-10-07, 11:27

Quote:

Originally Posted by LaurV

It repeats every 75 minutes. Ever since. I don't know if it is only cosmetic, and the connections works, or not, the assignments are taken, or not, or the results are reported correctly, or not, because there is no result to report, and it won't be for ~80 hours or so. Maybe it is only cosmetic (cleaning the code as George said).

That html is not coming from the program. It is the response it is receiving when it calls primenet API. And it is not primenet doing that either. It is a middleman in the network.

Quote:

Originally Posted by LaurV

Older version (renamed exe file in the same folder, using the same prime.txt/local.txt files) seems to work without any error, so it is not a network issue.

Are you running a firewall? Antivirus? Anit-malware? These things can prevent "suspicious" network traffic. Do you remember ever having to whitelist P95 previously?

2022-10-05, 05:53	#84
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 41·251 Posts	Got an angry message from 30.9b3 - don't know what it means, but the things work properly here - probably related to "I have to do a lot of code cleaning". Decided to let you know, just in case. Last fiddled with by LaurV on 2022-10-05 at 05:53

2022-10-07, 08:52	#87
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 41×251 Posts	It repeats every 75 minutes. Ever since. I don't know if it is only cosmetic, and the connections works, or not, the assignments are taken, or not, or the results are reported correctly, or not, because there is no result to report, and it won't be for ~80 hours or so. Maybe it is only cosmetic (cleaning the code as George said). Older version (renamed exe file in the same folder, using the same prime.txt/local.txt files) seems to work without any error, so it is not a network issue. Last fiddled with by LaurV on 2022-10-07 at 08:55

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
That's a Lot of Users!!!	jinydu	Lounge	9	2006-11-10 00:14
Beta version 24.6 - Athlon users wanted	Prime95	Software	139	2005-03-30 12:13
For Old Users	Citrix	Prime Sierpinski Project	15	2004-08-22 16:43
Opportunity! Retaining new users post-M40	GP2	Lounge	55	2003-11-21 21:08
AMD USERS	ET_	Lounge	3	2003-10-11 16:52

2022-10-04, 06:47	#82
Denial140 Dec 2021 2⁴·5 Posts	I had something similar happen when I first ran a curve with build 2. Since i forgot to change the memory settings from the default it crashed at the end of stage 1. I assumed this was because it was stopping because of insufficient memory, and hence trying to write a stage 2 savefile (and crashing as expected) - everything worked fine when there was enough memory allocated to run stage 2.

2022-10-05, 15:41	#86
chris2be8 Sep 2009 2⁵·7·11 Posts	That looks like a network glitch truncated the response. So you only received part of the header for the web page but none of the body. If it worked OK when it retried then don't worry about it. I've had many similar problems when scraping HTML in my scripts. And written work rounds to retry errors later in most cases.