When did PrimeNet begin rubberstamping results? - Page 6

retina · 2014-01-02, 05:14

Quote:

Originally Posted by TheMawn

Why ARE you still using V4 if you're fully aware of V5?

Why does it even matter? If V4 works then there is no reason to update. My computer is still running V4. I have no intention to upgrade it to a newer version. The current version works. A newer version might fail. Why should I take the risk of upgrading for no advantage at all? Old software doesn't rust or get worn, it still performs the same as the first day it was run.

Hmm ... perhaps I should be saying to TheMawn "FU for trying to force me to change my setup"? Nah, I won't say that, that would be rude. However I would say it to MS for trying to get me to stop using XP. I don't hate change, but I do hate change for no good reason.

LaurV · 2014-01-02, 05:29

Quote:

Originally Posted by retina

However I would say it to MS for trying to get me to stop using XP.

Same here! (being a M$ guy for years, but this time they made me angry with stopping XP support).

Brian-E · 2014-01-02, 09:26

Yes, there are those who always enthusiastically update their software to the newest version as soon as it's available, those who get round to that occasionally, and finally those who have the attitude "if it ain't broke, don't fix it". I, too, would put myself in the latter category, like I guess Retina and LaurV also would.

The discussion, quite rightly, has been focussing on those who are seen to hoard lots of premium assignments without working on them for a long time. It is nonsense to try to define this group by such a flimsily-related, and undoubtedly poorly correlated, set such as people who don't upgrade from v4 to v5.

I, like others, have complete confidence that tha in particular is being careful with the work he chooses to take from its assignees. I do, however, have difficulty with the idea that *anyone* can just ignore the project administrators' decisions about which assignments to expire and just choose their own work on the basis of their own personal judgment about which assignees are and are not actively working on the project. I think the "slippery slope" concept is relevant here.

patrik · 2014-01-05, 23:54

I just want to add that I had one machine which I couldn't upgrade to v5. It was a Linux machine from 2003, and I needed to upgrade a library, which I couldn't do without upgrading other libraries which then would break other things.

chalsall · 2014-01-06, 00:16

Quote:

Originally Posted by patrik

I just want to add that I had one machine which I couldn't upgrade to v5. It was a Linux machine from 2003, and I needed to upgrade a library, which I couldn't do without upgrading other libraries which then would break other things.

Fair enough.

Just for clarity, are you the patrick who is doing a great deal of work in the 33M range?

Nothing wrong with that, of course. Just wondering.

kladner · 2014-01-06, 00:34

Quote:

Originally Posted by chalsall

Fair enough.

Just for clarity, are you the patrick who is doing a great deal of work in the 33M range?

Nothing wrong with that, of course. Just wondering.

Note the spelling. 'Patrik' is the same on both the forum and PrimeNet. I'm pretty sure that "you have reached the party to whom you are speaking."

chalsall · 2014-01-06, 00:47

Quote:

Originally Posted by kladner

Note the spelling. 'Patrik' is the same on both the forum and PrimeNet. I'm pretty sure that "you have reached the party to whom you are speaking."

Why do so few get my humor?

My girlfriend would say "Because you're not funny....

patrik · 2014-01-06, 01:02

Quote:

Originally Posted by chalsall

Fair enough.

Just for clarity, are you the patrick who is doing a great deal of work in the 33M range?

Nothing wrong with that, of course. Just wondering.

That's me. I've been doing 10M digit double-checks for 2 years now.

With an estimated error rate of 10% just above exponent 33219280, and the DC's taking 1/4 the time of first time LL's, the chance of finding a prime is about 40% of what I would have if I were doing only first time tests.

Edit: That is, it is here that a missed Mersenne prime might be lurking!

Patrik

chalsall · 2014-01-06, 01:07

Quote:

Originally Posted by patrik

That's me. I've been doing 10M digit double-checks for 2 years now.

Thank you for that.

We've recently been dealing with those who might be interpreted as a little less than honest.

So, again, thank you.

patrik · 2014-01-06, 01:22

I just realized that my estimate of 40% is wrong (since every test with error code has been followed by a new one). It is really the fraction of tests with zero error code that is interesting. (It's a bit too late (0220 am) to calculate probabilities right now.)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Primenet Recent Results Request	garo	PrimeNet	7	2014-02-03 02:53
making PrimeNet accept duplicate TF results?	ixfd64	PrimeNet	5	2012-12-08 07:04
In which country do you suggest me to begin my act	Unregistered	Information & Answers	0	2010-11-30 23:34
V5 not reporting results to PrimeNet	edron1011	Software	8	2009-01-26 16:59
Problem with Reporting Results to Primenet	RMAC9.5	Software	3	2003-11-02 23:06

2014-01-02, 09:26	#58
Brian-E "Brian" Jul 2007 The Netherlands 3269₁₀ Posts	Yes, there are those who always enthusiastically update their software to the newest version as soon as it's available, those who get round to that occasionally, and finally those who have the attitude "if it ain't broke, don't fix it". I, too, would put myself in the latter category, like I guess Retina and LaurV also would. The discussion, quite rightly, has been focussing on those who are seen to hoard lots of premium assignments without working on them for a long time. It is nonsense to try to define this group by such a flimsily-related, and undoubtedly poorly correlated, set such as people who don't upgrade from v4 to v5. I, like others, have complete confidence that tha in particular is being careful with the work he chooses to take from its assignees. I do, however, have difficulty with the idea that anyone can just ignore the project administrators' decisions about which assignments to expire and just choose their own work on the basis of their own personal judgment about which assignees are and are not actively working on the project. I think the "slippery slope" concept is relevant here.

2014-01-05, 23:54	#59
patrik "Patrik Johansson" Aug 2002 Uppsala, Sweden 5²·17 Posts	I just want to add that I had one machine which I couldn't upgrade to v5. It was a Linux machine from 2003, and I needed to upgrade a library, which I couldn't do without upgrading other libraries which then would break other things.

2014-01-06, 01:22	#65
patrik "Patrik Johansson" Aug 2002 Uppsala, Sweden 5²×17 Posts	I just realized that my estimate of 40% is wrong (since every test with error code has been followed by a new one). It is really the fraction of tests with zero error code that is interesting. (It's a bit too late (0220 am) to calculate probabilities right now.)