Primenet eyesores
 

The length of the tail (OK cut off the straggily bit)
of the waves (LL and DC) is perfectly natural and to be encouraged:
The more tests in progress, the more will get finished.
***king formula omitted in case you are intelligent.

Why the **** are almost no LLs being dished out TFed to 72?

David

[QUOTE=davieddy;279784]The length of the tail (OK cut off the straggily bit)
of the waves (LL and DC) is perfectly natural and to be encouraged:
The more tests in progress, the more will get finished.
***king formula omitted in case you are intelligent.

Why the **** are almost no LLs being dished out TFed to 72?

David[/QUOTE]

Give it some time. GPUto72 has only been fully operational for a couple of weeks. Remember that the exponents they prepare for LL represent only a tiny percentage of the entire exponent pool. Indeed, remember that GPUers are only a tiny percentage of the GIMPS population.

What is/are *your* GPU(s) crunching? :smile:

OK. Gloves off!
 

[QUOTE=NBtarheel_33;279790]Give it some time. GPUto72 has only been fully operational for a couple of weeks. Remember that the exponents they prepare for LL represent only a tiny percentage of the entire exponent pool. Indeed, remember that GPUers are only a tiny percentage of the GIMPS population.

What is/are *your* GPU(s) crunching? :smile:[/QUOTE]
They represent 0% of the infinite pool Uncwilly etc would
like to trawl.

TF takes about 1% of the time of an LL.

My Celeron does a LL in ~40 days (24/7). I don't mind
getting my hands dirty occasionally.

Stop f***ing about with this breadth first, arse over tip approach.

David

[B]Already[/B]

[QUOTE=davieddy;279798]TF takes about 1% of the time of an LL.[/QUOTE]

so each participant can do 100 TF for each LL which means if 100 participants did tf we get 10000 tf results every 40 days according to your Celeron stats. so all numbers from 1 to 43112609 done tf in under 473 years and that's including composite exponents. not counting composites and assuming the average result is the same as yours we get all prime exponents under 840,354,253 but remember that's in 473 years I hope you survive it.

[QUOTE=davieddy;279798]Stop [redacted] about with this breadth first, arse over tip approach.[/QUOTE]

I don't know what you're complaining about. We've already released back to PrimeNet over 1000 candidates TFed to 72 and P-1ed.

Looking only at the 50M-60M Range...

Since last week 1894 exponent have been TF to 72. Yet in that same time frame 1042 exponent got LL tested for the first time. I think it's fair to assume most of those grabbed a new exponent? But as we know, only 10-25 or 50% (I don't remember the number lets say 25%) so we can assume 4000 exponent were handed out last week, a fair amount of those exponent were TF to 72, but they will never get completed. Next time they are released to the pool, we won't have to TF them.

We will catch up to the wavefront, just need to give it time...

[QUOTE=chalsall;279820]I don't know what you're complaining about. We've already released back to PrimeNet over 1000 candidates TFed to 72 and P-1ed.[/QUOTE]With 140, non-DC range, factors found.:xmastree:

Davieddy:
This "tip over arse" effort has yielded 50 factors in the last week between 45M and 50M and another 107 between 50M and 60M. That's a very direct effect on the LL wavefront, and it's exactly the kind of thing we hoped to do...by hitting the "worst offenders first"!

Patience grasshopper...patience....:beer::xmastree::batalov:

[QUOTE=Christenson;279841]Davieddy:
This "tip over arse" effort has yielded 50 factors in the last week between 45M and 50M and another 107 between 50M and 60M. That's a very direct effect on the LL wavefront, and it's exactly the kind of thing we hoped to do...by hitting the "worst offenders first"!

Patience grasshopper...patience....:beer::xmastree::batalov:[/QUOTE]

Is this to say that exponents are being eliminated prior to DC even if they have had a first time LL? Sorry...I've had a couple margaritas and now some sherry...so I may have misinterpreted the implications of the above posts!

[QUOTE=Primeinator;279933]Is this to say that exponents are being eliminated prior to DC even if they have had a first time LL? Sorry...I've had a couple margaritas and now some sherry...so I may have misinterpreted the implications of the above posts![/QUOTE]
My understanding is: [U]there have been[/U] some P-1 and/or TF (via GPU) successes (factors found) for exponents that have been LL'ed once.

[QUOTE=Primeinator;279933]Sorry...I've had a couple margaritas and now some sherry[/QUOTE]
I find that does wonders as regards appreciation
of this forum:smile:

David

To be serious for a bit...
 

[QUOTE=Primeinator;279933]Is this to say that exponents are being eliminated prior to DC even if they have had a first time LL?[/QUOTE]
This is one of the "eyesores".
Obviously TF/P-1 should be done before the first LL, and
the DC shouldn't need it.
Those "available" P-1s and "assigned" TFs in the 40-50M range are such (LLed once).
No urgency whatsoever to do anything with them.
George said to me:
"I goofed but no harm done".
True, but please erase them.
Or someone knock them on the head.

David

[QUOTE=davieddy;279943]This is one of the "eyesores".
Obviously TF/P-1 should be done before the first LL, and
the DC shouldn't need it.
Those "available" P-1s and "assigned" TFs in the 40-50M range are such (LLed once).
No urgency whatsoever to do anything with them.
George said to me:
"I goofed but no harm done".
True, but please erase them.
Or someone knock them on the head.

David[/QUOTE]

Thank you for clearing this issue up!

[QUOTE=Primeinator;279933]Is this to say that exponents are being eliminated prior to DC even if they have had a first time LL? Sorry...I've had a couple margaritas and now some sherry...so I may have misinterpreted the implications of the above posts![/QUOTE]

Ckdo and some helpers, including me, did some additional TF on exponents definitely in the DC range to 69 bits, and certainly turned in a few factors, thus completing several proofs of compositeness with only a single LL and eliminating the LL-DC. Primenet, of course, DCs factors when they are turned in. In the very low range, below 10M, I believe Petrw1 has been running P-1 when it wasn't run well, and finding factors, too. There's nothing to stop you from asking GPU to 72 for more of these 25M-30M exponents, and no reason to believe you won't find more factors if you do. But the optimum, in terms of minimising total effort to find M48 and prove it is indeed M48 and not M49, is to first try to elimnate 2 LL tests with TF'ing below 72 bits at approximately 50M, before first LL tests, until exponents at that level get hard to find with TF below 72 bits. We do try not to do TF on in-progress LL tests.

This is because there is a multiple bitlevel difference in optimal TF between a 25M exponent with one completed LL and a 50M exponent with no completed LL on it. First, the LL tests are 4 times as hard to do at 50M (2x as many iterations, 2x FFT size), and if a single LL test is eliminated on the DC range, the savings are only half. Second, 70 bits TF at 50M is approximately only half as much work as 70 bits at 25M because there are only half as many factor candidates. Finally, the probability of finding a factor by TF between 69 and 70 bits is approximately 1 in 70, regardless of whether we TF a 25M or 50M exponent.

So yes, we have TF'ed exponents that have had single LL tests, and will do so again in the future, as soon as the yield of work saved to GIMPS on exponents with no LL test drops far enough. Some of us also run DCs on our GPUs, since there is a very limited yield (10%-20% fewer exponents to LL test) to increasing the amount of TF work that can be done.

[QUOTE=Christenson;280060]Ckdo and some helpers, including me, did some additional TF on exponents definitely in the DC range to 69 bits, and certainly turned in a few factors, thus completing several proofs of compositeness with only a single LL and eliminating the LL-DC.

....

So yes, we have TF'ed exponents that have had single LL tests, and will do so again in the future, as soon as the yield of work saved to GIMPS on exponents with no LL test drops far enough. Some of us also run DCs on our GPUs, since there is a very limited yield (10%-20% fewer exponents to LL test) to increasing the amount of TF work that can be done.[/QUOTE]

Thank you. However, looking at the stats page it seems like there is a huge need for DCs at the moment. There are still over 300,000 exponents remaining before M47. Are there any rough estimates on how many years THAT is going to take?

I personally recently set myself the goal of keeping at least as many DC's completed as LL. That way, around 80% of my total CPU time goes to LL, while at the same time I do my fair share of DC. I also maintain a similar policy for P-1 work relative to the others.

[QUOTE=Dubslow;280270]I personally recently set myself the goal of keeping at least as many DC's completed as LL. That way, around 80% of my total CPU time goes to LL, while at the same time I do my fair share of DC. I also maintain a similar policy for P-1 work relative to the others.[/QUOTE]

For what it's worth; lifetime GIMPS I show 500 LL and 540 DC.
Prior to v5 server DC was not counted seperately so my LL is a bit overstated and DC a bit understated--- by 20 or so.
Over this past year I focused primarily on P-1 with a few diversions. I should be close to 2,000 for the year by year end.
My "tentative" goal for 2012 is 1,000 DC's.

[QUOTE=petrw1;280272]For what it's worth; lifetime GIMPS I show 500 LL and 540 DC.
Prior to v5 server DC was not counted seperately so my LL is a bit overstated and DC a bit understated--- by 20 or so.
Over this past year I focused primarily on P-1 with a few diversions. I should be close to 2,000 for the year by year end.
My "tentative" goal for 2012 is 1,000 DC's.[/QUOTE]

One thousand? How many computers do you have??

I think that's probably 2-3 high-end GPUs...especially if PetrW means "total 1000DCs by the end of 2012" and not "1000DCs completed during 2012".

And by the way, we'll be very upset (NOT!) if you help with the DC effort!!!:smile:
...whether by DC on P95, DC on GPU with CUDALucas, P-1, or TF... :smile:

[QUOTE=Christenson;280277]I think that's probably 2-3 high-end GPUs...especially if PetrW means "total 1000DCs by the end of 2012" and not "1000DCs completed during 2012".

And by the way, we'll be very upset (NOT!) if you help with the DC effort!!!:smile:
...whether by DC on P95, DC on GPU with CUDALucas, P-1, or TF... :smile:[/QUOTE]

If I can get a hold of an older machine I will for sure! I devote my quad core to first time LLs of record size due to its superior number crunching ability. I have an older laptop and an ancient desktop that if I can get them up and running again will dedicate them to the DC effort. I may also get my hands on a somewhat newer Mac desktop of some kind. Depending on its specs I will either dedicate it to DC or LL.

I'll need a laptop sometime soon, and when I get it, it's going on P-1 duty. (Gonna have 1600MHz native memory, hopefully!)

[QUOTE=Primeinator;280279]If I can get a hold of an older machine I will for sure! I devote my quad core to first time LLs of record size due to its superior number crunching ability. [/QUOTE]

Here's a wacky thought: The next mersenne prime to be found will be in the DCs...it had an erroneous non-zero residue turned in by someone.....it's not too wildly unlikely...especially with three different residues now on the last exponent to be DC'd below 24M....time to think about a GPU and seeing what it can do with CudaLucas...

[QUOTE=Christenson;280327]Here's a wacky thought: The next mersenne prime to be found will be in the DCs...it had an erroneous non-zero residue turned in by someone.....it's not too wildly unlikely...especially with three different residues now on the last exponent to be DC'd below 24M....time to think about a GPU and seeing what it can do with CudaLucas...[/QUOTE]

I hope not... it'll be a while before we get to it.

[QUOTE=Christenson;280327]Here's a wacky thought: The next mersenne prime to be found will be in the DCs...it had an erroneous non-zero residue turned in by someone.....it's not too wildly unlikely...especially with three different residues now on the last exponent to be DC'd below 24M....time to think about a GPU and seeing what it can do with CudaLucas...[/QUOTE]

I suppose this would depend on the percentage of reliable first-time LL tests currently turned into the serve. There are very few (less than 200) exponents untested before M47* currently. Does anyone have any information on those that have received a first-time LL test and how many of those are suspect?

[QUOTE=Dubslow;280330]I hope not... it'll be a while before we get to it.[/QUOTE]

How does this make any difference? It is going to take a LONG time to DC all of those exponents anyway. The wait is to discover that there are missing primes or aren't any missing primes is the same. We don't have a way of knowing.

AFAIK, any tests turned in with an error code are immediately reassigned to DC workers, but of course not all bad tests report error codes. (Also, the last sentence does not make sense: Less then 200 expos untested, then ask which of those have had a test completed?)

[QUOTE=Primeinator;280335]How does this make any difference? It is going to take a LONG time to DC all of those exponents anyway. The wait is to discover that there are missing primes or aren't any missing primes is the same. We don't have a way of knowing.[/QUOTE]

Well that's what I meant, it'll be a LONG time before we figure out if we missed one. I'm confident enough in the heuristics that I don't think there are any others below M47; however, if there's another one less than 50M (which I doubt, but it's more likely than another one below 43M) it's more likely than not that we've missed it on the first go, and it could easily be a decade before we check those twice.

[QUOTE=Dubslow;280336]AFAIK, any tests turned in with an error code are immediately reassigned to DC workers, but of course not all bad tests report error codes. (Also, the last sentence does not make sense: Less then 200 expos untested, then ask which of those have had a test completed?)[/QUOTE]

I guess what I was trying to say with this is that there could (very unlikely) be an undiscovered prime below M47 that has not been LL tested once yet. Most exponents below this value have been tested at least once...it is even less likely that one of these is prime and just had a bad LL. What defines "suspect LL?" Any with an error code? I have seen these on PrimeNet.

[QUOTE=Dubslow;280338]Well that's what I meant, it'll be a LONG time before we figure out if we missed one. I'm confident enough in the heuristics that I don't think there are any others below M47; however, if there's another one less than 50M (which I doubt, but it's more likely than another one below 43M) it's more likely than not that we've missed it on the first go, and it could easily be a decade before we check those twice.[/QUOTE]

It will be a very long time indeed. Why do you feel like there aren't likely any exponents below 50M that will yield a new Mersenne? How many in this range have yet to receive a first time LL test?

[QUOTE=Primeinator;280347]What defines "suspect LL?" Any with an error code? I have seen these on PrimeNet.[/quote]
Yes, you are correct. Any error code marks it as suspect, along with immediate retesting as I said earlier.

[QUOTE=Primeinator;280347]
It will be a very long time indeed. Why do you feel like there aren't likely any exponents below 50M that will yield a new Mersenne? How many in this range have yet to receive a first time LL test?[/QUOTE]
Over the course of this project, there have been a surprising number of primes already, and given the random but average nature of their location (technical term: poisson distribution) I suspect that the next one is very far off. As for data:

[url]http://mersenne.info/exponent_status_tabular_data/2/40000000/[/url]
[url]http://mersenne.info/exponent_status_line_graph/2/40000000/[/url]

[QUOTE=Dubslow;280348]
[url]http://mersenne.info/exponent_status_tabular_data/2/40000000/[/url]
[url]http://mersenne.info/exponent_status_line_graph/2/40000000/[/url][/QUOTE]

Beautiful. I did not know you could generate reports like that.

There are still 5,000+ exponents awaiting a first time LL test in that range... statistically speaking, not great odds of finding a new Mersenne, but don't count your chickens too soon. Perhaps we have miscalculated how frequently Mersenne primes appear and they are a little more common. Increasing p by 7 million is a rather substantial jump for not finding any new primes.

The gaps between M38,M39,M40 are close to 7 million, at much lower numbers. That the following 7* are closer together seems to me that we must swing to the other side of the average, and I don't think 7M or even 10+ is unreasonable. I have half a mind to think there are none between 43M and 60M, though that is rather extreme. Of course, this is all just speculation as you pointed out.

Props to chalsall for the tool, especially the percentage graph, that's particularly awesome :)

[QUOTE=Dubslow;280352]The gaps between M38,M39,M40 are close to 7 million, at much lower numbers. That the following 7* are closer together seems to me that we must swing to the other side of the average, and I don't think 7M or even 10+ is unreasonable. I have half a mind to think there are none between 43M and 60M, though that is rather extreme. Of course, this is all just speculation as you pointed out.[/QUOTE]

This sounds like an example of the gambler's fallacy to me.

Maybe. Gambling is truly random, whereas primes are random but average.