Ok. Now it started to make more sense. Thanks a lot axn and Dubslow.

So, going to the table of the overall progress, it means the project found a LL-front factor in about 352 GHzd, that is 3.5 to 4 days, which indeed is saving a lot of time, assuming my card can do one LL in 5-6 days (about 130 hours) at this LL-front-size, and we saved 2 LLs and some P-1 too. So, we do the 10-12 day job in 4 days. This really makes sense!!

Also, no discussion about P-1 factors, this is quite profitable, I never had any bit of doubt, and the proof is in the table (average, that is one factor at each half day, which is more than wonderful!).

But please allow me to still doubt about factoring at DC-front. That 220 GHzd per factor is [B]much over double of the time[/B] DC will take. It is two days and about a half for a factor, when a DC-LL would take 24 to 30 hours. Of course, this come from the fact that a lot of P-1 was already done there, which filtered out part of the exponents with factors in the targeted range.

[QUOTE=axn;279454]
That would be a colossal waste of resources. [/QUOTE]

Ok. Trusting in your judgement. I will unreserve them all when I arrive home after the working day ends (in about 5 hours), except for the expos which are currently under process, and re-reserve other, with more reasonable limits. Don't need to involve chalsall in this, it can be done from the assignment page. It will require some work from my side for wortodo files (cut, paste, blabla) as I spreaded the assignments for all mfaktc copies.

[QUOTE=LaurV;279458]Ok. Trusting in your judgement. I will unreserve them all when I arrive home after the working day ends (in about 5 hours), except for the expos which are currently under process, and re-reserve other, with more reasonable limits. Don't need to involve chalsall in this, it can be done from the assignment page. It will require some work from my side for wortodo files (cut, paste, blabla) as I spreaded the assignments for all mfaktc copies.[/QUOTE]

The reason I wanted you to talk to chalsall was that may be you don't need to do any release/assignment at all.

I am not sure, but if you just edit your worktodo and change the "72" to "68" (or "69" if some of them are already at "68"), and then report your work, the tool would still accept it. After that you can just release them.

But chalsall would know the best way.

Anyways... up to you. If you're comfortable just releasing/reassigning, go for it. :smile:

[QUOTE=Dubslow;279453]72 for first time LL's, 68 for DC. Keep in mind that it also takes twice as long to do a DC exponent at a certain bit depth as the 45M first timers.[/QUOTE]

Unverified exponents not at 68 already are rare, to put it mildly, below 32M. Since extending TF depth on expos above 32M is not exactly of any urgency, taking everything below to 69 (and no further) is the route to follow, for now.

Apples and oranges...
 

[QUOTE=chalsall;279446]Indeed.

GPUs have completely changed the metrics. The more advanced ones produce something like 100 times the output of a CPU for trial factoring. It's a bit like a Battleship circling an Ironclad...

But this is the metric which PrimeNet uses.

James and I are simply calculating what PrimeNet calculates.[/QUOTE]

The only thing we all agree on is that GPUs should
do the TF a bit further than a CPU would have done.
[url=http://www.youtube.com/watch?v=jT3_UCm1A5I]Say no more.[/url]

David

Another issue: something wrong with the visualization tool. For a [URL="http://www.mersenne.info/trial_factored_tabular_delta_7/2/400000000/"]certain zoom in, and further[/URL], it shows all numbers (prime or not) as "exponents TF-ed to 58".

[QUOTE=ckdo;279460]Unverified exponents not at 68 already are rare, to put it mildly, below 32M. Since extending TF depth on expos above 32M is not exactly of any urgency, taking everything below to 69 (and no further) is the route to follow, for now.[/QUOTE]
[url=http://www.youtube.com/watch?v=J3gWi9bBkHQ]Beginning to see the light[/url]
:smile:

David

[QUOTE=davieddy;279465][QUOTE=ckdo;279460]Unverified exponents not at 68 already are rare, to put it mildly, below 32M. Since extending TF depth on expos above 32M is not exactly of any urgency, taking everything below to 69 (and no further) is the route to follow, [COLOR=Red][B]for now[/B][/COLOR].[/QUOTE][URL="http://www.youtube.com/watch?v=J3gWi9bBkHQ"]Beginning to see the light[/URL]
:smile:

David[/QUOTE]

Emphasis mine. Video not available in Germany, as usual.

[QUOTE=ckdo;279468]Emphasis mine. Video not available in Germany, as usual.[/QUOTE]I really fancy Angela Merkl.
Ask her to let you hear [url=http://www.youtube.com/watch?v=UyhkBg8wOBo]The route to go[/url]
David

[QUOTE=ckdo;279460]Unverified exponents not at 68 already are rare, to put it mildly, below 32M. Since extending TF depth on expos above 32M is not exactly of any urgency, taking everything below to 69 (and no further) is the route to follow, for now.[/QUOTE]

Then probably, declaring "mission accomplished" and moving on to first time LL-TF would be the way to go.

I've told you once
 

[QUOTE=axn;279474]Then probably, declaring "mission accomplished" and moving on to first time LL-TF would be the way to go.[/QUOTE]
:smile:

Really minor buglet, but clicking on "Created Overal System Status report. " in the what's new page takes me to my assignments instead of the overall system status report as promised. I feel bad even pointing out something so insignificant ... but it's probably easier to fix than any other bug report so far :)

[QUOTE=LaurV;279464]Another issue: something wrong with the visualization tool. For a [URL="http://www.mersenne.info/trial_factored_tabular_delta_7/2/400000000/"]certain zoom in, and further[/URL], it shows all numbers (prime or not) as "exponents TF-ed to 58".[/QUOTE]

But only when you go there from the GPU to 72 Weekly Progress link...not the regular GIMPS VIS link

[QUOTE=petrw1;279508]But only when you go there from the GPU to 72 Weekly Progress link...not the regular GIMPS VIS link[/QUOTE]
No, when you drill down it still does that. (At least it does here.)

[QUOTE=Uncwilly;279539]No, when you drill down it still does that. (At least it does here.[/QUOTE]

Yeah, that's a known bug... If you're looking at the "Changed" reports, and there haven't been any changes in the ranges (and time period) you've asked for, it does that.

Relatively simple to fix, but right now I'm working on something else. [URL="http://gpu.mersenne.info/reports/available/lldc/"]This might give you all a hint....[/URL] :smile:

[QUOTE=chalsall;279540](...) right now I'm working on something else. [/QUOTE]

Wow... another brick in the wall!

GPU to 72 now ready to assign real LL work.
 

OK everyone, the GPU to 72 Tool is now ready to automatically assign low LL candidates to PrimeNet workers. Please see [URL="http://gpu.mersenne.info/account/getassignments/ll/"]http://gpu.mersenne.info/account/getassignments/ll/[/URL].

Each candidate has been trial factored to at least 72 "bits" by GPUs, and had P-1 work completed. There are currently 67 available in the 45M range, and 33 in the 46M.

The system is currently retaining the lowest 100 for this service, and will be topped up as GPU and P-1 workers continue their completions. This number will be adjusted over time as we get a feel for how much fire power this work type will have brought to bear.

Note that this is intended for those LL workers who are "trusted" to complete the work in a reasonable time, and thus access must be authorized. Most GPU to 72 Workers already have access, but if anyone is not yet in the system but would like access, please sign up and then PM me here.

The DC version of this will be available later today.

[QUOTE=chalsall;279584]Note that this is intended for those LL workers who are "trusted" to complete the work in a reasonable time[/QUOTE]

Are you using George's method ... ie if your CPU's "Reliability and Confidence" are at a certain level..Per the Assignment Rules page they need to be 0.9 and 2.

[QUOTE=petrw1;279587]Are you using George's method ... ie if your CPU's "Reliability and Confidence" are at a certain level..Per the Assignment Rules page they need to be 0.9 and 2.[/QUOTE]

No, I don't have access to that data. (I don't think -- if there's a publicly available page on PrimeNet for all users, please let me know.)

Instead I'm going by the amount of work someone has done over the last year.

BTW, the [URL="http://gpu.mersenne.info/account/getassignments/dc/"]Low Double Check work type[/URL] is now available as well.

[QUOTE=chalsall;279588]No, I don't have access to that data. (I don't think -- if there's a publicly available page on PrimeNet for all users, please let me know.)

Instead I'm going by the amount of work someone has done over the last year.

BTW, the [URL="http://gpu.mersenne.info/account/getassignments/dc/"]Low Double Check work type[/URL] is now available as well.[/QUOTE]

Not that I know of.

My current plan is to do GPUto72 P-1 till about year end...I should complete close to 500. And then I think I will move to this DC.

[QUOTE=chalsall;279588]the [URL="http://gpu.mersenne.info/account/getassignments/dc/"]Low Double Check work type[/URL] is now available as well.[/QUOTE]I've grabbed one to see how it works. ETC: 01-Dec-2011

Our last discussions were quite fruitful. Following the argue here on the forum, I took 60 exponents from DC front and crunched them today for a couple of hours on both GPU's from 67/68 to 69 bits. It was one of that lucky days:

[CODE]
M25300493 has a factor: 482498669664986964257
found 1 factor(s) for M25300493 from 2^68 to 2^69 [mfaktc 0.17-Win barrett79_mul32]

M25381229 has a factor: 444817498910828872201
found 1 factor(s) for M25381229 from 2^68 to 2^69 [mfaktc 0.17-Win barrett79_mul32]

M25386671 has a factor: 257812533279578085649
found 1 factor(s) for M25386671 from 2^67 to 2^68 [mfaktc 0.17-Win barrett79_mul32]

[/CODE]I also saw that the "score" decreased within the last 36 hours from "2.4 days per exponent cleared (=factor found)" to "1.9 days per expo cleared" (current score on the overall system progress is "197.3 GHz-days per factor). That is very good, but I am still not convinced, the score should be half of that (i.e. the time should be half per factor, or the amount of factors found in this same time should be double) to really worth TF-ing at DC front, comparing with the time CudaLucas takes to clear an exponent in 26M range (24 hours).

Now as DC work is available, I will switch to that type of work after my current bunch of exponents (137 in the queue, plus 10 at LL front to 72, but that type of work I will continue, as each factor saves 2 LL tests, as discussed earlier) will be all finished. I am for sure "trusted producer", being gimpster from the beginning, but here I still could have a problem if the exponents were first time LL-ed by a third party application: the server will not accept my results. So I will have to manually walk through the Primenet's report to check if the first LL was done by P5 or by other program, for each exponent I would get from GPU272, before starting any work on it, and in case not, to un-reserve it back. And as I know, this is more like a guessing now, like check the uppercase/lowercase of the hex digits, scratch your head, turn around, do whatever else.... then take the risk...

Can anything be done in this direction? Like "assignments for DC work which had first LL done by P95 only"? (here maybe George could help to publish the "shift amount" for the LL's for each exponents? all CudaLucas tests have the shift zero, all P95 work have a random shift which is not zero). And have "assignments for DC work which had first LL done by third parties", which therefore could by double checked with P95 only, for who wants to run DC's on CPU.

Is it possible? Or I am dreaming...

Why don´t you keep the GPUs for TF (that´s where they truly excel) and do the DC work on CPU cores (with Prime95)?
Or, if you really want to use GPUs for LL testing, do first-time tests on them.
Both alternatives would save you (or someone else...) the trouble of analyzing the record of previous LL tests performed.

[QUOTE=LaurV;279594]Can anything be done in this direction? Like "assignments for DC work which had first LL done by P95 only"? (here maybe George could help to publish the "shift amount" for the LL's for each exponents? all CudaLucas tests have the shift zero, all P95 work have a random shift which is not zero). And have "assignments for DC work which had first LL done by third parties", which therefore could by double checked with P95 only, for who wants to run DC's on CPU.

Is it possible? Or I am dreaming...[/QUOTE]

It is possible, but I'm not going to do it...

To do what you ask would involve spidering a separate page for each exponent being returned to PrimeNet. Including those I grab and then immediately return because they're not "interesting" enough to have GPU/P-1 work done on them. This would put significant additional load on the PrimeNet server for very (very) little net benefit.

I would advise you take lycorn advice, and use your GPUs for TFing (preferably in the LL range, because I agree with you DC TF work to 69 is only marginally efficient), and any left over CPU cycles you apply to real DCing.

[QUOTE=LaurV;279594]Our last discussions were quite fruitful. Following the argue here on the forum, I took 60 exponents from DC front and crunched them today for a couple of hours on both GPU's from 67/68 to 69 bits. It was one of that lucky days:

[CODE]
M25300493 has a factor: 482498669664986964257
found 1 factor(s) for M25300493 from 2^68 to 2^69 [mfaktc 0.17-Win barrett79_mul32]

M25381229 has a factor: 444817498910828872201
found 1 factor(s) for M25381229 from 2^68 to 2^69 [mfaktc 0.17-Win barrett79_mul32]

M25386671 has a factor: 257812533279578085649
found 1 factor(s) for M25386671 from 2^67 to 2^68 [mfaktc 0.17-Win barrett79_mul32]

[/CODE]I also saw that the "score" decreased within the last 36 hours from "2.4 days per exponent cleared (=factor found)" to "1.9 days per expo cleared" (current score on the overall system progress is "197.3 GHz-days per factor). That is very good, but I am still not convinced, the score should be half of that (i.e. the time should be half per factor, or the amount of factors found in this same time should be double) to really worth TF-ing at DC front, comparing with the time CudaLucas takes to clear an exponent in 26M range (24 hours).[/QUOTE]

You need to remember not all of us have a GPU like yours. On my 560 I can complete a DC in 2.4 days where my 450 can do one in 4.3 and I'm guessing the 560 Ti to be around 2.1. Compared to that, my i5 2400 can do a DC in a little under 4 days while my i7 Q740 takes 10 days.

In terms of GHz Days/Day, doing TF, my 560 does ~140, the 460 ~95 and the 560 Ti doing ~150. ~385 GHzDays/Day at your 197.3/factor means I'd average ~2 factors per day compared to completing ~5 DC/4 days. 5 vs 8 tells me that TF outweighs LL at this point in terms of GPU usage on my systems. I would think your GPU to be similiar in terms of factors vs DC.

[QUOTE=lycorn;279607]Why don´t you keep the GPUs for TF (that´s where they truly excel) and do the DC work on CPU cores (with Prime95)?
Or, if you really want to use GPUs for LL testing, do first-time tests on them.[/QUOTE]

The GPU to 72 project cater in big part to people having that kind of hardware (GPU).

If I like to TF number or do LL, DC or god forbid do OBD or further TF on already DCed exponent. I think, those who pay for the hardware and electricity should have the right to do as they please?

Coming here and asking for a legitimate feature, given the problem encountered by the user. But being rebuked with a, "why don't you do something else" is not very helpful.

Software change and get better... Why wouldn't CUDALucas get so much better that doing any LL on GPU wouldn't be wasteful.

The issue of identifying where the first LL result came from is still relevant.

[QUOTE=lycorn;279607]Both alternatives would save you (or someone else...) the trouble of analyzing the record of previous LL tests performed.[/QUOTE]

Please let him or them determine what is too much trouble

Already
 

[QUOTE=diamonddave;279613]
Software change and get better... Why wouldn't CUDALucas get so much better that doing any LL on GPU wouldn't be wasteful.
[/QUOTE]:smile:

[QUOTE=diamonddave;279613]Software change and get better... Why wouldn't CUDALucas get so much better that doing any LL on GPU wouldn't be wasteful.[/QUOTE]

When (if?) that happens, the metrics once again change and a re-evaluation will happen.

At this point in time, it is far more efficient to have GPUs do TF, and CPUs do LL / DC (or P-1).

Also, please note, to the best of my knowledge, no low candidates were ever LLed by any program other than Prime95 or mprime.

But, as always, I'm happy to be corrected.

[QUOTE=davieddy;279623]:smile:[/QUOTE]

Smile back...

[URL="http://www.youtube.com/watch?v=mXsmyLtpxlA"]Live's What you Make It[/URL]

:cool:

[QUOTE=diamonddave;279613]

If I like to TF number or do LL, DC or god forbid do OBD or further TF on already DCed exponent. I think, those who pay for the hardware and electricity should have the right to do as they please? [/QUOTE]
Sure "they" (we all) have, and have always had.

[QUOTE=diamonddave;279613]
Coming here and asking for a legitimate feature, given the problem encountered by the user. But being rebuked with a, "why don't you do something else" is not very helpful. [/QUOTE]

What you mean "rebuked"? I was only suggesting a way of overcoming the issue raised by LaurV himself, not by me...

[QUOTE=diamonddave;279613]
Software change and get better... Why wouldn't CUDALucas get so much better that doing any LL on GPU wouldn't be wasteful. [/QUOTE]

CUDALucas [U]is already[/U] good at LL testing. That´s not the point. That´s why I suggested doing LLs on a GPU. But being first timers avoids the problem of determining the origin of the first result, once there isn´t any.

[QUOTE=diamonddave;279613]
Please let him or them determine what is too much trouble[/QUOTE]

As you can see from chalsall´s response, the request would require an exaggerated amount of work for what it is worth. And in his post, LaurV was already admitting that managing that manually would be a large task for himself. So...
Keep cool man, we both are around here for enough time so that you should already know me well enough to hint that I was dismissing or "pushing" someone to do this or that. That´s not my style, and I think it was apparent from my post that the idea was just being helpful to a "fellow cruncher".

[QUOTE=diamonddave;279613]If I like to TF number or do LL, DC or god forbid do OBD or further TF on already DCed exponent. I think, those who pay for the hardware and electricity should have the right to do as they please?[/QUOTE]
Or even work on the 100M digit range.:whistle:

[QUOTE=lycorn;279635]

(...) so that you should already know me well enough to [B]not [/B]hint that I was dismissing or "pushing" someone to do this or that. [/QUOTE]
Typo corrected... :wink:

[QUOTE=Uncwilly;279637]Or even work on the 100M digit range.:whistle:[/QUOTE]
It can't. CudaLucas is limited to about half of that size for the exponent :P

[QUOTE=LaurV;279654]It can't. CudaLucas is limited to about half of that size for the exponent :P[/QUOTE]But [i]Uncwilly[/i] can, he has no such limitations. :smile:

Just produced the 4th factor overnight. If it continues like that, I will break even, and then I will not be in the position to argue anymore :smile:
I just saw on the overall report a new drop, from 197 to 192 Gd/factor on DC front. It just pops under my hair that in fact, my "error" in the beginning, when I was (stupidly) factoring DC front to 72 bits (see above in this thread) is heavily weighting on that number. There were 11 exponents, with an additional ~35Gd for each (additional unwanted work done from 69 to 72), that is about 385Gd, which means about 18Gd for each of the 21 factors found to date. So, the "real profitability" of DC-front factoring should be about 174Gd/factor, or about 1.7 days. We are coming close to breakeven. :smile:

[QUOTE=chalsall;279588]BTW, the [URL="http://gpu.mersenne.info/account/getassignments/dc/"]Low Double Check work type[/URL] is now available as well.[/QUOTE]

I noticed in today's LL availability report there is one 49M candidate without P-1 and only(?) at 69 bits. Wasn't the intent to TF a lot deeper (hence the projects name) before releasing to LL?

Also the DC availability shows 100 in the 25M range but the Active Assignments report in GIMPS show you only have 35. Is my problem just timing; i.e. did someone grab 65 in the last hour?

[QUOTE=petrw1;279705]I noticed in today's LL availability report there is one 49M candidate without P-1 and only(?) at 69 bits. Wasn't the intent to TF a lot deeper (hence the projects name) before releasing to LL?[/QUOTE]

Yes. This is the opposite result of the "race condition" which I described to davieddy in post # 181 above. My "anonymous" spider grabbed an "interesting" candidate which my "GPU Factoring" spider had not only a couple of seconds earlier.

These will scroll off (and hopefully be picked up again by "Spidy") as additional lower candidates are completed by GPU / P-1 workers.

[QUOTE=petrw1;279705]Also the DC availability shows 100 in the 25M range but the Active Assignments report in GIMPS show you only have 35. Is my problem just timing; i.e. did someone grab 65 in the last hour?[/QUOTE]

I'm note entirely sure what report you are looking at. [url]http://www.mersenne.org/primenet/[/url] shows ~3200 DCs assigned at the moment.

[QUOTE=chalsall;279712]I'm note entirely sure what report you are looking at. [url]http://www.mersenne.org/primenet/[/url] shows ~3200 DCs assigned at the moment.[/QUOTE]

This one: [url]http://mersenne.org/assignments/?exp_lo=25000000&exp_hi=26000000&execm=1&exfirst=1&exp1=1&extf=1&B1=Get+Assignments[/url]
It lists 35 belonging to GPU dd_work

100+
 

Just surpassed 100 P-1 completions in GPU to 72. 3 Factors found.

[QUOTE=petrw1;279718]This one: [url]http://mersenne.org/assignments/?exp_lo=25000000&exp_hi=26000000&execm=1&exfirst=1&exp1=1&extf=1&B1=Get+Assignments[/url]
It lists 35 belonging to GPU dd_work[/QUOTE]

These reports are limited to 1000 rows.... :smile:

Also, keep in mind that reservations for real LL and DC work are reserved by an Anonymous account to facilitate the automatic transfer of ownership to the worker by PrimeNet.

[QUOTE=chalsall;279721]These reports are limited to 1000 rows.... :smile:[/QUOTE]

:redface:

Will (or can) there be a LL/DC summary page?

[QUOTE=petrw1;279869]Will (or can) there be a LL/DC summary page?[/QUOTE]

There can (and will).

I'm just waiting for a completion.... :smile:

[QUOTE=chalsall;279875]There can (and will).

I'm just waiting for a completion.... :smile:[/QUOTE]

:doh!:

I'm trying to hurry...
 

[QUOTE=chalsall;279875]There can (and will).

I'm just waiting for a completion.... :smile:[/QUOTE]

I'm hurrying as fast as I can.....

[CODE][Nov 25 18:44] Iteration: 12840000 / 45136699 [28.44%]. Per iteration time: 0.034 sec.
[Nov 25 18:49] Iteration: 12850000 / 45136699 [28.46%]. Per iteration time: 0.034 sec.
[Nov 25 18:55] Iteration: 12860000 / 45136699 [28.49%]. Per iteration time: 0.034 sec.
[/CODE]

[QUOTE=Chuck;279883]I'm hurrying as fast as I can.....
[Nov 25 18:55] Iteration: 12860000 / 45136699 [28.49%]. Per iteration time: 0.034 sec.[/QUOTE]I'm hurrying faster :smile:[code][Nov 25 19:13] Iteration: 9400000 / 25132963 [37.401081%]. Per iteration time: 0.021 sec.[/code]

[QUOTE=Chuck;279883]I'm hurrying as fast as I can.....

[CODE][Nov 25 18:44] Iteration: 12840000 / 45136699 [28.44%]. Per iteration time: 0.034 sec.
[Nov 25 18:49] Iteration: 12850000 / 45136699 [28.46%]. Per iteration time: 0.034 sec.
[Nov 25 18:55] Iteration: 12860000 / 45136699 [28.49%]. Per iteration time: 0.034 sec.
[/CODE][/QUOTE]

Thanks. :smile:

[QUOTE=Chuck;279883]I'm hurrying as fast as I can.....

[CODE][Nov 25 18:44] Iteration: 12840000 / 45136699 [28.44%]. Per iteration time: 0.034 sec.
[Nov 25 18:49] Iteration: 12850000 / 45136699 [28.46%]. Per iteration time: 0.034 sec.
[Nov 25 18:55] Iteration: 12860000 / 45136699 [28.49%]. Per iteration time: 0.034 sec.
[/CODE][/QUOTE]

my math puts it at about 12 days 16 hours 50 minutes and 8 seconds from time of copying.

[QUOTE=James Heinrich;279887]I'm hurrying faster :smile:[code][Nov 25 19:13] Iteration: 9400000 / 25132963 [37.401081%]. Per iteration time: 0.021 sec.[/code][/QUOTE]

my math puts it at about 3 days 19 hours 46 minutes and 33 seconds from time of copying.

[QUOTE=James Heinrich;279887]I'm hurrying faster :smile:[code][Nov 25 19:13] Iteration: 9400000 / 25132963 [37.401081%]. Per iteration time: 0.021 sec.[/code][/QUOTE]

So, now the question becomes: do we open a "book" on who will finish first?

There could be money made (and, thus, lost, since betting on a fair game is a zero-sum-game) on the probabilities. But who will be the book-maker?

LOL...

Thanks for all the participation everyone.

This is fun!

[QUOTE=chalsall;279891]So, now the question becomes: do we open a "book" on who will finish first?[/QUOTE]Well then, I bet on me, and I'll turn up the heat :flex:[code][Nov 25 19:45] Iteration: 9499000 / 25132963 [37.794986%]. Per iteration time: 0.007 sec.[/code]That brings estimated completion to about 1d 7h.

[QUOTE=James Heinrich;279892]Well then, I bet on me, and I'll turn up the heat :flex:[code][Nov 25 19:45] Iteration: 9499000 / 25132963 [37.794986%]. Per iteration time: 0.007 sec.[/code]That brings estimated completion to about 1d 7h.[/QUOTE]

No fair! Mine was already at
[code][Nov 25 18:50] Iteration: 11900000 / 25132697 [47.34%]. Per iteration time: 0.016 sec.[/code]

[QUOTE=bcp19;279893]No fair![/QUOTE]

As my mother always told me, much to my anger (but later understanding): "Life isn't fair."

[QUOTE=chalsall;279896]As my mother always told me, much to my anger (but later understanding): "Life isn't fair."[/QUOTE]
Ever read any Hilaire Belloc?

[QUOTE=davieddy;279901]Ever read any Hilaire Belloc?[/QUOTE]

No.

Ever read any Terry Pratchett?

IMHO, his "Going Postal" and "Making Money" are his best novels.

[QUOTE=Chuck;279883]I'm hurrying as fast as I can.....[/QUOTE]

[QUOTE=James Heinrich;279887]I'm hurrying faster :smile:[/QUOTE]

[QUOTE=James Heinrich;279892]Well then, I bet on me, and I'll turn up the heat :flex:That brings estimated completion to about 1d 7h.[/QUOTE]

[QUOTE=bcp19;279893]No fair! Mine was already at <snip> 47.34%.[/QUOTE]

[QUOTE=chalsall;279896]As my mother always told me, much to my anger (but later understanding): "Life isn't fair."[/QUOTE]

Is that a race, or what? :P
May I join?

[CODE]
e:\[snip]\CudaLucas>CL4 -c10000 -t -D0 25052227
CUDALucas: Could not find a checkpoint file to resume from
Iteration 10000 M( 25052227 )C, 0x4d3b77675236265c, n = 2097152, CUDALucas v1.3alpha_eoc (0:37 real, 3.7499 ms/iter, ETA 23:04:58)
Iteration 20000 M( 25052227 )C, 0xc2017ca7f1bc661b, n = 2097152, CUDALucas v1.3alpha_eoc (0:38 real, 3.7169 ms/iter, ETA 22:50:34)[/CODE]

:P:P:P

[edit] did not check if the original test was done by P95, so I am under the risk to work in vain, but I believe at that time when this expo was first time Lucas-Leh-hammered there was no GPU around...

Wait, there's a 1.3 alpha? When did that happen?

Sorry I'll feel that again
 

[QUOTE=chalsall;279906]No.

Ever read any Terry Pratchett?

IMHO, his "Going Postal" and "Making Money" are his best novels.[/QUOTE]

I'm afraid I suffer from poor eyesight.
Fortunately we are blessed with the wonderful gift of Broil.

I'm limited to Derek and Clive ATM.

"As I was walking down the street one day,
I saw a house on fire...."

David

Grrrrr! I really begun to HATE the number 77, with all its multiplies, which I already learned by heart up to now: 154, 231, 308, 385, 462...
The hell with them all!!!:smile:

[QUOTE=chalsall;279891]So, now the question becomes: do we open a "book" on who will finish first?

There could be money made (and, thus, lost, since betting on a fair game is a zero-sum-game) on the probabilities. But who will be the book-maker?

LOL...

Thanks for all the participation everyone.

This is fun![/QUOTE]

Well, I am in the ballpark too, but using CUDALucas I'm afraid I won't be "booked"... :smile:
[code]
Iteration 21160000 M( 25977901 )C, 0xd7fcc438d40420e1, n = 2097152, CUDALucas v1.2
[/code]
Luigi

[QUOTE=Dubslow;279929]Wait, there's a 1.3 alpha? When did that happen?[/QUOTE]
Long ago, and there is the only one working well for drv 4.x with cc 2.0. If you have anew card (580, 590, Tesla) then I highly recommend it to you.

[QUOTE=ET_;279965]Well, I am in the ballpark too, but using CUDALucas I'm afraid I won't be "booked"... :smile:
[code]
Iteration 21160000 M( 25977901 )C, 0xd7fcc438d40420e1, n = 2097152, CUDALucas v1.2
[/code]Luigi[/QUOTE]

Now THAT's not fair! Unless your ETA is longer then 16 hours, which would be ok for me :P:P (ETA 15:53)

[QUOTE=LaurV;279926]Is that a race, or what? :P
May I join?
:P:P:P[/QUOTE]:cry:
A faster CPU may be unfair, but a GPU is just cheating! :razz:
Actually, it will be fairly close, we'll see who gets there first...

[QUOTE=James Heinrich;279978]:cry:
A faster CPU may be unfair, but a GPU is just cheating! :razz:
Actually, it will be fairly close, we'll see who gets there first...[/QUOTE]

What kind of CPU goes that fast if I may inquire? I thought from looking at the CPU benchmarks that the 2400 would be close to the top but you've doubled it's speed. Or did you just stop all the other workers and use all the cores?

[QUOTE=bcp19;279981]What kind of CPU goes that fast if I may inquire?
Or did you just stop all the other workers and use all the cores?[/QUOTE]i7-920 @ 3.2GHz. On 3 cores I get 0.007s iterations. On 4 cores I get 0.006 :ermm:

[QUOTE=James Heinrich;279978]:cry:
A faster CPU may be unfair, but a GPU is just cheating! :razz:
Actually, it will be fairly close, we'll see who gets there first...[/QUOTE]

And a faster GPU would be cheating cheating. But in math, minus and minus makes plus, so I am not cheating at all :smile:

But anyhow, ET is at almost the end, and I just started, so he will beat us even with the hands of his CPU tied on its back... :smile:
edit: and he is on GPU too!

[QUOTE=LaurV;279986]And a faster GPU would be cheating cheating. But in math, minus and minus makes plus, so I am not cheating at all :smile:

But anyhow, ET is at almost the end, and I just started, so he will beat us even with the hands of his CPU tied on its back... :smile:
edit: and he is on GPU too![/QUOTE]

Well, I'm not sure about that... :smile:
I asked for the exponent 25977901 BEFORE the official opening of the low LL-D system.
Since I was warned that the feature was still not running, I decided to TF the exponent from 67 bits to 69 bits and then do a LL-D with CUDA by myself.

The net result is that I can't see neither credit for the TF done nor the assignment on the Low LL-D on GPU to 72. :smile:

Hope Chalsall will help out, or you will be the first one to officially enter the stats... :bow:

Luigi

[QUOTE=ET_;279991]The net result is that I can't see neither credit for the TF done nor the assignment on the Low LL-D on GPU to 72. :smile:

Hope Chalsall will help out, or you will be the first one to officially enter the stats... :bow:[/QUOTE]

You do see the credit for the TF work you did.

And today I will create the "Assigned" records for your "tricked" assignment, davieddy's test assignment (both of which are special cases because GPUFactor still owns them) as well as the half dozen handed out by hand to requesters before the web interface came online.

So, short answer, whoever finishes first should be determinable....

[QUOTE=chalsall;279998]You do see the credit for the TF work you did.
[/QUOTE]
Yes, I do :redface: :blush:
For strange reasons I thought the wxponents were sorted.
[QUOTE=chalsall;279998]
And today I will create the "Assigned" records for your "tricked" assignment, davieddy's test assignment (both of which are special cases because GPUFactor still owns them) as well as the half dozen handed out by hand to requesters before the web interface came online.

So, short answer, whoever finishes first should be determinable....[/QUOTE]

:et_:
Luigi

Finally finished...
 

This afternoon I finally finished up a big bunch of TF exponents in the 28M range that ckdo sent me a month or so ago — this was just before the GPU to 72 project got up and running.

Now at last I'll be able to turn both instances of mfaktc to work on this project.

Chuck

delta_t wins!!!
 

With regards to who will complete the first "GPU to 72" assigned DC, please see:

[url]http://www.mersenne.org/report_exponent/?exp_lo=25094677[/url]

Or, at least, this is the first I have noticed manually, as I update the database with the appropriate "Assigned" records for those few assignments I provided via PMs.

Congratulations delta_t. And thanks for playing!!! :smile::smile::smile:

Is 25M the lower bound for the DC project, or will you be rounding up 24Ms as well?

While we're there, what about those <40M expos? Will those get picked up by the spider if they go up for reassignment? Certainly those should be assigned to a very reliable worker.

Edit: Just looked at the overall status page, and suddenly I haz a happy :):)
But a suggestion: "Note that for DC and LL TF work, divide the GHz Days by 100 [U]to get the approximate time spent[/U] when comparing with other work types" or something to that effect. Otherwise, awesome! :)
Edit 2: If we divide GPU work by 100, then we should divide CPU work by 3 to get a roughly equal comparison. Maybe instead of a footnote a whole extra column?

[QUOTE=NBtarheel_33;280039]Is 25M the lower bound for the DC project, or will you be rounding up 24Ms as well?[/QUOTE]
[QUOTE=Dubslow;280056]While we're there, what about those <40M expos? Will those get picked up by the spider if they go up for reassignment? Certainly those should be assigned to a very reliable worker.[/QUOTE]
Unfortunately those don't look possible. Primenet calls DCs below 25,000,000 and first-time LLs below 45,000,000 [URL="http://www.mersenne.org/thresholds/"]"preferred exponents"[/URL] which chalsall's anonymous spider can't get. :sad:

[QUOTE=chalsall;280033]delta_t wins![/QUOTE]
Congratulations! You were lucky I had to bring my daughter from piano lessons. When I was back the DC was finished for about 15 minutes and patiently waiting for me to report it. This was about 1:15 my time (GMT+7) so I almost could beat you (remember "need for speed" movies, with Vin Diesel and the other guy, forgot his name: "I almost had you, dude!", hehe).

[QUOTE=markr;280064]Unfortunately those don't look possible. Primenet calls DCs below 25,000,000 and first-time LLs below 45,000,000 [URL="http://www.mersenne.org/thresholds/"]"preferred exponents"[/URL] which chalsall's anonymous spider can't get. :sad:[/QUOTE]

Perhaps George can somehow tweak that part of the assignment system.

@chalsall: I believe I lost 25162937. P95 unreserved it in a very strange way, I had 4 workers, each of them with 2 (two) DC assignments, and P95 unreserved BOTH assignments of worker 2 (no joke, I am still crossed about it!) and he got a new assignment from primenet ONLY for this worker (29006071, which I am NOT going to finish, I stopped the worker and I will return the exponent back). He did not touch the work queued for workers 1, 3, and 4. What do you make of it? Were the keys for the two exponents wrong? I doubt, because in spite of the fact 25162937 was assigned from GPU-2-72, the second one (26xxxxxx) was assigned by P95/primenet (auto, I like to have more days queued just in case I run dry). That is very strange.

So I tried to recover it by using the key. Not valid anymore. It appears assigned to some anonymous user. If your spider is the one, I want it back. If not (or you already assigned it to someone else), then say so, and I have to unreserve it from GPU-2-72 page.

edit: and 25052227 which I cleared by completing DC (residue match) did not disappear from my gpu272 DC-assignment list. Are we supposed to unreserve them when we finish DC, or how it is working? (I did not get this part).

[QUOTE=chalsall;280033]Congratulations delta_t.[/QUOTE]I tried, but apparently not fast enough. :no:[code][Nov 27 01:53] M25132963 is not prime. Res64: 95A0D39CF17E690B.[/code](time is GMT-4)

@chalsal again: Now your spider completely disrespected my 8th factor! :smile:
[CODE]
M52431343 has a factor: 2313673646385899219839
found 1 factor(s) for M52431343 from 2^70 to 2^71 [mfaktc 0.17-Win barrett79_mul32]
M52431343 has a factor: 3771030653667250842383
found 1 factor(s) for M52431343 from 2^71 to 2^72 [mfaktc 0.17-Win barrett79_mul32]
[/CODE]
I can't be blamed because I took the exponent to the promised bitlevel and completed all the range without stopping, can I? :smile:

[QUOTE=LaurV;280090]@chalsal again: Now your spider completely disrespected my 8th factor! :smile:
[CODE]
M52431343 has a factor: 2313673646385899219839
found 1 factor(s) for M52431343 from 2^70 to 2^71 [mfaktc 0.17-Win barrett79_mul32]
M52431343 has a factor: 3771030653667250842383
found 1 factor(s) for M52431343 from 2^71 to 2^72 [mfaktc 0.17-Win barrett79_mul32]
[/CODE]
I can't be blamed because I took the exponent to the promised bitlevel and completed all the range without stopping, can I? :smile:[/QUOTE]

Hummm... Now that's something I had forgotten about and didn't design for... mfaktc will continue searching the entire "bit" range, even after a factor is found.

I'm afraid I'm going to have to let this (very rare) situation pass through undetected; as in, you won't get credit for the second factor on G72. You will, however, get the credit on PrimeNet.

To detect this unusual case would involve quite a bit of code, and a lot of additional spidering.

[QUOTE=LaurV;280082]What do you make of it? Were the keys for the two exponents wrong? I doubt, because in spite of the fact 25162937 was assigned from GPU-2-72, the second one (26xxxxxx) was assigned by P95/primenet (auto, I like to have more days queued just in case I run dry). That is very strange.[/QUOTE]

It is indeed very strange. Did you have your settings appropriate for the number of days of work you were asking Prime95 to hold? Obviously this is entirely within the domain of Prime95 / PrimeNet -- I have no control. But I assure you, the AIDs were valid.

Out of interest, how did you add the candidates to your worktodo.txt? By direct editing, or by using worktodo.add? And when did the candidates get unreserved? Lastly, you mentioned that a candidate which you had received directly from PrimeNet was also unreserved. This seems very strange as any candidate which has had work done on it should not be unreserved even if the AID is invalid.

[QUOTE=LaurV;280082]So I tried to recover it by using the key. Not valid anymore. It appears assigned to some anonymous user. If your spider is the one, I want it back. If not (or you already assigned it to someone else), then say so, and I have to unreserve it from GPU-2-72 page.[/QUOTE]

Nope -- which ever Anonymous account grabbed that exponent was not my spider's.

[QUOTE=LaurV;280082]edit: and 25052227 which I cleared by completing DC (residue match) did not disappear from my gpu272 DC-assignment list. Are we supposed to unreserve them when we finish DC, or how it is working? (I did not get this part).[/QUOTE]

Nope, don't unreserve completed DCs / LLs from your account. I still have a bit of spidering / parcing / database update work to do for this to be automated. Hopefully today.

[QUOTE=NBtarheel_33;280074]Perhaps George can somehow tweak that part of the assignment system.[/QUOTE]

Yes. If George set my registered spider "gpufactor" / "GPU Factoring" to be a "trusted" account, Spidy could get the lowest candidates to be TFed (and P-1'd if needed) up. The system would then release them back to PrimeNet, and my anonymous spider would [B][I][U]not[/U][/I][/B] get them -- they would go to "trusted" workers via the usual route.

[QUOTE=LaurV;280082]@chalsall: I believe I lost 25162937. P95 unreserved it in a very strange way.[/QUOTE]

I don't have an answer but I can add that there have been times in the past when I have had one worker on a 2 or 4 core system get all the work unassigned and then immediately new work assigned. IIRC...When I reported to George (a couple years ago) he suspected that for a very brief moment during server communication that either the PC reported a very low Mhz or the Server interpreted it as such and determined there was too much work for the UnreserveDays setting.

I can't say if this will work for sure but I would suggest setting your UnreserveDays higher....I often will use 200 days or more.

[QUOTE=chalsall;279875]There can (and will).

I'm just waiting for a completion.... :smile:[/QUOTE]

Now that there have been a few where might I see the results?

[QUOTE=chalsall;280099]Yes. If George set my registered spider "gpufactor" / "GPU Factoring" to be a "trusted" account,.[/QUOTE]

CPUs are trusted - not user accounts. I hacked the get-assignment code to trust your spider's CPU.

[QUOTE=Prime95;280104]CPUs are trusted - not user accounts. I hacked the get-assignment code to trust your spider's CPU.[/QUOTE]

Coolness. Thanks a lot George!!! :smile:

And, just for clarity, does that cover both the LL "CPU" (CID: "ll_work", CGUID: "29ea6276de4907906baa2f15ec0xxxx") and the DC (CID: "dd_work", CGUID: "736c310ec061628e49aee96bae82xxxx")?

[QUOTE=Prime95;280104]CPUs are trusted - not user accounts. I hacked the get-assignment code to trust your spider's CPU.[/QUOTE]

I guess this will work for TF and P-1, but DC and LL will be left out if GPUto72 need to do the assignment shuffle (release with known account, re grab with ANONYMOUS account) since the second account is ANONYMOUS?

Edit: Not fast enough on the post button! :-)

25977901 was finished this morning with CUDALucas. :smile:

The exponent wasn't obviously prime.

The residue was different from the previous run :sad:

The exponent is still hanging on my assignment page. :smile:

Luigi

[QUOTE=ET_;280109]The exponent is still hanging on my assignment page. :smile[/QUOTE]

Check your "Completed" page.... :smile:

[QUOTE=diamonddave;280107]I guess this will work for TF and P-1, but DC and LL will be left out if GPUto72 need to do the assignment shuffle (release with known account, re grab with ANONYMOUS account) since the second account is ANONYMOUS?[/QUOTE]

Which is not a problem -- it simply means that the system will do the appropriate level of TF / P-1'ing, and then return them to PrimeNet to assigned to "preferred" workers.

[QUOTE=chalsall;280111]Check your "Completed" page.... :smile:[/QUOTE]

Didn't mean to be mean... I know you actually have a lot to do :bow:

Luigi

[QUOTE=petrw1;280101]Now that there have been a few where might I see the results?[/QUOTE]

[URL="http://gpu.mersenne.info/reports/workers/"]Workers' Progress[/URL] and [URL="http://gpu.mersenne.info/reports/overall/"]Overall System Progress[/URL], plus individuals' "Completed" report.

Obviously the Overall report needs to be reworked to show the number of attempts for DC (and, soon LL). The DC row was simply put there to show that the system is picking them up.

The Overall report needs some TLC anyway, in order to show the number of LL runs (and the associated GHz Days) saved because of the TF and P-1 work which found factors.

Please note that I'm still "training my spider" ("Fetch boy, fetch. Sit. Shake paw. Good boy..."), so the completed DCs are not yet picked up "real time". But they will be soon....

[QUOTE=ET_;280116]Didn't mean to be mean... I know you actually have a lot to do[/QUOTE]

Please trust me: I have very thick skin. I take no offence from requests for promised features.

Quite to the contrary in fact; I appreciate the eagerness and all the participation. :smile:

[QUOTE=chalsall;280119][URL="http://gpu.mersenne.info/reports/workers/"]Workers' Progress[/URL] and [URL="http://gpu.mersenne.info/reports/overall/"]Overall System Progress[/URL], plus individuals' "Completed" report.

Obviously the Overall report needs to be reworked to show the number of attempts for DC (and, soon LL). The DC row was simply put there to show that the system is picking them up.

The Overall report needs some TLC anyway, in order to show the number of LL runs (and the associated GHz Days) saved because of the TF and P-1 work which found factors.

Please note that I'm still "training my spider" ("Fetch boy, fetch. Sit. Shake paw. Good boy..."), so the completed DCs are not yet picked up "real time". But they will be soon....[/QUOTE]

:bow wave:

We are not sure what your click-thru rate and earnings are for the ads on your page, but we would be willing to kick in some money to eliminate them.

:cat:

[SIZE=1]Edit: Aha![/SIZE]
[quote][SIZE=1]As an example, over on my GIMPS Visualization Tool site for example, I have installed two Google Adsense ads. Here are the statistics:

Page impressions: 192,910.
Clicks: 58.
Click Through Rate (CTR): 0.03%.
Total Earnings: $8.88. (Less than $0.50 a month; my serving costs me $10 a month. And it would take me more than 20 years before I earn the $100 payment threshold if this was the only place I had advertising.)[/SIZE][/quote]

[QUOTE=Xyzzy;280129]We are not sure what your click-thru rate and earnings are for the ads on your page, but we would be willing to kick in some money to eliminate them.[/QUOTE]

Well, to be perfectly honest and open, in the last 30 days I've had 21,614 page views, 38 click-throughs (0.18% CTR), and made a total of $18.58 USD.

The server infrastructure, meanwhile, cost me $60 USD hard cash (since this is on a dedicated server rather than a shared server), plus the approximate $6,000 USD of my development time which I consider donated (AKA "pro bono").

But if the ads are offensive, I'll remove them.

They are not offensive.

We just spend a lot of time on your pages and they get in the way a little.

Your new tool is one of the coolest ideas we have seen in a long time.

:smile:

I just noticed it records all work completed, not just stuff assigned by the system.