Sweet. Glad to hear we're able to fix this before our friend David returns.

[QUOTE=TheMawn;371911]Sweet. Glad to hear we're able to fix this before our friend David returns.[/QUOTE]

Some women I know refer to their period as "my friend is in town".

Post-menopausal women can be much more fun (if you know how to manage them)....

[QUOTE]Post-menopausal women can be much more fun (if you know how to manage them)....[/QUOTE]

Playing with fire, are we? SWMBOKd doesn't read the forum?

[QUOTE=kladner;371933]Playing with fire, are we? SWMBOKd doesn't read the forum?[/QUOTE]

She doesn't. But she was who I was referring to as being more fun (but needs to be managed). She'll kill me if she ever reads this.

This is a picture of her in the ocean at the end of the day of a birthday a year or so ago....

[QUOTE=chalsall;371936]
This is a picture of her in the ocean at the end of the day of a birthday a year or so ago....[/QUOTE]

Hummm... I see good management did wonders.

[QUOTE=lycorn;371942]Hummm... I see good management did wonders.[/QUOTE]

Let's just say I drove home.... :smile:

See, I never knew a person could use one of those balloons to float.

That stunning sunset setting is a fantastic place to wrap up a birthday celebration.

[QUOTE=TheMawn;371947]See, I never knew a person could use one of those balloons to float.[/QUOTE]

Things work a little differently near the equator. Centrifugal force and all...

We actually lose a child or two a year because some idiot gives them more than two helium balloons, and they just float away...

There's a public education program in place to try to prevent this....

[QUOTE=chalsall;371949]Things work a little differently near the equator. Centrifugal force and all...

We actually lose a child or two a year because some idiot gives them more than two helium balloons, and they just float away...

There's a public education program in place to try to prevent this....[/QUOTE]

SNORT! :goodposting:

Haha, good one, but not entirely true... You are losing them because they eat the helium and start talking funny... :razz:

When I grow up I want to have access to the firepower Oliver does!
 

[URL="https://www.gpu72.com/reports/worker/6e67460a77a11a707a665a6270df1a82/"]Wow!!![/URL] Thanks Oliver!

Thanks to this, we're now able to start going to 74 again (although we may still release a few candidates at only 73 for a while longer)! :smile:

[QUOTE=chalsall;372138][URL="https://www.gpu72.com/reports/worker/6e67460a77a11a707a665a6270df1a82/"]Wow!!![/URL] Thanks Oliver!

Thanks to this, we're now able to start going to 74 again (although we may still release a few candidates at only 73 for a while longer)! :smile:[/QUOTE]

That's not your "normal" firepower... :razz:

If he didn't double his computer park overnight, it could mean we can expect some heavy improvement of the speed for the new version of mfaktc? :w00t:

It is interesting, and somewhat clearer for me when I look at the [URL="https://www.gpu72.com/reports/worker_exact/6e67460a77a11a707a665a6270df1a82/"]Exact Graphs,[/URL] in this case. The regularity of the spikes, with the consistent P-1 red cap is impressive. :smile:

[QUOTE=LaurV;372152][...]it could mean we can expect some heavy improvement of the speed for the new version of mfaktc? :w00t:[/QUOTE]

false hope, sorry! :sad:

[QUOTE=kladner;372162]It is interesting, and somewhat clearer for me when I look at the [URL="https://www.gpu72.com/reports/worker_exact/6e67460a77a11a707a665a6270df1a82/"]Exact Graphs,[/URL] in this case. [B]The regularity of the spikes[/B][...][/QUOTE]

Chris has asked me to report my results "at least once per week" so I dump my results each Sunday.

Oliver

Grrr... Hey Chris/[George], where [URL="http://www.mersenne.info/exponent_status_tabular_data/2/30000000/"]those 15k[/URL] come from? (it seems they are the [URL="http://www.mersenne.org/report_LL/default.php?exp_lo=33000000&exp_hi=33010000&exp_date=&user_only=0&user_id=LaurV&exdchk=1&exbad=1&exfactor=1&B1=Get+LL+data"]same on PrimeNet[/URL], but why? were they forgotten, or some brain fart of the server when we moved to new web page format? Should we launch a fast TC?

It is always possible that this is deliberate to clear out the huge amount of extra errors in the 10M digit range.

How many verified & unverified did you see then in the [URL="http://www.mersenne.org/report_LL/?exp_lo=33000000&exp_hi=33010000&exbad=1&exfactor=1&B1=Get+LL+data"]primenet LL result report[/URL] you linked to, LaurV? Now, it shows 112 distinct exponents with verified LL and 94 with unverified LL, for the same total as mersenne.info with its 110 2LL, 71 1LL and 25 0LL. The primenet summary does not show that bump either. I think it's a glitch in mersenne.info; it's happened before. Trust nothing!

[QUOTE=markr;372502] I think it's a glitch in mersenne.info; it's happened before. Trust nothing![/QUOTE]
[STRIKE]What glitch? There is no discrepancy between the numbers. What I linked on gimps page is just a different range, because I could not link all the range, too much time to display. But both PrimeNet and mersenne.info show the [B][U]same[/U][/B] data, i.e. the fact that there are [B][U]15k[/U][/B] exponents [B][U]between 33M and 34M[/U][/B] which are [B][U]not LL-tested, not even once[/U][/B], which is absurd. According with the milestones, everything below [B][U]51M[/U][/B] or so, is tested at least once. I think this was an error introduced once the PrimeNet was ported to the new format, or some chunk of the database is lost/destroyed. We may need, or not (George should know better where this problem comes from) to TC some of those exponents, just to be on the "safe" side. Do we? That was my question.[/STRIKE]

Ok, scrap that! I think this is intentional. Doing some totals on gpu72 candidates, I believe some exponents were just manually moved to "no LL test done" category, to increase their minimum TF bitlevel (from 70 to 71). If they "are DC-ed, i.e. done", then the GPU72 spider would not pick them for TF. This means we did (and still doing) TF on GPU72 for some 33M expos which were already DC-ed. Well, it would be nice to move everything to 71 there, but as we are low on resources, we better stop doing this, and concentrate on expos in the higher ranges, 36M, or so, which really didn't have DC done and need TF...

[QUOTE=LaurV;372522]Ok, scrap that! I think this is intentional. Doing some totals on gpu72 candidates, I believe some exponents were just manually moved to "no LL test done" category, to increase their minimum TF bitlevel (from 70 to 71).[/QUOTE]

Nope...

Keep in mind that Mersenne.info and GPU.com are completely independent. The database behind the former is actually based on a server here in Barbados which spiders Mersenne.org once a night, and uploads the summaries after the spidering is completed.

I'm not sure what happened here, but I'm going to wait until tonight's spidering has finished, and see if the problem corrects itself.

And, lastly, GPU72 is not assigning anything for DCTF'ing which is not still a real DC.

That's good to know! :razz:

[QUOTE=TheJudger;372203]Chris has asked me to report my results "at least once per week" so I dump my results each Sunday.[/QUOTE]

Bloody hell!!! A new [URL="https://www.gpu72.com/graphs/worker/ghzdays_exact/6e67460a77a11a707a665a6270df1a82/"]GPU72 record: just short of 100,000 GHz Days done by one participant in seven days![/URL]

Thanks Oliver! We should now being able to start bringing back in candidates released at only 73 as well as those at 72 as they're recycled!

[QUOTE=chalsall;372623]Bloody hell!!! A new [URL="https://www.gpu72.com/graphs/worker/ghzdays_exact/6e67460a77a11a707a665a6270df1a82/"]GPU72 record: just short of 100,000 GHz Days done by one participant in seven days![/URL][/QUOTE]

Jaysus... What is his hardware normally doing?

Overall worker reports are sane again...
 

Just to let everyone know...

I finally had some cycles yesterday to figure out the SPE which was causing the [URL="https://www.gpu72.com/reports/workers/"]Worker's Overall Progress Reports[/URL] to include in the "Done" column work which was actually unreserved instead (naturally, it was all of about eight characters in a part of a very complicated SQL query's conditional).

Please let me know if anyone sees anything strange. I'm Only Human, and might have very well broken something else fixing this....

[QUOTE=chalsall;372641]Just to let everyone know...

I finally had some cycles yesterday to figure out the SPE which was causing the [URL="https://www.gpu72.com/reports/workers/"]Worker's Overall Progress Reports[/URL] to include in the "Done" column work which was actually unreserved instead (naturally, it was all of about eight characters in a part of a very complicated SQL query's conditional).

Please let me know if anyone sees anything strange. I'm Only Human, and might have very well broken something else fixing this....[/QUOTE]

Thanks for fixing this — I just noticed today that the report had been corrected.

The nine that show tf'd to 72 on this page are actually tf'd up to 73 (as of awhile ago)

[url]http://www.mersenne.info/trial_factored_tabular_data/4/336000000/[/url]

[QUOTE=pdazzl;372663]The nine that show tf'd to 72 on this page are actually tf'd up to 73 (as of awhile ago)

[URL]http://www.mersenne.info/trial_factored_tabular_data/4/336000000/[/URL][/QUOTE]

The Spider only updates once a day (assuming you believe this to be a bug).

Tf'd to 73 about eight days ago, so yes appears buggy

[QUOTE=pdazzl;372675]Tf'd to 73 about eight days ago, so yes appears buggy[/QUOTE]

Some might consider this a bug (of Mersenne.info, not GPU72.com), but it's actually by design...

To lessen the load on Primenet, the Spider only "looks" at ranges (by 1M blocks) which have had a change in state as it appears on the [URL="http://mersenne.org/primenet/"]PrimeNet Activity Summary[/URL] report, plus another 30 or so sorted by last updated date.

It would be pointless and wasteful to spider all 1,000 ranges every day, and for those ranges actually important the reports update daily (or, more specifically, when there's a "state change"). The "state change" which triggers a scan is a factor being found, a LL or a DC being completed, or a change in the number of assignments.

So, if you really want to see this specific sub-range reflect the newly TF'ed candidates from 72 to 73 "bits", either do some more TF'ing and find a factor, manually get an assignment from that range and hold it for 24 hours, or wait about a month....

[QUOTE=chalsall;372641]Please let me know if anyone sees anything strange. [/QUOTE]
Still small numbering problem (like double 14, double 25, etc, even if the credits are totally different). But generally the page looks much better now.

[QUOTE=LaurV;372705]Still small numbering problem (like double 14, double 25, etc, even if the credits are totally different). But generally the page looks much better now.[/QUOTE]

Yes, I know about that... I'm working on a better fix, but right now two users who haven't been updated recently (in addition to any which have had a "state change" (read: completed work, received work)) are updated every time the script runs (twice an hour).

As the numbering issue only occurs when someone passes another, this should largely mask the issue until I implement the better solution (I have it modeled, but it's a bit of code, and I also have "real work" to do...).

[QUOTE=petrw1;369444]Did you happen to check about another 20 spots higher....?[/QUOTE]
Which "two spots" are you talking about? :huh:
(going to be "minus one" spot soon:boxer::showoff:)

[QUOTE=LaurV;373114]Which "two spots" are you talking about? :huh:
(going to be "minus one" spot soon:boxer::showoff:)[/QUOTE]

Wow, either you have a good memory, or was reading the history, or... or....

No, I could say nothing that time, so I put it on a black list on the wall and kept it till I had something to say :razz: And why do you care? This is between me and Syd, hehe...

[QUOTE=LaurV;373114]Which "two spots" are you talking about? :huh:
(going to be "minus one" spot soon:boxer::showoff:)[/QUOTE]

Hmmm...moving fast....About 15,000 in 90 days...No I can't touch that.

I am about to be humbled.... :deadhorse:

FWIW, my system will be off line for about 3-4 days. We are taking a modest road trip down to the southern end of Illinois to visit a favorite State Park (Giant City), and National Forest (Shawnee).

I don't like to leave things running unattended for too long, so about 850 GHz-days/day will drop out. A couple of dozen TF assignments may get a little long in the tooth, but should get finished before they age out.

[QUOTE=kladner;373199]FWIW, my system will be off line for about 3-4 days. We are taking a modest road trip down to the southern end of Illinois to visit a favorite State Park (Giant City), and National Forest (Shawnee).[/QUOTE]

OK, thanks for the head's up. And enjoy your trip!

So everyone knows, Oliver has done it again -- 133,000 GHz days in seven days! Thanks Oliver!!! (Although he warned me yesterday that his throughput is going to "return to normal" (normal for Oliver, that is... :wink:) for the foreseeable future).

Further, with our current firepower, we've been able to sustain about 300 LLTF's to 74 per day, mostly from 72. As in, we, are now able to sustain going to 74 and keep the LL'ers and P-1'ers fed, plus bring in those candidates released at below 74 as they are recycled.

Lastly, some interesting statistics... Although the Cat 4 "churners" ask for approximately 1,000 assignments per day, only approximately 72 are completed. Thus, we should be able to recover ~93% of those sub-optimally released.

This does *not* mean we're out of the woods yet -- please keep the cycles coming. In about two months we should have some breathing room again.

Thanks for everyone who's helping with this effort. :smile:

[QUOTE=chalsall;373211]Although the Cat 4 "churners" ask for approximately 1,000 assignments per day, only approximately 72 are completed. [/QUOTE]

That's insane...

[QUOTE=TheMawn;373277]That's insane...[/QUOTE]

No. That's reality.

These two things are not mutually exclusive.

For some of us, they are exclusively mutual :razz:

[QUOTE=TheMawn;373277]That's insane...[/QUOTE]

I reissue my request to have interim LL results saved on the server (maybe an option for discussion since George opened the "next version of PrimeNet" can-of-worms.

Corallory....only if there is a minimum of X% progress ... maybe 25% ... enough to make it worth while storing and passing on the interim work vs starting over.

I wonder how many of the other 99% that would get....maybe not much if most are abandoned with no progress.

[QUOTE=petrw1;373333]I wonder how many of the other 99% that would get....maybe not much if most are abandoned with no progress.[/QUOTE]

Most are abandoned with no, or very, VERY little, progress.

While a good idea in theory, I suspect the bandwidth, and programmer effort, involved is not worth it.

I'm thinking the bandwidth issues would be the toughest.

If 100 million digits ever gathered a bigger following, they might have more interest in something like that. Saved checkpoints would be the holy grail of doublechecks.

[QUOTE=TheMawn;373367]I'm thinking the bandwidth issues would be the toughest.[/QUOTE]

Bandwidth (and hardware) are cheap. Humans are expensive (and error-prone).

[QUOTE=chalsall;373368]Humans are expensive.[/QUOTE]

The good ones, anyway.

34,000,000 - 35,000,000 69bit
 

Could you bring in the exponents in the 34M-35M DC range, those that are TFed to 69bits so I can take them to at least 70bit (maybe even 71bit if we have the time). I can't request them on GPU72 and they don't show up as assigned on the factoring effort report.

[url]http://www.mersenne.org/report_factoring_effort/default.php?exp_lo=34000000&exp_hi=35000000&bits_lo=69&bits_hi=69&exassigned=1&B1=Get+Data[/url]

All exponents in the range which are not yet DC appear in [URL="http://www.gpu72.com/reports/current_level/"]this list[/URL] (second table), and they ARE reserved by GPU72 (or "own" as some say) already. You can take them from GPU72 by manual reservation (the automatic will give you expos in 36M, which help gimps more right now). Other expos in your link which are not "proper" TF-ed, they already are double-checked, and please be aware that TF-ing them to 70 or 71 do not help GIMPS at all. [edit: of course, everybody is free to do what work type he likes, I am still strongly adhering to this thinking!]
[edit2: the factoring effort reports don't usually show if the expo is reserved or not. The "[URL="http://www.mersenne.org/report_exponent/default.php?exp_lo=34000000&exp_hi=34500000&full=1"]expo status[/URL]" reports, usually do show that]

My mistake, I didn't realize they were already DCed. I should have checked their status (LL/DC) before posting.

[QUOTE=VictordeHolland;373429]My mistake, I didn't realize they were already DCed. I should have checked their status (LL/DC) before posting.[/QUOTE]

They aren't already DC'ed -- LaurV misspoke... :razz: :wink:

I actually "threw those back" to Primenet a couple of days ago. The reason was that the 34M range is at the "tail end" of the DC "Category 3" range -- we have so much of a buffer for Cat 1, 2 and 3 that these won't be assigned by Primenet for another six months to a year from now. As in, working on them is not currently a high priority.

On the other hand, we were falling behind with appropriate TF'ing for the Category 4 range. I actually had to bring some of our LLTF'ing firepower (Jerry, LaurV and myself) down there for a couple of days to build up some candidates -- and we really can't spare the LLTF'ing firepower at the moment.

However, I have brought the 34M range back into GPU72 for those who really want to work that range. Please know that the "What Makes Sense" and "Lowest TF level" options default to 35M or higher. "Lowest Exponent" will honor the request (read: 34M candidates, as long as they are available).

Based on the above, while you are (of course) free to do whatever work tickles your fancy, it would be appreciated, and of greater short-term benefit to GIMPS, if you did at least half your TF'ing work at 35M or above.

Whoops! :redface:
Sorry, I didn't know the details, but the posted lists (I mean both, mine included) contained a lot of already-DC-ed exponents too.
Sorry for this confusion, and for any harm I caused :smile:

Regarding this discussion, is Cat4 DCTF or Cat4 LLTF a bigger concern at the moment? I would have to imagine the LLTF is where the biggest strain is, unless we've built up enough of a buffer for the churners.

Thanks for the clarification. I'll try to split my humble power between 34M and 35M, not that my 300 GHzdays/day is going to make a huge impact.

[QUOTE=TheMawn;373450]Regarding this discussion, is Cat4 DCTF or Cat4 LLTF a bigger concern at the moment? I would have to imagine the LLTF is where the biggest strain is, unless we've built up enough of a buffer for the churners.[/QUOTE]

Cat 4 LLTF is the biggest issue at the moment.

Surfing the edge of the wave, baby! Just the way we like it! :wink:

[QUOTE=VictordeHolland;373452]Thanks for the clarification. I'll try to split my humble power between 34M and 35M, not that my 300 GHzdays/day is going to make a huge impact.[/QUOTE]

Every cycle is valued. And you might be surprised by how important only 300 GHd/d is to feed the DC Cat 4.

[QUOTE=VictordeHolland;373452]not that my 300 GHzdays/day is going to make a huge impact.[/QUOTE]

I've got about 600 myself and I'm in the top-20 yearly GIMPS contributors for TF, so 300 itself is quite substantial. Every bit helps, but don't be surprised by your impact over time.

[QUOTE=TheMawn;373519]I've got about 600 myself and I'm in the top-20 yearly GIMPS contributors for TF, so 300 itself is quite substantial. Every bit helps, but don't be surprised by your impact over time.[/QUOTE]
Yes, I'm well aware of that. I'm currently ranked #60 this year and #49 overall for TF. A few days ago I reached the 100,000 GHzdays overall milestone :). I can scale up to 1000GHzdays/day with GPUs that are collecting dust at the moment. But the heat, noise and electricity costs are preventing me from doing that (electricity is very expensive in the Netherlands).

Yes, every little bit helps(I suppose).

But I still feel/[I]am [/I]microscopic with 360GHz.

[QUOTE=kracker;373555]But I still feel/[I]am [/I]microscopic with 360GHz.[/QUOTE]

Hey, I can currently bring less GPU firepower to bear than that (because the only machine which I currently have control over and can host a GPU does "real work" during the day), and I run GPU72! :smile:

As helpful as ~600 GHz-days/day might be, I am going offline starting tomorrow. I'll be up and running again on May 24.

I shall be off to Paris. Looking forward to the show at Le Moulin Rouge.


To be honest, the little superstitious part of me has a great feeling about this trip. I grab my TF assignments from GPU72 in batches of 100 per GPU, and both GPUs are looking like they will finish [I]just[/I] before I have to turn off my PC. I have a workout scheduled for today, and I have exactly enough episodes of the TV show I am watching to finish the season today, and exactly enough ingredients to make my post-workout smoothie.

[QUOTE=TheMawn;373576]As helpful as ~600 GHz-days/day might be, I am going offline starting tomorrow. I'll be up and running again on May 24.

I shall be off to Paris. Looking forward to the show at Le Moulin Rouge.[/QUOTE]

Thanks for the head's up (and you already pre-announced this, but thanks for the follow up).

I hope you can speak French to the French, to their tolerances.

I understand (from my sister, who can speak four languages fluently, and lived in France for four years) that the French generally refuse to understand English, even if they can.

And that the French in France look down on those who speak "Quebec French" as being "too colloquial", and the Quebec French consider the France French as "too snooty".

What is the Chinese saying? Oh, yeah... "May you live in interesting times....

[QUOTE=chalsall;373591]I hope you can speak French to the French, to their tolerances.

I understand (from my sister, who can speak four languages fluently, and lived in France for four years) that the French generally refuse to understand English, even if they can.

And that the French in France look down on those who speak "Quebec French" as being "too colloquial", and the Quebec French consider the France French as "too snooty".

What is the Chinese saying? Oh, yeah... "May you live in interesting times....[/QUOTE]

Hahahaha that is my favourite Chinese saying ever. I'll be sure to use it whenever I can from now on.

Well I belong to the Québec French in all above aspects. The only notion that isn't 100% correct is that they will refuse to understand English. In my experience, they will prefer English to Québecois.

On the other hand, if you starve them of French for long enough, they'll take whatever they can. Our French summer students at the lab last year started speaking to us in French again after a few days.

[SUB][SUB]Nyehh nyehh heh heh hehehehehe[/SUB][/SUB]


EDIT: What is interesting however is I find it a lot easier to step my French up to acceptable levels than English, despite being better at English. My English is already at a higher level, but I could please a snooty Frenchman much more easily than a snooty Englishman. Still, I'm sure the snooty Englishman would accept to understand me. A few years ago, we had a French summer student who was pretty high on the snootiometer. I was six or seven sentences in until he realized I was speaking French. And no, my French isn't that bad...

yep, this is the worst curse ever....

If anyone is successful in getting CUDA running on Ubuntu 12.04.4 or 14.04 please let us know.

[url=http://www.mersenneforum.org/showpost.php?p=365319&postcount=2301]We had it working via the package manager on 12.04.1[/url] but it is horribly broken now.

:help:

[QUOTE=Xyzzy;373606]If anyone is successful in getting CUDA running on Ubuntu 12.04.4 or 14.04 please let us know.

[URL="http://www.mersenneforum.org/showpost.php?p=365319&postcount=2301"]We had it working via the package manager on 12.04.1[/URL] but it is horribly broken now.

:help:[/QUOTE]
Works for me with driver and CUDA-Package from nvidia site.

BTW: I just finished my overdue assignments today and will take down my cards for a while (about 2500 GHz/d per day) due to heat issues. I definitely need to go for water cooling...

[QUOTE]Works for me with driver and CUDA-Package from nvidia site.[/QUOTE]You are talking about the .RUN installer? If so, we tried that and it failed to install numerous times.

[url]https://developer.nvidia.com/cuda-toolkit-55-archive[/url]

[QUOTE=blip;373649]Works for me with driver and CUDA-Package from nvidia site.

BTW: I just finished my overdue assignments today and will take down my cards for a while (about 2500 GHz/d per day) due to heat issues. I definitely need to go for water cooling...[/QUOTE]

You could try under-clocking them to keep thermals under control. I've pulled down one of my overclocks for that reason, though the weather has turned back cool* after a week which hit 90 F, so it probably isn't necessary just now. [INDENT]*including reports of snow in the Chicago suburbs, away from the lake.
[/INDENT]

[QUOTE=kladner;373657]You could try under-clocking them to keep thermals under control. [/QUOTE]
How do you do that on a headless linux system?

[QUOTE=Xyzzy;373655]You are talking about the .RUN installer? [/QUOTE]
CUDA 6:
[CODE]https://developer.nvidia.com/cuda-downloads[/CODE]

We did not know CUDA 6 worked. Did you use the .deb or .run install?

[QUOTE=blip;373658]How do you do that on a headless linux system?[/QUOTE]

Oops. Sorry.

[QUOTE=blip;373658]How do you do that on a headless linux system?[/QUOTE]

I can't remember where on the forum it is, but owftheevil gave some guidelines as to how to change the clock settings in the BIOS for nvidia cards. Needs Windows (or, at least, a DOS prompt) and a display and keyboard during the changes, but not then again -- the changes are retained during reboots / power cycles.

It worked great for me (fortunately the machine currently hosting my GPU is dual-boot) -- I had to use the technique to down-clock the card's memory to get stable CUDA LLing.

[QUOTE=Xyzzy;373660]We did not know CUDA 6 worked. Did you use the .deb or .run install?[/QUOTE]
Try .run. But be sure to also use the "right" driver, and get rid of nouveau first.

[CODE]http://www.nvidia.com/download/driverResults.aspx/75340/en-us[/CODE]

[QUOTE=VictordeHolland;373554] (electricity is very expensive in the Netherlands).[/QUOTE]

Just out of curiosity, how expensive?

In Portugal I pay 20.18c / Kwh (day rate) and 10.7c / Kwh (night rate). Prices include 23% VAT.

[QUOTE=lycorn;373673]Just out of curiosity, how expensive?

In Portugal I pay 20.18c / Kwh (day rate) and 10.7c / Kwh (night rate). Prices include 23% VAT.[/QUOTE]

Wow. Here it is 10-12 Kwh. (without other charges, which...)

[QUOTE=kracker;373675]Wow. Here it is 10-12 Kwh. (without other charges, which...)[/QUOTE]

From what utility? The marginal rate is what matters, as your daily life uses up all the tier 1 and likely much of the tier 2 power rates. If you powered down the computer, you'd be using less of the most-expensive power, but still all of the cheap 10c/kwh stuff.

For my Riverside muni power, tier 2 is 16c/kwh, and I rarely hit tier 3 which is 21. For SoCal Edison, most of my friends are on tier 3 which is ~24c/kwh, while some larger families are on tier 4 ~27c/kwh.

-Curtis

Wow, I never realized how cheap our electric is compared to most.
The first 900 kWh per month is 6.4313 c/kWh, with additional energy costing 5.2575c in winter, and 7.5589c in the summer. Of course, additional charges are in there, too, for infrastructure and such.

about 21 c(€) per KWh + VAT, and still increasing...

Electricity prices have deceptively low numbers when looking at the utility bill here, but then there's a multitude of fees and surcharges that bring it up. Speaking in Eurocents for ease of comparison, the advertised off-peak rate is as low as about €0.05/kWh, but once you include delivery and debt reduction and servicewhatevertheycallit charges, the average here becomes about €0.125/kWh (incl 13% VAT), which globally isn't bad. Especially when you'd be burning that power anyway to heat the room, may as well heat it with a GPU :smile:

Last I looked at my bill, a cursory glance might have given the impression of a rate of US $0.06/kWh. A closer look showed 2 lines which had value tied to usage, as opposed to the various fixed surcharges. This brought the /kWh part of the bill to something over 12 cents per.

We were never so clever to understand how the Thai tariffs works related to the electricity, and we never understood anything on the invoice/bill except few numbers, like how many kWh we spent, and how much money we paid. Usually we divide them, but it seems that we are stupid with division, because every month we get a different cost per kWh, always between 13 and 20 US cents, mostly around 16. We believe that the tariffs are differentiated by day/night, summer/winter, rain/sunny (?!?), being enterprise/company or private person, being Thai or farang. No joke about the last one, in Thailand the practice is quite common, the government encourage the locals to practice differentiate prices for foreigners (public parks, cinema, different shops, etc, have prices for tourists which are from double to 20 times higher than for local people - and because we are pale faces, we can't hide it :smile:, hehe. We use to have a special card from the revenue department saying that "this farang works here and pays its taxes as local people", which goes together with the passport helping us to pay 20 baht to visit some park/temple/etc, like local people, instead of 400 baht, like tourists).

[QUOTE=lycorn;373673]Just out of curiosity, how expensive?

In Portugal I pay 20.18c / Kwh (day rate) and 10.7c / Kwh (night rate). Prices include 23% VAT.[/QUOTE]1-10.000KWh :
07:00-21:00 (high tarif)
7.223c + 11.65c (energytax) + 0.11c (green energytax something) + 3.986c (21% VAT)= 22.97c / KWh

21:00-07:00 (low tarif)
5.603c + 11.65c (energytax) + 0.11c (green energytax something) + 3.646 (21% VAT)= 21.01c / KWh

Above 10.000KWh:
Energytax reduced to 4.24c and the green energy tax is a bit higher 0.14c .

So, its mostly tax, not the energy itself that makes it so expensive.

[QUOTE=VictordeHolland;373694]So, its mostly tax, not the energy itself that makes it so expensive.[/QUOTE]

Here in Barbados, the cost of electricity is a little scary...

For "Domestic service", there is an automatic $10 monthly "Customer Charge" for most consumers. $6 if you use less than 150 kWh (but who does?), and $14 if you use over 500 kWh.

Then:
First 150 kWh, the rate is $0.150 per kWh.
Next 350 kWh, the rate is $0.176 per kWh.
Next 1000 kWh, the rate is $0.200 per kWh.
Next 1500 kWh, the rate is $0.224 per kWh.

That might appear inexpensive, but there is then the "Fuel Adjustment Clause" (FAC) charge, which is the cost "The Company" paid for the fuel burnt to generate the power... (Yeah, we're 13.2 degrees above the equator, and yet we still mostly import and burn long-dead creatures for our power rather than use sunlight or wind.)

Last month the FAC charge was $0.392921 per kWh, regardless of the consumption range. As in, the FAC charge is the same for every kWh consumed.

All of the above prices are in Barbados dollars, and then subjected to a 17.5% VAT. To convert Barbados dollars to US of A dollars (to which we are "tied"), divide by 1.98.

So, for my very modest home which consumed 392 kWh's last month, the bill was $269 Barbados, or $136 USD.

Or... Approximately $0.347 USD per kWh....

Well the prices I posted for Portugal are in eurocents (~1.37 x US dollar cents), and are just for the energy actually consumed. You also have to pay an overhead of some 10-11€ a month for infrastructure. The cost of the Kwh is independent of the quantity you use. We have that system for water, but not for electricity.

@blip: the price you posted is really scary, as VAT is not included! Where in Europe are you posting from?

Last month in UK charge was equivalent to 23.184 us cents/kWh plus 5% VAT ...

Living in Sweden with a large part hydroelectric power and nuclear power gives me a price of 43 öre/kWh, circa 4.5 euro cent or 6,15 Us cent.
I also have to pay a fix price for my connection to the power grid, at my current consumption that is circa 10% more.

Now I understand you guys complaining of the high cost of running GPU:s :razz:

I forgot to convert it to US Dollars, 0.2297 euro/KWh is approximately 0.31 US dollars/KWh.
Still a bit cheaper than Barbados and our electricity gets cheaper (in euro/KWh) if you use more, while yours gets more expensive :/ .

Our "green" electricity is not that green at all, the Netherlands still has 5 or 6 power plants that burn coal! The rest is mostly gas powered and 1 is nuclear I believe. And all those gigantic wind turbines only provide 5% of our power needs. Luckily we still have our Heineken beer and can smoke pot legally.

Nice, I've an avatar, don't know who put it there, but thanks, I like it!

Last month 12.45¢/kWh bottom line — just the total bill divided by usage. There are no tiered charges from my electric cooperative. The water company began using that pricing structure some time back.

Somewhat related to all this: On my [url=http://www.mersenne.ca/mfaktc.php]mfaktc performance chart[/url] the JVR / JVR2 columns factor in electricity usage, and I've set the value of that at $0.25/kWh. Is that a reasonable average?

A few AMD price updates :smile:

R7 240 is [URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600473872&IsNodeId=1&name=Radeon%20R7%20240"]$70[/URL].
R7 250 is [URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600473873&IsNodeId=1"]$90[/URL].
R7 250X is [URL="http://www.anandtech.com/show/7751/amd-announces-radeon-r7-250x-shipping-today"]$99[/URL].
R7 260 is $109.
R7 260X is $139.
R7 265 is ~[URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600489396&IsNodeId=1"]$160[/URL].

R9 270 is [URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600481061&IsNodeId=1"]$180[/URL].
R9 270X is [URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600481061&IsNodeId=1"]$199[/URL].
R9 280 is [URL="http://www.anandtech.com/show/8019/amd-cuts-radeon-r9-280-to-249"]$250[/URL].
R9 280X is ~[URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600473876&IsNodeId=1"]$300-$310[/URL].

[URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600473877&IsNodeId=1"]R9 290[/URL] and [URL="http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20600473871&IsNodeId=1"]290X[/URL] pricing is all over the place... but not $552 and $674.

GPU pricing is in Canadian Dollars. Pricing is taken semi-live as an average selling price from a major (Canadian) retailer. If you happen to live in the land of cheap then lucky you. USA prices are probably 25% less (our government needs the money to pay the utility bill to keep this place frozen all year long).

Getting a bit more on topic of the website again: The "Individual Overall Statistics"-page doesn't show up any assignments completed in the 70M or 71M-range.
That should be only pulling over one or two switches, right? ;)

Wow!!!
 

I'm beginning to get the impression that Oliver might 0wnz EC2.... :wink:

[URL="https://www.gpu72.com/graphs/worker/ghzdays_exact/6e67460a77a11a707a665a6270df1a82/"]20,000 GHz days/day for the last week![/URL]

Thanks mate! This helps a lot.

[QUOTE=manfred4;373770]Getting a bit more on topic of the website again: The "Individual Overall Statistics"-page doesn't show up any assignments completed in the 70M or 71M-range. That should be only pulling over one or two switches, right? ;)[/QUOTE]

Good catch! Thanks.

I've commented out the "... and Exponent<70000000" constraint in the SQL query. (GPU72 has been running for a lot longer than I ever imaged -- it was originally envisioned to be short-term sub-project.)

In an "order by fire power", I see a "[URL="http://www.gpu72.com/reports/worker/c87ea45fa5e7920bc6d4a431af1c3698/"]NickOfTime[/URL]" coming very strong from the back. With about 2THzD/D, that sounds like 4 Titans or a lot of AMD fire power. With ~29G/factor, it means he is doing the last bit, for which the 39 factors from 46 expected sounds statistically good. So, he is not cheating, he [U]has[/U] a lot of fire power.

So.. Who is he? Same "Nick" from the forum, or new meat? :razz:
Congrats anyhow, and kotgw!