Hi to BBC Climate Crunchers - New to Team.

PeteMcCann · 13 Sep 2006 at 22:33

Hi there. I'm not sure if this is the right place to post, but I'm sure you'll put me right if not. I was reading the BBC forums and they were encouraging people to join a team. As I have just started on my second pair of models, I thought I'd join up as I also want to get into overclocking. Yes all right, I am an overclocking virgin (except for AIBooster) but you have to start somewhere. I'm currently running 2 models on my main PC and a 3rd on my kids. Hoping to build some other PCs to boost the output further. I'm also rather new to this kind of forum, so you'll have to forgive me if I'm not up to speed.

Anyone out there say Hi.

Cheers

Pete

PeteMcCann · 13 Sep 2006 at 22:56

You're into folding are you? I've read about that in Custom PC mag. One of your competitors? Is folding proteins more fun than watching the polar ice caps melt? Did you finish your model through to 2080? How long did it take you and what spec machine did you run it on? It doesn't seem to be the most organised of projects especially the great **** up when everyone had to restart new models from the beginning. I was sick as a parrot. Still it felt good to finish 2 complete models, but it sure takes some time.

Pete

PeteMcCann · 14 Sep 2006 at 11:47

shadowscotland said:
HI
- but with three PC's your'll be clocking up the points/models in no time.

Well not three exactly. My main PC is an X2 4400+ on an A8N-SLI Premium with 3GB of Corsair 3200C2-Pro. This was able to run two of the BBC climate models very effectively, one on each core, and only just slower than running a single model. Even running 24/7 it still took from April to September to finish these first model runs which are a 160 year climate simulation from 1920 to 2080, so I don't think I'll be zipping through anything very quickly at this rate of knots. It would be nice to squeeze some more juice out of the processor though. I seen that some people have been able to raise this chip from 2.2 to 2.6-2.7Ghzs by overclocking. I'll settle for that!! :cool:

Why do you find folding so great? Are there good graphics? You'll have to work hard to convert me to folding though.

Cheeers. Pete.

PeteMcCann · 14 Sep 2006 at 12:03

joeyjojo said:
Folding is the big craze at the moment.

Well it does give all these massively powerful machines something to do. When you think that even a single machine has more computing power than was used to send man to the moon, and most of the time they sit there with system idle at 99%. Certainly the speed of my typing doesn't tax my machine very much!

I've known about SETI for ages but it never really appealed. I think there's didly squat up there, and even if there is, why would they come here!! I did get rather more interested in DC when I started reading Custom PC Mag as they have a folding team aswell. I was going to start that untill the BBC climate experiment came along, which was much more akin to my interests, so I signed up and have been crunching since February this year. It felt good to be giving my PC a really good workout. Before I could never get it to break a sweat. Now it is always maxing out at 100%. It will keep my study nice and warm this winter as a fringe benefit!

If you haven't seen the graphics on the BBC model, they are worth a look and fun to play with.

Cheers. Pete

PeteMcCann · 14 Sep 2006 at 21:55

Amp34 said:
Are you Pete? If so you just pushed me from 4th place

Yes I am Pete!! Sorry for storming in like that. I had not seen the team table before I joined, so was rather suprised to have wizzed up to second place. I thought I'd be way down the table with all you fanatical crunching overclockers. It sound like most people are into folding though. Still, someone has to save the planet!! There's not much use in advancing medical science if we are all under water now, is there!!

Perhaps we can get these folders to get their priorities right!!!

Are you crunching on one machine or multiple? Did you have to reset your model in April or did you start after that. Do you just do the BBC DC or are you doing others as well?

Cheers. Pete

PeteMcCann · 15 Sep 2006 at 00:34

PhilthyPhil said:
Welcome to the forums and Team OcUK
Personally I'm not into BBC Climate Change at the moment although I might be in the future since I have moved around four or five BOINC projects so far and it could be on my list of projects to do next I might be being a bit thick, but where are the stats for this project? I can't see them on climateprediction.net or bbc.cpdn.org

Hi there. I think you should find the info you want at the following

http://bbc.cpdn.org/advanced.php

Yes, do come and run a climate model. Our team is currently 9th overall and could move up to 5th with a few more producing members. It has some great graphics too, which you can have as a screen saver, though I think it takes quite a lot of processing juice.

Mind you, the models are not for the faint hearted or if you want instant results. They run 160 year simulations which take fast computers 5-6 months to complete 24/7!! That's some comittment, though they run very sucessfully on proper dual core cpu's, one on each core, so 2 for the price of one! And they were interesting to watch too. Both my models had no arctic ice during the summer from about 2050 onwards. That's only 40 years away.

"We're going to need a bigger boat!"

Pete

PeteMcCann · 16 Sep 2006 at 23:29

joeyjojo said:
Not a criticism of the project, but isn't it ironic that you leave your crunchers on 24/7, which uses more electricity, which produces more CO2 (indirectly ), which ends up changing the climate more?

There has been much discussion of this on the BBC forums. The easy answer is "turn off your monitor, speakers, and some lights, and balance up the extra useage".

A slightly deeper answer is to say that making any environmental improvement involves initial cost outlay which results in long term benefits / savings. Fitting loft insulation / doubleglazing / draft excluders all involves cost and effort, but gives a long term benefit of lower heating bills and reduced use of energy. It actually costs more to collect, sort and process recyclable rubbish than to throw it unsorted into landfill. It costs money to propery dispose of waste / sewage to have clean rivers etc. There is a large cost to setting up a wind farm and then having a renewable source of energy. However you have to weigh up the initial cost against the long term benefits.

I can recycle all my rubbish, buy local food, make my own energy all I want but globally my familly impact is just a drop in the ocean. However, when you multiply up indiviual actions becoming widespread general behavior, then you can start having an impact on the problem. CFC's have been eliminated from most fridges and freezers globally and in quite a short space of time.

I personally see the BBC climate model as a way in which I can use my own resourses (ie my fast and underused computers) to add to the science of climate change, which hopefully will feed into the political debate and then actually effect some change in global policy making. Yes, I am using some extra energy now to run the models, but if it results in the scientists having a much better global climate model and being able to make more certain predictions, and the probabilities of likely outcome being better know, then I think that would be well worth it. A little more energy now, for the chance of pulling us back from the brink, is money well spent in my book.

PeteMcCann · 16 Sep 2006 at 23:31

br83taylor said:
humm tempted too, maybe put one of my SETI machines to other work, once i've passed my next milestone

Go on, you know it makes sense, and the graphics are great too. Watch that globe melting down!

Pete

PeteMcCann · 19 Sep 2006 at 01:22

Pumpkinstew said:
I'm quoting this as one of the most succinct and well structured arguments for running CPDN/BBC Climate/Seasonal Attribution I've read.
Nice job.

Thanks. I won't feel to smug though as my Windows installation crashed over the weekend, and the Windows System Restore has not only buggered up both my current models, but also taken out my backups too. **** **** ****. Being new models I hadn't got round to burning a backup to disk either. Looks like I'm in for starting two new models, unless anyone is an expert in Windows System restore. It's not giving me the option to reverse the restore. Any ideas guys?

Pete

PeteMcCann · 19 Sep 2006 at 01:42

rich99million said:
I would go and have another look at the BBC/CPDN project but if WUs are several months long that's gonna be turn off for me I'm afraid
Are they still split into sections/phases or whathaveyou? Is there any way of quitting part way through a unit without ruining the science completely?

Each model runs a climate simulation over a 160 year period from 1920 to 2080. Each day runs an atmospheric model in half hour slices, hence 48 per day, and also an oceanic model in one hour slices, hence 24 per day. The total model is about 4,500,000 time slices in all. The model runs in six day periods (432 timeslices) when it reaches a check point and saves the date in case the computer is turned off. Every year that the model runs , it contacts the project online and 'trickles up' data for which you then recieve credit.
Every 10 years of the model it then sends up a much larger data Zip file to the project which summarises the decade. 16 Zip files in all for a complete model run. Credit is continuously acquired as a reasonable computer can do one model year per day, or two models on a dual core.

Obviously a completed model is the ultimate objective and would be most usefull for the scientists, but they do say that the first few decades are useful as well, as they are trying out your climate model against known information ie 1920 - 2005, and building up a better overall global climate model. 2005 - 2080 is running the prediction phase, to see where your particular climate model leads.

It might well be a long crunching haul, but given the way things are heading if you saw The Independent last week, about a 20% loss of permanent ice from the arctic in one single year, I kind of think that this could be one of the most important DC projects out there.

Sorry all you folding fanatics!!

Pete

PeteMcCann · 19 Sep 2006 at 01:48

Berserker said:
Not sure how many could cope with the 4-6 month workunits though. Actually, one of my two crunchers has made 26% in 5 months, so god knows what's going to happen there.

All I would stress to anyone thinking of starting the BBC climate model, is to back it up regularly. Man are you ****** when your 4 months of crunching goes down the toilet!!! Just don't go there. Look at the BBC forum to see the pain, wailing and nashing of teeth caused by model loss!!

Pete

PeteMcCann · 19 Sep 2006 at 01:54

Peter-B said:
Hi Pete & welcome to the team.
I don’t wish to curb your enthusiasm but we are currently 11th (world) 8th (UK) and unless some of the slackers switch back to the BBC from there cosy Folding “fire side” we are stuck there, well not quite we will stomp “AMD users” in 2380 days!! In fact thinking about it, the cosy fire side seem a much better place to be!

Hi there Pete. I was looking at the BBC team list which ranks acording to RAC, which I assumed was the norm, but then maybe not. A few more BBC crunchers could easily see us up to 5th on RAC on the BBC list. Is there another ranking list somewhere? Maybe you could point me in the right direction.

Cheers.

Pete

PeteMcCann · 19 Sep 2006 at 22:02

joeyjojo said:
That made me laugh

Having to backup regularly is a turn off for me. I'm not organised enough at all. How do you back up then? CD or DVD or what? How big is the model?

Single model about 500Mb. 2 models about 1GB. but it does vary a bit. I was relying on my raid 1 setup to keep me out of too much trouble of disk failure. Trouble always comes from the least expected direction.

Pete

PeteMcCann · 19 Sep 2006 at 22:12

Peter-B said:
Hi Pete,
Stats here

http://www.boincsynergy.com/stats/teams.php?project=cce
http://boincstats.com/stats/team_stats.php?pr=cce&st=0

It’s a shame about the loss but that’s the BBC!
Oceaness is right system restore is a waste of space, I never use it. I back up all models daily and have had to use them twice this week, including to-day. If it gets any worse I’m quitting having only completed 2 models in 8 months. I us to be a folder and it’s a lot easier than this!!

Hi again Pete. Thanks for the info. I think your No2 spot is safe for a while now as I have ground to a halt trying to find a solution to the current promlems. Yes , the BBC project has not been without its problems, and major cockups for that matter, but when you think the Met office run their global climate simulation models on of the most powerful computers around, it's a pretty mean feat to get a decent model running on a single PC. I just focus on the end result, and think 'No Pain, No Gain'! If it was easy, it would be called 'Folding'!!!!!

PeteMcCann · 19 Sep 2006 at 22:14

oceaness said:
System restore is disabled on all my rigs, it's one of the first things I do after installing Windows.

Now he tells me.!!

Pete