climateprediction.net home page
Posts by Jim1348

Posts by Jim1348

41) Message boards : Number crunching : Dr Lisa Su says up to 192MB L3 on newer Ryzen -- hope it's true (Message 64999)
Posted 24 Jan 2022 by Jim1348
Post:
Now, for the real kicker. The 5800X3D has a 96MB L3 cache (32+64) as compared to the 32MB cache found on the standard 5800X. That’s a whole 64MB more, putting it above even the Ryzen 9 5950X in terms of cache.
https://appuals.com/ryzen-7-5800x3d-is-the-first-ryzen-chip-to-use-the-3d-v-cache-tech-and-its-faster-than-the-core-i9-12900k/

Looks good to me. I would go for it if OpenIFS ever comes along, and needs a lot of cache.
But the glaciers may have receded by then anyway.
42) Message boards : climateprediction.net Science : Scientists see stronger evidence of slowing Atlantic Ocean circulation, an ‘Achilles’ heel’ of the climate (Message 64982)
Posted 19 Jan 2022 by Jim1348
Post:
The general conception of a tipping point is that there’s some fixed physical threshold beyond which the system trips into a different state. But they found that a lesser-known phenomenon known as a rate-­induced tipping point, triggered by a sudden increase in the system’s rate of change, might halt the currents as well. In other words, too much change occurring too fast could cause the system to break down. The Atlantic circulation could be susceptible to this if the water flowing from ice sheets increases rapidly enough, according to the study, which was published in the Proceedings of the National Academy of Sciences in March.

It’s just one model and one study, but it suggests that the climate system could be more fragile than previously appreciated. These “chaotic dynamics” mean that “we maybe cannot expect, even if our models get much better, to be able to predict with 100% confidence whether such an element of the climate system will go into another state or not,” Lohmann says.

https://www.technologyreview.com/2021/12/14/1041321/climate-change-ocean-atlantic-circulation/
43) Message boards : Number crunching : Lost tasks (Message 64924)
Posted 7 Jan 2022 by Jim1348
Post:
Sadly, it has been suggested several times to the project that they reduce the length of deadlines. I am not sure what the reluctance is about.

I attributed it once to the Medieval timescale. I don't think it was accepted formally.
44) Message boards : Number crunching : New work Discussion (Message 64832)
Posted 20 Nov 2021 by Jim1348
Post:
... with adding a second getting close to doubling throughput but the third and fourth tasks only gave marginal gains.

Thanks. I think that is a good first indication. It is not surprising that they use a lot of cache.
45) Message boards : Number crunching : New work Discussion (Message 64829)
Posted 20 Nov 2021 by Jim1348
Post:
Some have been as low as 4GB/task in the past so this small testing batch of three tasks is no guarantee that they will be as heavy on RAM when they finally make it to the main site or it may be like the testing ones, some are as bad and others are lower. But I am ordering some more RAM as with 8 real cores, it is pretty clear 32GB does not cut the mustard any more.

Can you determine anything about cache requirements yet? That often determines how many work units we can run (efficiently), rather than the RAM requirements.
I am all in favor of using lots of RAM, but there is no point in buying it if it can't be used.
46) Message boards : Number crunching : New work Discussion (Message 64824)
Posted 19 Nov 2021 by Jim1348
Post:
Peak usage for latest OpenIFS tasks in testing is about 12GB!

I have already retired one machine (64 GB) waiting for this project. I am now up to 96 GB, and can do 128 GB if needed.
They are just trying to support the memory companies.
47) Message boards : Number crunching : New work Discussion (Message 64758)
Posted 1 Nov 2021 by Jim1348
Post:
It's amazing how much math my generation can do in our heads compared to kids today that need a calculator to do the most rudimentry arithmetic.

At my engineering school, order-of-magnitude calculations were emphasized, to catch the mistakes that people did with more precise methods.
Also, it gave you a greater physical feel for the subject matter. I think many political mistakes are made by people who have not the slightest idea of the magnitude of what they are talking about.
48) Message boards : Number crunching : Site problems (Message 64697)
Posted 25 Oct 2021 by Jim1348
Post:
We don't see much evidence of it in the eastern U.S., just some normal rain, though it is making the news.
https://news.yahoo.com/record-breaking-california-bomb-cyclone-linked-to-climate-change-183607985.html

(Probably every topic on the CPDN forums will have to be sub-titled "Climate Change in the News" from now on.)
49) Message boards : Number crunching : Computation Errors (Message 64355)
Posted 13 Aug 2021 by Jim1348
Post:
Running only two HadAM4 at N144 on a Ryzen 3900X, with the other 22 cores on SiDock, gives less than 9 sec/TS.
https://www.cpdn.org/results.php?hostid=1520871&offset=0&show_names=0&state=0&appid=31

That is the best I have seen. It is usually around 12 sec/TS when running four N144 and maybe a couple of N216.
50) Message boards : Number crunching : Computation Errors (Message 64346)
Posted 12 Aug 2021 by Jim1348
Post:
Any thoughts about my questions on CPU cache limits and using sec/TS?

On my Ryzen 3600 and 3900X, I limit the N216 to no more than four at a time, though two is a bit better.
I don't know how that scales for your machine, but be advised.

You should be getting around 20 sec/TS.
https://www.cpdn.org/results.php?hostid=1520871&offset=0&show_names=0&state=4&appid=33

The N144 take less cache, you could probably run twice as many, and are almost twice as fast.
https://www.cpdn.org/results.php?hostid=1520871&offset=0&show_names=0&state=4&appid=31

I think the Ryzens, with their larger caches, will probably do better than the Intel CPUs.
51) Message boards : climateprediction.net Science : Climate change in the News (Message 64320)
Posted 10 Aug 2021 by Jim1348
Post:
An additional problem is that even the changes may not be stable. Civilizations can adapt to long-term changes by migration; that is why humans left Africa, etc. But with wandering jet streams and ocean currents, even the migrations might not work as a long-term solution. This is an unprecedented situation. Even the pandemic looks easy by comparison. We know how to deal with that.
52) Message boards : climateprediction.net Science : Nuclear fusion start-up backed by Jeff Bezos to build first reactor in UK (Message 64290)
Posted 8 Aug 2021 by Jim1348
Post:
China broke the record by keeping the Experimental Advanced Superconducting Tokamak (EAST) by achieving plasma temperature at 120 million Celsius for 101 seconds and 160 million Celsius for 20 seconds, a major step toward the test run of the fusion reactor.
https://www.globaltimes.cn/page/202105/1224755.shtml
53) Message boards : climateprediction.net Science : Climate change in the News (Message 64279)
Posted 7 Aug 2021 by Jim1348
Post:
The real problems will come when the expected revenue from an acre of ground is less than the expected cost of planting (seed, fertilizer, equipment, etc.).
Then they won't plant at all.
54) Message boards : climateprediction.net Science : Climate change in the News (Message 64276)
Posted 7 Aug 2021 by Jim1348
Post:
We are having a fairly mild summer in the eastern U.S. I don't think that will bring much comfort to the West.
There is no paradox at all. The heat has to go somewhere. It is the average that is rising, with variations around the mean.

Surely you studied that in your statistics course.
55) Message boards : climateprediction.net Science : Climate change in the News (Message 64263)
Posted 6 Aug 2021 by Jim1348
Post:
And now a paper is suggesting the AMOC might stop working.

I am hoping OpenIFS can include that, though the end of civilization might be difficult to account for.
56) Message boards : Number crunching : Project Outage (Message 64262)
Posted 6 Aug 2021 by Jim1348
Post:
I had two that had competed a day or two ago, but had not been reported yet.
A manual update fixed it, and I am now all up to date.
https://www.cpdn.org/results.php?hostid=1520871
57) Message boards : climateprediction.net Science : Nuclear fusion start-up backed by Jeff Bezos to build first reactor in UK (Message 64228)
Posted 28 Jul 2021 by Jim1348
Post:
I don't think there has been a lot new in the science of fusion. It has been an engineering problem for a long time now.

Then ITER is spending a huge amount of money for no conceivable purpose. They will barely get to breakeven, and will leave the engineering challenges to the next version.
But it is not my money.
58) Message boards : climateprediction.net Science : Nuclear fusion start-up backed by Jeff Bezos to build first reactor in UK (Message 64225)
Posted 28 Jul 2021 by Jim1348
Post:
Great stuff! I think the "completion date" can be taken with a grain of salt; maybe they will get a plasma by then, but not much more.
However, the fact that private investors in the U.S. are willing to spend money on it shows that they think the basic science is far enough along, and it is just an engineering project now.
That is still VERY ambitious, but it gives signs of hope.
59) Message boards : climateprediction.net Science : Nuclear fusion start-up backed by Jeff Bezos to build first reactor in UK (Message 64212)
Posted 25 Jul 2021 by Jim1348
Post:
Here is another discussion of the General Fusion project, along with a number of other startups.
https://www.economist.com/science-and-technology/2021/06/24/the-race-to-build-a-commercial-fusion-reactor-hots-up

You will need an account to read it, but it is free.
60) Message boards : climateprediction.net Science : Nuclear fusion start-up backed by Jeff Bezos to build first reactor in UK (Message 64211)
Posted 25 Jul 2021 by Jim1348
Post:
Thanks. Putting smart minds together will help.
But they will have a practical solution long before the large-scale projects even get results.
Then they can follow up with scientific studies if they want to.


Previous 20 · Next 20

©2024 climateprediction.net