GPU to 72 status... - Page 481

chalsall · 2022-04-28, 16:36

Quote:

Originally Posted by LOBES

If I were to start up a new Colab session, it would assign me new work (as opposed to picking up where I left off).

You misunderstand how the Colab assignments work. Assignments don't expire (unless they're obviously abandoned) and will be reassigned to your next instance so long as the last update is more than 30 minutes ago. The checkpoint files are sent to the server every five minutes, so on average 2.5 minutes of work is "lost" per run.

Quote:

Originally Posted by LOBES

Thanks as always for your help! I think GPU72 is a great service.

Thanks for the kind words. GPU72 has been a wonderful test-bed for me over the years. Colab just made it even more interesting!

Uncwilly · 2022-04-28, 16:50

I think I spotted an anomaly:
Looking at this graph: https://www.gpu72.com/reports/workers/lltf/graph/69-78/
The line for the user in place 70 appears to have a flat production in the last week or two. I believe that that is not the case. Credits are not the goal, but tracking data should be accurate.

chalsall · 2022-04-28, 17:01

Quote:

Originally Posted by Uncwilly

I believe that that is not the case. Credits are not the goal, but tracking data should be accurate.

I /think/ what you are seeing on that graph is simply aliasing -- you should see the line jump another pixel shortly.

The tracking data is accurate. Please see this report, for example.

chalsall · 2022-04-30, 23:40

Quote:

Originally Posted by chalsall

I /think/ what you are seeing on that graph is simply aliasing -- you should see the line jump another pixel shortly.

Yup. As modeled, the pixel bump happened for that graph over the last couple of days.

And, I thank you sincerely for pointing out that incredibly expensive query involved with that graph. I ***so*** need to refactor the code used for that. There are many code paths I have long ago forgotten about, but are run from time to time.

So much to do. So little time.

Uncwilly · 2022-05-06, 01:50

Can you poke the beehive again? There are some exponents (LMH-TF) at 100% that are stuck.

bayanne · 2022-05-19, 08:06

Quote:

Originally Posted by bayanne

Could you give the system a nudge to accept automatic results please :)

Could you give the system a nudge again

Thanks

LaurV · 2022-05-24, 14:50

Hey Chris, reporting results is stuck, I think it needs some pedals? Paddles? Oars? Slaves to pull them?

mrk74 · 2022-05-26, 23:54

Quote:

Originally Posted by LaurV

Slaves to pull them?

Really? Too far. WAY too far.

chalsall · 2022-05-27, 01:32

Quote:

Originally Posted by mrk74

Really? Too far. WAY too far.

Please don't!

We are in a hyper-sensitive environment. I *know* LaurV didn't mean anything negative by his post.

To share... When I first came down to Barbados I was doing an RF survey with some very cool dudes... I said "So, this area is black... I mean dark... I mean, there's no signal...

The darkest man I had ever met came next to me and said "It's OK. We know what you mean. It's all OK.

PhilF · 2022-05-27, 01:38

Quote:

Originally Posted by mrk74

Really? Too far. WAY too far.

I thought he meant hamsters running in their wheels...

I don't think it is too far. It's just what you read into it.

Dr Sardonicus · 2022-05-27, 01:43

Quote:

Originally Posted by mrk74

Really? Too far. WAY too far.

In the words of Joe Biden, "C'mon, man!"

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Status	Primeinator	Operation Billion Digits	5	2011-12-06 02:35
62 bit status	1997rj7	Lone Mersenne Hunters	27	2008-09-29 13:52
OBD Status	Uncwilly	Operation Billion Digits	22	2005-10-25 14:05
1-2M LLR status	paulunderwood	3*2^n-1 Search	2	2005-03-13 17:03
Status of 26.0M - 26.5M	1997rj7	Lone Mersenne Hunters	25	2004-06-18 16:46

2022-04-28, 16:50	#5282
Uncwilly 6809 > 6502 """"""""""""""""""" Aug 2003 101×103 Posts 25217₈ Posts	I think I spotted an anomaly: Looking at this graph: https://www.gpu72.com/reports/workers/lltf/graph/69-78/ The line for the user in place 70 appears to have a flat production in the last week or two. I believe that that is not the case. Credits are not the goal, but tracking data should be accurate.

2022-05-06, 01:50	#5285
Uncwilly 6809 > 6502 """"""""""""""""""" Aug 2003 101×103 Posts 10895₁₀ Posts	Can you poke the beehive again? There are some exponents (LMH-TF) at 100% that are stuck.

2022-05-24, 14:50	#5287
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 24051₈ Posts	Hey Chris, reporting results is stuck, I think it needs some pedals? Paddles? Oars? Slaves to pull them?