The Scottish Boinc Team

Are these tasks optimized the same same at the solo? Large estimating over 7 hours to crunch?

Yes they are. The tasks have not changed any, just the size and the name. You should use exactly the same optimisation as you did for the solo collatz, but you will have to cut and paste it into each of the 3 new config files.

You can also choose which of the 3 new sizes to do on your collatz application page within your collatz account.

Need to do this when I get home as they started just as I left and judging by the points I am generating, doubt my GPU's are running at much more than 50 - 60 percent utilisation.

I think I've got most of my machines, except maybe one, fully optimised.

I've not ran enough of the new types of units on each machine but it does look as though the large units are more profitable: On a 7970 I get about over 10 credits per second crunching the large_collatz units as opposed to just 7-9 credits with solo. No data yet on mini or micro units.

I'll report back when I have clearer data.

Megacruncher wrote:I think I've got most of my machines, except maybe one, fully optimised.

I've not ran enough of the new types of units on each machine but it does look as though the large units are more profitable: On a 7970 I get about over 10 credits per second crunching the large_collatz units as opposed to just 7-9 credits with solo. No data yet on mini or micro units.

I'll report back when I have clearer data.

That's my observations as well. That's better than a 10% efficiency crunching the large over the solo. If you're currently tallying 1,000,000 solo collatz credits per day, that can mean another 100,000 if you stick with the large WU.

Is any one else having an issue with the large tasks locking up the gpu ?
Having an issue similar to what we used to have on Dirt where the gpu would stop working, 0 % load but the unit would just keep on counting up time.
Only seems to happen on the large units.

Not sure I've seen that exact behavior. Both GPUs on one system did crash recently though. I don't recall if it was running the large tasks.

Anyone know what's been up with Collatz for the last week?

Jon is away on holiday and true to form as soon as he left, server down. Boy do I know that feeling. I believe he is away for at least another week. He did post about the power company replacing the meter and if the outage lasted longer than the ups then all would be down.

Oh well. We can look forward to a nice windfall when he gets back from his hols.

I think quite a few of the projects are one man bands.

I might augment the Einstein with a bit of Primegrid.

I don't think it was the leccy company and the meter because it went down at 4AM that day. I was online. I got into Collatz and about 10 minutes later it went kaput. I thought in his message before he left he said it if it did go down then he would be back by the 10th to fix it. Maybe it was the 10th he was leaving. Confusing message.

iirc he posted that if anything more than a simple reboot of the system was required that it would be down until the 15th as the sitter didn't know how to do anymore than reboot it.

So if i am remembering right it should be back on within the next 24.

and you are in the team top 100 for Einstein so don't jump too quick

robmacagain wrote:iirc he posted that if anything more than a simple reboot of the system was required that it would be down until the 15th as the sitter didn't know how to do anymore than reboot it.

So if i am remembering right it should be back on within the next 24.

Woo hoo my memory works .

Woo hoo my memory works .

Glad yours works. Mine falls into the same category as the rest of my busted body.