So are all the important WUs like 0-2 in the chain assigned to the fast cards and the last one assigned to the slower cards? That would make sense as for whatever reason you can only run them in a serial chain rather than all parallel.

No, WU's are assigned randomly.

Each WU 0-4 has to be completed because 1-4 is generated from its result etc.

... Even though the server status shows 0~3 unsent workunits there is work until the number of workunits in progress is higher than the number of the active hosts (2000~2500) as new workunits generated from a finished workunit

Zoltan, just for me to know: where can one see the number of active hosts?
I would have thought that this is the figure shown in the upper right hand corner of the project status page (i.e. approx. 5600 right now) - or am I mixing up something?

Zoltan, just for me to know: where can one see the number of active hosts?

Nowhere.

I would have thought that this is the figure shown in the upper right hand corner of the project status page (i.e. approx. 5600 right now) - or am I mixing up something?

We can't have 5600 active hosts while there's only 3300 workunits are in progress, right? As a host can have only 2 workunits per GPU, and there are negligible number of hosts with multiple GPUs, the real number of the active hosts is about the half of the number of workunits in progress. There are 5600 hosts with recent credit. Now what does "recent" mean is unclear, perhaps it is the past 30 days, but I wouldn't call a host active if it's crunching only a few workunits in a month.

We can't have 5600 active hosts while there's only 3300 workunits are in progress, right?

that's totally right, I didn't see it from this angle.

Thank you for explaining to me.

I noticed lately that the tasks are quite demanding lately on my GPU and I'm starting to get very close to missing the 24 hour mark. Is it time to upgrade my GPU? Which should I buy if I upgrade?

---

Cruncher/Learner in progress.

Get a 1070 if you can afford it, 1060 if not.

Which should I buy if I upgrade?

Available funds?

---

I would love to see HCF1 protein folding and interaction simulations to help my little boy... someday.

I'd like to know from the scientists too if I should get an upgrade, though. I don't want to buy a new GPU and come back later to see that my gtx 960 works just fine again under 24 hours (meaning these long demanding tasks are gone)

I probably won't be getting a GPU upgrade until I get myself my first part-time job after High School while I'm in college, so I can have a nice steady income. I like to look forward in the future purchases though to see what I want and set a goal.

Basically, I'm willing to upgrade if I should, but do not have a budget in mind at this time.

---

Cruncher/Learner in progress.

I'd like to know from the scientists too if I should get an upgrade, though. I don't want to buy a new GPU and come back later to see that my gtx 960 works just fine again under 24 hours (meaning these long demanding tasks are gone)

I think the long demanding tasks are almost gone now. So if you are just concerned about the 24 hour limit, I would keep the GTX 960 for now.

Are these new Pablo WUs depleted already or has he not yet added the rest?

I think the long demanding tasks are almost gone now. So if you are just concerned about the 24 hour limit, I would keep the GTX 960 for now.

I second that. The gtx 960 is a good mid-range card and energy efficient (Maxwell). It is somewhat comparable to the new 1050ti in terms of crunching speed. So if at all, you need to upgrade to the gtx 1060 to have a noticeable speed improvement. Price is from 200 USD (for the 3G model) and up. But for the moment, if you miss the 24h deadline just narrowly, you may want to overclock your 960 a little.

---

I would love to see HCF1 protein folding and interaction simulations to help my little boy... someday.

Okay, I have overclocked my card and have tested it several times. I'm now awaiting a task to see how things go.

---

Cruncher/Learner in progress.

Here was my last task that missed it by around 2 hours :(

http://www.gpugrid.net/result.php?resultid=15946821

---

Cruncher/Learner in progress.

And the queue is empty again. Come on, how long is this going to take?

---

I would love to see HCF1 protein folding and interaction simulations to help my little boy... someday.

You seem to have gotten 2 tasks right now. To avoid the waiting in these cases, I'd change your settings to "store at least 0.1 days of work" instead of having it set to 0. I just changed mine myself.

If that doesn't work, try 0.2 and keep going 0.1 higher to see which is good (until the servers offer lots of projects of course)

---

Cruncher/Learner in progress.

I do 0 days of work for both regular storage and additional

I was wondering why the living room was a tad colder than usual this morning... ^_^

I was wondering why the living room was a tad colder than usual this morning... ^_^

You definitely need backup projects with this project :) You just need to set the priority for GPUGrid to something ridiculous, I set mine to 10000 and everything else to 1

Or ... leave GPUGrid at 100, and set backup projects to the "backup resource share" value, of 0, which literally means "Only use this project when work cannot be retrieved from other projects and I need work now ... and only get enough tasks to satisfy my idle resources."

0 Resource Share, is very useful.

Putting up some new adaptive runs today