SMP2 appears to have switched to -smalladv

Soldato
Joined
25 Dec 2008
Posts
5,976
Location
Sheffield/Norwich
I was just getting confused at why my frame times were hovering around the 9:36 times yet everything seemed normal. I even restarted the client. Then I noticed the project size was 10,353 :o (including bonuses). Anyone else had one of these units? (P6040)

No -smalladv I'm afraid, it just seemed an appropriate name :p
 
Nope this is on Windows, I switched to that soon after SMP2 came out. A quick Google's come up with this information:


Peter Kasson has added project 6040 to the list of SMP2 projects.

This represents an increase in the system load for SMP2 projects, evidenced in the number of atoms contained within this new project: a grand total of 330,706. Most of the projects thus far have contained around 75,000 atoms, projects 6011 and 6013 being exceptions at 146,000 atoms each. This is not a sufficiently large number of atoms for the project to be classed "bigadv" (the current A2 bigadv units contain over a million atoms) but it is a step closer to bigadv units for the A3 core.

This project examines in detail the substance known as hemagglutinin. This protein is responsible for setting the flu virus upon potential host cells.

In order to fold these new units, version 2.17 of the A3 core is required. The core should be updated automatically upon receiving one of these units, if it hasn't already through another project.

Project 6040 is distributed by a Stanford server located at 171.64.65.54. Each unit is worth 1395 points, and has a bonus factor of 4.63. The units have a preferred deadline of 6 days, with the final deadline 2 days later.

Along with the increase in the number of atoms, you will notice an increase in the amount of memory consumed by the core. The memory usage may rise to 375MB with the -smp 8 configuration, and 300 MB with the -smp 4 one. It will also require an additional 60MB of disk space when outputting the unit results. Judging by initial observations, these units can be folded to completion in less than a day on a Core i7 processor, and within one to one and a half days on a Core 2 Quad or Phenom II X4.


Oddly enough despite the (improved?) bonus factor, I'm not seeing an improved PPD compared to other A3 units.

Looks like I'm curing flu :D
 
Had a couple. These are quite slow on older chips - was getting 4500ppd on my Q6600 compared to 6000 with the smaller WUs. Supposed to be better on i7, though.
 
Back
Top Bottom