Well i wonder if with my little means i shud swap to SCC instead of trying just to get a badge on MIP (in fact just to show i am participing)
is it useful or not...

Hi TonyEllis,
are your machines running Windows or Linux?
On my side, in case of MIP1 only computation, the both platforms are impacted, nonetheless Linux seems to be more negatively impacted than Windows.
If you monitor the used RAM, you will notice that MIP1 is really demanding, close to 8 GB RAM, incl. Windows 7 Pro x64, for 8 threads.
On Linux hosts, I did noticed that the CPU temperature drops down even if the CPU load is still reported to be 100%.
MIP1 has definitely a memory management problem.
Maybe it could be corrected with other compiling options of the science, maybe a better (more efficient) software design should be considered, avoiding cache faults as far as possible.
Cheers,
Yves

But FX8350 is one generation old from AMD now (Bulldozer architecture was a notorious low-IPC:er). Ryzen is much, much better (even if it is not up to par on IPC when compared to latest Intel chips). But, since WCG crunching is an application that scales Very Well with increased core/thread count capabilities of a CPU, you will probably get more out of an AMD today in terms of WCG performance/$.

I'm very much looking forward to ryzen notebook CPU's. Does anyone have info on them/release date?
I think exciting times are ahead.

In an attempt to see difference between SCC1 and MIP1, since have two virtually identical machines, ran SCC1 on one and MIP1 on the other. They each have 2G memory, same motherboard, same processor with 4 CPUs (2 Intel cores hyper-threaded). An application known to run continuously was added to see what normal variations could be expected...
The boinc files for each project and the space taken for each project under "projects'
SCC1 17 files 7.6 MB
MIP1 140 files 334 MB
First ,lets see the difference in page faults and memory use over a 30 minute (1800 secs) period.

Command in this format "pidstat -p pid1,pid2,pid3,pid4 -r -t 1800 1
-r Report page faults and memory utilization. (truncated from 'man' pages)
... snipped
minft/s Total number of minor faults the task has made per second,
those which have not required loading a memory page from disk.
majflt/s Total number of major faults the task has made per second,
those which have required loading a memory page from disk.
VSZ Virtual Size: The virtual memory usage of entire task in kilobytes.
RSS Resident Set Size: The non-swapped physical memory used by the task in kilobytes.
... snipped

Average:         TGID       TID  minflt/s  majflt/s     VSZ    RSS   %MEM  Command
Average:         4427         -      1.27      0.00  284780  92280   4.53  wcgrid_mip1_ros
Average:            -	   4427      1.20      0.00  284780  92280   4.53  |__wcgrid_mip1_ros
Average:            -	   4428      0.07      0.00  284780  92280   4.53  |__wcgrid_mip1_ros
Average:         4376         -      0.58      0.00  286016  93604   4.59  wcgrid_mip1_ros
Average:            -	   4376      0.51      0.00  286016  93604   4.59  |__wcgrid_mip1_ros
Average:            -	   4385      0.07      0.00  286016  93604   4.59  |__wcgrid_mip1_ros
Average:         4365         -     20.90      0.00  405988  17660   0.87  boinc_client
Average:            -	   4365     20.86      0.00  405988  17660   0.87  |__boinc_client
Average:          603         -     10.35      0.00   99792   1756   0.09  gkrellmd
Average:            -       603      7.75      0.00   99792   1756   0.09  |__gkrellmd

Average:         TGID       TID  minflt/s  majflt/s     VSZ    RSS   %MEM  Command
Average:        17305         -      0.18      0.00  127864  51200   2.50  wcgrid_scc1_vin
Average:            -     17305      0.00      0.00  127864  51200   2.50  |__wcgrid_scc1_vin
Average:            -     17310      0.06      0.00  127864  51200   2.50  |__wcgrid_scc1_vin
Average:            -     17350      0.12      0.00  127864  51200   2.50  |__wcgrid_scc1_vin
Average:        15302         -      0.12      0.00  125376  48708   2.38  wcgrid_scc1_vin
Average:            -     15302      0.00      0.00  125376  48708   2.38  |__wcgrid_scc1_vin
Average:            -     15303      0.06      0.00  125376  48708   2.38  |__wcgrid_scc1_vin
Average:            -     15322      0.06      0.00  125376  48708   2.38  |__wcgrid_scc1_vin
Average:         6788         -     24.01      0.00  185164  15740   0.77  boinc_client
Average:            -	   6788     24.01      0.00  185164  15740   0.77  |__boinc_client
Average:         3687         -      9.34      0.00   98984   1860   0.09  gkrellmd

Not a massive memory diference between SCC1 and MIP1 - roughly 2x - no substantial difference in page faults
OK - lets; look at disk IO
Commnd in this format "pidstat -p pid1,pid2,pid3,pid4 -d 1800 1
-d Report I/O statistics (kernels 2.6.20 and later only). The following values are displayed:
..snipped
kB_rd/s Number of kilobytes the task has caused to be read from disk per second.
kB_wr/s Number of kilobytes the task has caused, or shall cause to be written to disk per second.
kB_ccwr/s Number of kilobytes whose writing to disk has been cancelled by the task. This may occur when the task truncates some dirty pagecache. In this case, some IO which another task has been accounted for will not be happening.
...snipped

Average:         TGID       TID   kB_rd/s   kB_wr/s kB_ccwr/s  Command
Average:         4427         -    115.21      0.47	 0.00  wcgrid_mip1_ros
Average:            -	   4427    105.04      0.12	 0.00  |__wcgrid_mip1_ros
Average:            -	   4428     10.18      0.36	 0.00  |__wcgrid_mip1_ros
Average:         4376         -    290.83      0.34	 0.00  wcgrid_mip1_ros
Average:            -	   4376    271.76      0.00	 0.00  |__wcgrid_mip1_ros
Average:            -	   4385     19.07      0.34	 0.00  |__wcgrid_mip1_ros
Average:         4365         -     28.41      3.83	 0.00  boinc_client
Average:            -	   4365     27.64      3.65	 0.00  |__boinc_client
Average:          603         -      3.77      0.00	 0.00  gkrellmd
Average:            -       603      2.30      0.00	 0.00  |__gkrellmd

Average:         TGID       TID   kB_rd/s   kB_wr/s kB_ccwr/s  Command
Average:        17305         -      0.00      0.55	 0.06  wcgrid_scc1_vin
Average:            -     17305      0.00      0.00	 0.00  |__wcgrid_scc1_vin
Average:            -     17310      0.00      0.17	 0.00  |__wcgrid_scc1_vin
Average:            -     17350      0.00      0.38	 0.06  |__wcgrid_scc1_vin
Average:        15302         -      0.00      0.49	 0.06  wcgrid_scc1_vin
Average:            -     15302      0.00      0.00	 0.00  |__wcgrid_scc1_vin
Average:            -     15303      0.00      0.16	 0.00  |__wcgrid_scc1_vin
Average:            -     15322      0.00      0.33	 0.06  |__wcgrid_scc1_vin
Average:         6788         -      0.00      3.13	 0.72  boinc_client
Average:            -	   6788      0.00      3.13	 0.72  |__boinc_client
Average:         3687         -      0.00      0.00	 0.00  gkrellmd
Average:            -	   3687      0.00      0.00	 0.00  |__gkrellmd

Well that is a difference! - looks like a really fast disk system helps MIP1.
If anyone has other ideas - will see what can be measured... sorry for the length...
EDIT: Sorry about the loss of formatting - it was correct in the editor copy-and-pasted from..
Update: trying the code tag suggestion below...