big beta 8103

My 62xx rig pick up one: 8:24 TPF. Now that everyone is nice and excited, it is time for Stanford to pull the plug and make them non-existent like the 8102.
 
Nicolas_orleans said:
Thanks guys for the data, it's confirmed Runs 0 and 1 are available for 8103. Any of you could confirm whether 8102 was Run 0 only ?
Click to expand...

Looking through my logs, all the 8102s in the last couple of months have been Run 0. Unfortunately I recently had to move my HFM box, so don't have all my history online. Perhaps some others can chime in.
 
402blownstroker said:
My 62xx rig pick up one: 8:24 TPF. Now that everyone is nice and excited, it is time for Stanford to pull the plug and make them non-existent like the 8102.
Click to expand...

The last three or four runs on my 2P rig have actually been 8102's. I had not seen them in about 2-3 months it seems, and just out of the blue a get a string of them. Hopefully, that's a sign of greater things to come! :cool:
 
Nathan_P said:
Only ever had run 0 for 8102, have managed to get run 0 clone 40 though
Click to expand...

This supports the theory of having 2 runs of 8103 against 1 run of 8102. Though, I do not understand well enough the Run/Clone/Gen process to be sure this means more WUs to fold :confused:

Speaking of what... back to 8101 since yesterday :eek:
 
Two almost identical rigs (both have set of 4x6172 run at same refclock - 241, 16x2GB DDR3-1600, same board) but

1- sm6172_1 runs on Ubuntu 12
2- sm6172_2 runs on Ubuntu 10

8103u.jpg


Any idea? suggestion?
 
Check the load balancing on the Ubuntu 10 machine. I had an 8103 processing slowly on an Ubuntu 10 machine today, much slower than previous 8103 units. I checked the ~.log file, and sure enough, the load imbalance was at 14%. A reboot and Folding restart fixed the problem.
 
I'm running bigadv, not bigbeta, and I snagged a 8103 on a dual E7-2860 rig today.

HFM benchmarks says:
8101 166K PPD
8102 246K PPD
8103 239K PPD (estimate on this WU only)

Pretty sweet.
 
Well they just launched 8104 to beta, 1.8 days to deadline, that just ruled out most 2p rigs
 
Nathan_P said:
Well they just launched 8104 to beta, 1.8 days to deadline, that just ruled out most 2p rigs
Click to expand...

Depends on the number of atoms in the project. If it is like the 6901 still floating around, then 2P should do fine with it.
 
It has less base points than an 8101/3.. 17153 vs 22607

Seems it's like the 8103 which I do in ~21:40 for bout 36hr run so I'd make the 8104 by a couple hrs.

Why are they jacking us on short deadlines!
 
Last edited:
402blownstroker said:
Depends on the number of atoms in the project. If it is like the 6901 still floating around, then 2P should do fine with it.
Click to expand...

Atoms are not a good predictor of tpf. 8102/3 have the same number of atoms as 8101.

Given the way the points are now, if tpf on 8104 is 75% of tpf on 8103, it will give ~same PPD. Slower than that, less, and if it's the same tpf, then PPD will equal 8101 levels.
 
You must log in or register to reply here.
Back
Top