3.03 vs 3.08. About a 7% speed increase or so with 3.03

[Kelemvor]

Well here's the results of my tests.

I have a P4 1.8 with 640 RAM.

Originally I was running 3.08 with SetiHide. 4 WUs were copmleted with the following times:
4:38:16
4:37:56
4:30:43
4:33:53

Then I put on 3.03 with SetiHide and completed 4 WUs:
4:04:41
4:20:40
4:28:54
4:11:49

AVERAGES
3.08: 4:35:12
3.03: 4:16:31

I think that's about a 7% difference or so.

 

[cakin]

I wonder if these result are the same for linux?

 

[Kelemvor]

I guess I should put in that I'm on XP Home...

Since my post the next couple were 4:21:xx and 4:26:xx which will bring up the average so it seems that 3.03 isn't all that much faster than 3.08. At least for me...

 

[Spacehead]

Well, the angle range(AR) of the work unit(WU) has alot to do with it. Here's the results of my testing so far. I made copies of the 'work_unit.sah' file so i could run the exact same WU with both versions, on both computers.

Windows 98SE
3.413AR
v3.03 - 3:03:55
v3.08 - 3:05:26

0.834 AR
v3.03 - 3:21:33
v3.08 - 3:38:24

0.417AR
v3.03 - 3:35:24
v3.08 - 4:00:33

-----------
Windows XP Home
3.413AR
v3.03
v3.08 - 3:03:53

0.834 AR
v3.03
v3.08 - 3:30:20

0.417AR
v3.03 - 3:30:23
v3.08 - 3:51:43

0.072AR
v3.03 - 3:45:54
v3.08 - 3:37:31

0.380
v3.03 - 3:39:37
v3.08 - 4:06:36

 

[Assimilator1]

Just stumbled upon this thread!😱

Franky jnr
Yeah the Angle Range of the WU makes a big difference to WU times ,3.03 is faster at average AR & VeryHighAR WUs ,whilst 3.08 is faster at VLAR WUs.
Most WUs are average & VHAR WUs so were better off with v3.03🙂