I'm looking for, but can't seem to find, a technical reason why my dual Xeon 24 thread Windows server can only receive BigAdv WU 6900. Which is also odd because the 6900 seem to not exist any more but I'm still getting them.
Yet, if I hack my I7-2600K that runs Linux to make it seem like I have 16 threads I only get the new 8101 WU that would actually fold properly on my dual Xeon Windows server (I'm assuming this part as I having actual been able to fold this WU's).
What gives? I have a machine that can fold BigAdv but because it's running Windows I can't? Doesn't make sense to me.
It has always been that way. The projects that run in Linux are not necessarily avail on the Mac or Windows and vice-versa. The projects require specific cores and those are always coded for the OS they are run on. Cores between the different OS's aren't necessarily identical in capability (because the OS's have different API's and capabilities) or version.
Hardware configuration: Intel i7-4770K @ 4.5 GHz, 16 GB DDR3-2133 Corsair Vengence (black/red), EVGA GTX 760 @ 1200 MHz, on an Asus Maximus VI Hero MB (black/red), in a blacked out Antec P280 Tower, with a Xigmatek Night Hawk (black) HSF, Seasonic 760w Platinum (black case, sleeves, wires), 4 SilenX 120mm Case fans with silicon fan gaskets and silicon mounts (all black), a 512GB Samsung SSD (black), and a 2TB Black Western Digital HD (silver/black).
No 64 bit win fahcore for bigadv, so no WUs assigned that use more than 4 GB. Sorry, don't know why either. Always seems to be a programmimg issue with making fahcores for win and 64 bit.
7im wrote:No 64 bit win fahcore for bigadv, so no WUs assigned that use more than 4 GB. Sorry, don't know why either. Always seems to be a programmimg issue with making fahcores for win and 64 bit.
I'm still confused buy all this. I can run A5 WU on Windows. These are the supposedly finished 6900's that I have been folding none stop for about 4 months now. My Linux (Ubuntu 11.10) setup is configured to use only 4GB memory in VirtualBox. I can fold 6903 with no problem on this and I get the new 8101 WU's that I can't fold due the fact that I'm running it on an i7-2600K. But all with just 4GB of memory.
So, it can't be anything to do with A5 bigadv needing more than 4GB of memory. I'm really thinking that there may be a bias toward Linux happening here. Which in the long run just cuts out the literally millions of Windows Server boxes that have the power to fold bigadv. Sad and disappointing.
patonb wrote:Use a vbox linux install.... I use it on my 2p xeon setup, works great.
I already had VirtualBox setup on this server to I went ahead and setup an Ubuntu 11.10 guest on it. Guess what? I got a 6900 WU. I really don't understand the logic of how WU's are assigned. The Linux session is setup for SMP 24 (v7 client), client-type=bigadv and max-packet-size=big and I get a WU that isn't supposed to exist anymore. But on my VirtualBox Linux session running on my i7-2600k (SMP 16) I get the 8101 that can't be properly folded on an i7.
Hardware configuration: Intel i7-4770K @ 4.5 GHz, 16 GB DDR3-2133 Corsair Vengence (black/red), EVGA GTX 760 @ 1200 MHz, on an Asus Maximus VI Hero MB (black/red), in a blacked out Antec P280 Tower, with a Xigmatek Night Hawk (black) HSF, Seasonic 760w Platinum (black case, sleeves, wires), 4 SilenX 120mm Case fans with silicon fan gaskets and silicon mounts (all black), a 512GB Samsung SSD (black), and a 2TB Black Western Digital HD (silver/black).
Sorry, upon review, win cores were puking at 2 GB.
And yes, there is a bias towards llinux, but not from Pande Group! The whole world tries to improve open source software performance, whereas only Microsoft works on windows. Linux software has always been ahead of windows when it comes to multithreading.
