As a sidebar: Have you ever considered teaching your students the differences between all of the Linux versions out there? The Good, the Bad, and Not-So-Good?
Just thought I'd ask.
Not at this point. I am not sure I am versed enough in Linux to do this. I have tried Mint, canonical Ubuntu, and Linux Clear. It's a good idea and something I will think about incorporating. I am just happy to get them exposed to Linux/Ubuntu in the first place. Unless they are really into computers/coding and learning it on their own, they have most likely never used it. It is basically not used in schools.
mikey wrote:
I like the 1000 second tasks you are doing on Meerkat though!!!
Yeah, that's running 2x so I have been fairly impressed with it. Right now, it has been stable at 2x. Time will tell.
Anyone know how the B580 stacks up to the Nvidia A4000? Looks like the half-precision is higher on the Intel card but single is higher on the Nvidia, which should I be looking at for Boinc?
Anyone know how the B580 stacks up to the Nvidia A4000? Looks like the half-precision is higher on the Intel card but single is higher on the Nvidia, which should I be looking at for Boinc?
honestly you should be more concerned with the memory bandwidth for Einstein
Anyone know how the B580 stacks up to the Nvidia A4000? Looks like the half-precision is higher on the Intel card but single is higher on the Nvidia, which should I be looking at for Boinc?
I can't speak specifically for the A4000 but we have the A4500 on a Windows platform (running MeerKAT 2x) and the B580 on a single platform (running MeerKAT 1x). It is somewhat comparing apples to oranges when it comes to price and watts. Personally, I would go with the Arc (in Windows) when it comes to lower power usage and overall cost. The A4000 is nice that it is one slot, but it is not an amazing GPU for its cost. All of this is just my opinion though.
EDIT: Keep in mind, right now you will not be able to run some of the work (O3AS) with the Arc as there is no application (yet...).
mikey wrote:That's GREAT info
)
he can't run O3AS since there's no app for it. and the scheduler will not send work since there's no Intel plan class to ask for it.
_________________________________________________________________________
GWGeorge007 wrote: As a
)
Not at this point. I am not sure I am versed enough in Linux to do this. I have tried Mint, canonical Ubuntu, and Linux Clear. It's a good idea and something I will think about incorporating. I am just happy to get them exposed to Linux/Ubuntu in the first place. Unless they are really into computers/coding and learning it on their own, they have most likely never used it. It is basically not used in schools.
Yeah, that's running 2x so I have been fairly impressed with it. Right now, it has been stable at 2x. Time will tell.
my B580 works a treat without
)
my B580 works a treat without tweaking
https://einsteinathome.org/host/12769171
GPU tasks: https://einsteinathome.org/host/12769171/tasks/4/19
Anyone know how the B580
)
Anyone know how the B580 stacks up to the Nvidia A4000? Looks like the half-precision is higher on the Intel card but single is higher on the Nvidia, which should I be looking at for Boinc?
Ryan wrote: Anyone know how
)
honestly you should be more concerned with the memory bandwidth for Einstein
_________________________________________________________________________
Ryan wrote:Anyone know how
)
I can't speak specifically for the A4000 but we have the A4500 on a Windows platform (running MeerKAT 2x) and the B580 on a single platform (running MeerKAT 1x). It is somewhat comparing apples to oranges when it comes to price and watts. Personally, I would go with the Arc (in Windows) when it comes to lower power usage and overall cost. The A4000 is nice that it is one slot, but it is not an amazing GPU for its cost. All of this is just my opinion though.
EDIT: Keep in mind, right now you will not be able to run some of the work (O3AS) with the Arc as there is no application (yet...).