I have a 550W antec PSU, and while I have a 5 drive raid, I don't have SLI gfx cards so I thought I was oversized. I'll be ordering a 750W PSU in the near future that should hopefully provide a longer period of overkill than this one.
No, this looks like the science app got blocked from writing when trying to checkpoint and had to start over from scratch on the next task switch (check the RID data). The reason the time seems to "double" is the Total CPU run time for the result so far is stored in the client_state file, not the checkpoint file.
Where you see that in stdout err? The talk about 'no usable checkpoint' is clearly form the first start of the workunit.
OOPs, my bad! I must have been looking at the wrong host. :-O
Now that I'm looking at yours (735729, just to make sure), that is kind of strange. It looks you had two in row from the same data pack, and what's really strange is your partner for 57075104 ran it in what looks to be the nominal time for it. Can't tell about the other one since the host doesn't have enough result showing to make any comparisons.
I have a 550W antec PSU, and while I have a 5 drive raid, I don't have SLI gfx cards so I thought I was oversized. I'll be ordering a 750W PSU in the near future that should hopefully provide a longer period of overkill than this one.
I'm beginning to think we need to read the spec's of power supplies more closely. About a year ago I had to replace a 650W ThermalTake psu when the computer was upgraded. It was replaced with a 480W Antec, that had a greater output on the 5V and 3.3V rails. The 650W unit, was tested and is up to spec, it now powers a new computer where the mobo and graphics power requirements are on the 12V rails.
I'm beginning to think we need to read the spec's of power supplies more closely. About a year ago I had to replace a 650W ThermalTake psu when the computer was upgraded. It was replaced with a 480W Antec, that had a greater output on the 5V and 3.3V rails. The 650W unit, was tested and is up to spec, it now powers a new computer where the mobo and graphics power requirements are on the 12V rails.
I think we need to have better specs available to read! Some load curves would be nice, as that is what actually matters. I suspect many PSU's would look pretty ordinary. I too have replaced many a PSU to cure an apparent mobo/card problem, so many that it's on the short list in my troubleshooting algorithm. Mind you, some graphics cards probably ought to declare their need for a separate power input ( other than via the slot ), as some do....
As I've yet to have a bad Antec, I have a couple of those spare ready to drop in at short notice.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
RE the 3.3/5v vs 12v loading, every mobo with a 4pin 12v connector will power the cpu and most of the mobo off of the 12v rail. That spec includes all p4's, all amd's post 774 (maybe some late model 462 boards as well?). Unless you have a large number of drives and PCI cards drawing 5/3.3v the load is almost always heavily biased in favor of the 12v load. Having problems with inadequate 3.3/5v loading is rarely a problem, what exactly did you have in that box. The normal problem is someone who bought a psu back designed in 01 that can't put out nearly enough 12v for a modern system. Those tend to be 50/50 people who bought based solely on wattage and have either kept thier PSU for the last 5 years or bought a clearance model, or those who got a cheap PSU bundled with thier case. The bundles are almost exclusively obsolencent designs that're being unloaded on suckers, or cheap noisy models with crap regulation and a very low MTB(catastrophic)F.
I have a 550W antec PSU, and
)
I have a 550W antec PSU, and while I have a 5 drive raid, I don't have SLI gfx cards so I thought I was oversized. I'll be ordering a 750W PSU in the near future that should hopefully provide a longer period of overkill than this one.
RE: No, this looks like the
)
Where you see that in stdout err? The talk about 'no usable checkpoint' is clearly form the first start of the workunit.
OOPs, my bad! I must have
)
OOPs, my bad! I must have been looking at the wrong host. :-O
Now that I'm looking at yours (735729, just to make sure), that is kind of strange. It looks you had two in row from the same data pack, and what's really strange is your partner for 57075104 ran it in what looks to be the nominal time for it. Can't tell about the other one since the host doesn't have enough result showing to make any comparisons.
Alinator
RE: I have a 550W antec
)
I'm beginning to think we need to read the spec's of power supplies more closely. About a year ago I had to replace a 650W ThermalTake psu when the computer was upgraded. It was replaced with a 480W Antec, that had a greater output on the 5V and 3.3V rails. The 650W unit, was tested and is up to spec, it now powers a new computer where the mobo and graphics power requirements are on the 12V rails.
RE: I'm beginning to think
)
I think we need to have better specs available to read! Some load curves would be nice, as that is what actually matters. I suspect many PSU's would look pretty ordinary. I too have replaced many a PSU to cure an apparent mobo/card problem, so many that it's on the short list in my troubleshooting algorithm. Mind you, some graphics cards probably ought to declare their need for a separate power input ( other than via the slot ), as some do....
As I've yet to have a bad Antec, I have a couple of those spare ready to drop in at short notice.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
You can find that sort of
)
You can find that sort of data if you look hard enough.
Here's one example, i've seen others which provided detailed crossloading graphs but wasn't able to find one now.
http://www.jonnyguru.com/review_details.php?id=43&page_num=3
RE the 3.3/5v vs 12v loading, every mobo with a 4pin 12v connector will power the cpu and most of the mobo off of the 12v rail. That spec includes all p4's, all amd's post 774 (maybe some late model 462 boards as well?). Unless you have a large number of drives and PCI cards drawing 5/3.3v the load is almost always heavily biased in favor of the 12v load. Having problems with inadequate 3.3/5v loading is rarely a problem, what exactly did you have in that box. The normal problem is someone who bought a psu back designed in 01 that can't put out nearly enough 12v for a modern system. Those tend to be 50/50 people who bought based solely on wattage and have either kept thier PSU for the last 5 years or bought a clearance model, or those who got a cheap PSU bundled with thier case. The bundles are almost exclusively obsolencent designs that're being unloaded on suckers, or cheap noisy models with crap regulation and a very low MTB(catastrophic)F.