Posts by HAL9000

1) Message boards : Number crunching : 1060 3GB Vs 6GB (Message 1850310)
Posted 4 days ago by Profile HAL9000
Post:
I wonder why the 3 GB version of the 1060 GPU is higher in the Top GPU models even though it has it has less cores,

Do you think this is because of more powerful Rigs would have been installing the 3GB since it was released earlier? or is there an actual spec deference between the cards,

Do you think the 6GB will slowly move up the ranking to be similar to the 3 GB.

cheers

Steven

There is a slight difference between the 3GB and 6GB cards. In that the 3GB ones have 1152 cores and the 6GB have 1280 cores.

I would expect the higher performance number for the 3GB cards would more likely be that there are more factory OC'd versions of them running .
2) Message boards : Number crunching : GPU FLOPS: Theory vs Reality (Message 1850309)
Posted 4 days ago by Profile HAL9000
Post:
^hi all, just seen something strange in my results

Boinc self determine théorical Gflop to my GPU ( HD7750 ) a 2048 Gflop, , on technical websites and wiki , the value is about 819 Gflops ...

but now on somme task , it set to 2304 Gflops , is it self calculated with time by result ?

look at theses ..

http://setiathome.berkeley.edu/result.php?resultid=5423626195 >2048
http://setiathome.berkeley.edu/result.php?resultid=5516955845 >2304
http://setiathome.berkeley.edu/result.php?resultid=5514924288 >2048
http://setiathome.berkeley.edu/result.php?resultid=5522146789 >2304

thks for your thoughts ...

As I recall BOINC always reported the GFLOPs of all of my older CAL Radeon GPUs as double the actual value.
It probably has something to do with them having issues with the GPU detection using CAL.
For the numbers you are seeing you are seeing I would guess your GPU is running 1000MHz base and then 1125MHz for turbo?
3) Message boards : News : Web site upgrade (Message 1849357)
Posted 7 days ago by Profile HAL9000
Post:
Yes, I notice this also about a week ago. I'll tweet them and maybe that will get their attention more than this forum. :-)

That issue has been fixed already.
4) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1849173)
Posted 8 days ago by Profile HAL9000
Post:
On paper for similar core, thread, thread configurations AMD and Intel are pretty close for the types of systems I normally build.
Name		c/t	Clock	L3	TDP	MSRP
R5-1400X	4/8	3.5 GHz	8MB	65W	$199
R3-1200X	4/4	3.4 GHz	8MB	65W	$149

i7-7600		4/8	3.6GHz	6MB	65W	$312
i5-7500		4/4	3.4GHz	6MB	65W	$202


Given the AMD chips have a similar TDP and don't have an integrated GPU does lead me to think they will run a fair bit hotter under full load.
5) Message boards : Number crunching : Panic Mode On (104) Server Problems? (Message 1849161)
Posted 8 days ago by Profile HAL9000
Post:
That's because you have told the server you want AP work in your prefs but you have no app for it.

And even when I don't have that option selected I still get the messages in the Event log. Just getting it in the Manager Notices is new.

You can get a message for either MB or AP depending on your app_info.xml no matter what your app selections are.
The quirk in the messages tab/event log has been driving me nuts for years.
My AP only machines would get "you don't have a usable version of MB" message and my MB only machines would get the "you don't have a usable version of AP.
I found if I configured the app that I didn't want to run for hardware I didn't have then the message didn't seem to show up.
Like setting up the NV AP app on a machine that only ran CPU & ATI GPU MB work.
6) Message boards : Number crunching : Panic Mode On (104) Server Problems? (Message 1848929)
Posted 9 days ago by Profile HAL9000
Post:
Two of my computers had full caches early this morning but my daily driver still hadn't reloaded. I had to go through the project toggling exercise to get it to get tasks. What I find interesting is that the Windows 10 computer is able to fill out its cache within a few hours of the project coming back online. Not the case with the Windows 7 machines which always take 1-2 days to recover.

There doesn't seem to be anything specific to those that are having the issue. One of my machines is only allowed network traffic for a 2.5hr time period each day and it isn't having any issues maintaining its cache.
7) Message boards : Number crunching : Panic Mode On (104) Server Problems? (Message 1848870)
Posted 9 days ago by Profile HAL9000
Post:
Over 12 hrs since the outrage ended and the RTS remains drained. This is what I predicted.

I have full caches now, which means many others should as well.
So, if the splitters are able to maintain present creation rate of around 38/second, RTS should start to build in the next hour or so.
That is MY prediction.

Meow.

I've been bouncing off of the limits pretty much right after everything came back online after maintenance.
I guess some hosts just have better odds with work requests than others.
8) Message boards : Number crunching : Problem joining / finding a team (Message 1848784)
Posted 9 days ago by Profile HAL9000
Post:
There seems to be a problem with the team finding function.
If i want so find a team, there is no button for searching.
Is this possible a server problem?

You can press enter after entering your search text, but when I tried it returned nothing.
I'm not sure if that happened because I'm already in a team.
9) Message boards : Number crunching : Panic Mode On (104) Server Problems? (Message 1848718)
Posted 10 days ago by Profile HAL9000
Post:
Why don't we start a fund for a robotic disk library? I used to do Tech Support on commercial tape backup robotic libraries and they aren't all that big. Should be able to find space for a robot at the CoLo or SSL or wherever the splitters are physically located. That would solve the issue of having to have a corporeal entity physically change out the disks as is done currently.

I believe it is just a matter of running a command to load the data from the storage array into the splitting queue.
When they receive a shipment of hard drives from one of the antennas they do have to copy the data to the storage array.
I don't recall the exact process for that, but I believe it is still transferred from the SSL to the CoLo over the 100Mb shared connection that the SSL has to the rest of the campus.
IIRC 2TB drives were what they asked for in the last hardware donation drive. So each drive would hold about 40 data sets (aka tapes) and a full case of 20 drives would be about 800 data sets.
10) Message boards : Number crunching : Panic Mode On (104) Server Problems? (Message 1848691)
Posted 10 days ago by Profile HAL9000
Post:
Hmmm and here we are again with empty caches and it's soon tuesday outage.
I really hope they can fix Centurion and that it would run more then a few days without crashing.

Do we know what the problem is with it?

Unclear to me whether this is a problem with Centurion, or whether it's just down due to not having any work to do.
SSP shows no BLC tapes ready to split.
What's not clear is what process places tapes into place so PFB and GBT splitters can find them and do work.

The name of the process that loads the data for splitting is normally Eric or Jeff.
Unless they have some interns to delegate the task to.
11) Message boards : Number crunching : User achievements thread......... (Message 1848073)
Posted 13 days ago by Profile HAL9000
Post:
i've just reach 2.5M credits now \o/

hard to download 2 Wu at a time with a 56k modem dial up connexion ( adsl outage till 2/2/2017) and i do not find where to modify this setting ..

There is a cc_config.xml option <max_file_xfers_per_project>N</max_file_xfers_per_project>. You could set to 1 instead of the default of 2.
12) Message boards : Number crunching : My Computer Builds (Message 1848069)
Posted 13 days ago by Profile HAL9000
Post:
Intel has a chip with 24 cores that costs $8,898
Intel's Xeon E7-8894 v4 is a monster server chip with 24 cores, 60MB of cache and a maximum clock frequency of 3.4GHz

No Intel chip is as expensive as the new Xeon E7-8894 v4 server processor.

The Xeon E7-8894 v4, which is based on the Broadwell architecture, is priced high for many reasons. It is targeted at fault tolerant servers used by financial or retail companies which need highly reliable systems for transactions and fraud detection. These companies could lose hundreds of millions of dollars if a server crashed.

That even out prices their 72 core chips that are only $6400.
http://ark.intel.com/compare/94035,95829,95830,95831
13) Message boards : Number crunching : Pine A64+ 2GB running 24/7 (Message 1847707)
Posted 15 days ago by Profile HAL9000
Post:
Just an update for anyone interested. I installed the CPU fan. But the numbers aren't adding up.
It's 5V and the sticker on it says it draws 0.1A. That should mean the fan consumes 0.5w. But the meter is now reading 6.6W. An increase of 0.9W. Also meaning this tiny little CPU fan the size of my thumbnail is increasing the power load by 16%! That's ridiculous.

Seems like they left off a digit on the current rating. Maybe it should be 0.18A!
More likely I would guess the 5v supply is on the high end of its tolerance and the resistance of the fan is on the lower end of its tolerance.
14) Message boards : Number crunching : Accept work from other applications? doesn't work (Message 1847701)
Posted 15 days ago by Profile HAL9000
Post:
Yeah that has been broken for a while now. Eric has looked at it a few times but it seems no luck in fixing it as of yet.
15) Message boards : Number crunching : Mac OS Sierra (Message 1847643)
Posted 15 days ago by Profile HAL9000
Post:
If iGPU in use CPU is somehow throttled. Bottleneck ?

Either bottleneck (CPU/iGPU sharing caches & data paths?), or power related- Power limit for complete package; power required to run iGPU reduces available power for CPU?
Or both?

Here are some posts starting from 2014 about the same issue.
A journey: iGPU slowing CPU processing
iGPU tuning
Loading APU to the limit: performance considerations - ongoing research

I believe the leading theory was L3 cache contention when it was being looked into last.
16) Message boards : Number crunching : Mac OS Sierra (Message 1847449)
Posted 16 days ago by Profile HAL9000
Post:
Intel has changed the clock domain for iGPU with their Skylake CPUs. Its now bound to the baseclock (L3-cache, RAM, iGPU) instead of the DMI-connection (chipset and PCIe).
I do see the same behaviour on Linux on a KabyLake i7 : if iGPU in use CPU tasks are slowed down to a fourth, else the CPU cores without iGPU usage are fast.
Doesn't matter if one or more CPU threads are in use, if iGPU used for crunching the CPUs are slowed. The speedloss is bigger than running 8 threads on CPU.
The "sweet spot" seems to be running 6 or 7 threads without iGPU.

Seems they made the issue worse. The 4000 series CPUs would only be slowed by about 50% when using the iGPU. :(
17) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1847410)
Posted 16 days ago by Profile HAL9000
Post:
Despite what MS wants manufactures are not cutting support for Windows7 any time soon. It is still nearly half on the computers on the planet.
To get the support of the newest devices in an older OS you just grab the latest drivers from the manufacture. Sometimes that does require rolling drivers into the OS installation, but that is a pretty simple process.
18) Message boards : Number crunching : GPU FLOPS: Theory vs Reality (Message 1847329)
Posted 16 days ago by Profile HAL9000
Post:
Why no 1050Ti?
Got one on my computer & one on Stephen's... ;)

The most recent data does show the 1050 Ti.
I would guess that since the 1050's were released in 2016 October is the reason they don't show up in 2016 July data.
19) Message boards : Number crunching : Windows 10 - Yea or Nay? (Message 1845874)
Posted 23 days ago by Profile HAL9000
Post:
New bit of M$ nonsense.
All 5 boxes here interact for some tasks, and since these are scheduled (non-SETI) tasks it is expected that time & date will be correct to keep them from stepping on each other. Been having some issues and discovered the following:
1) While Internet Time Update remains enabled, all machines have been changed from time.nist.gov to time.windows.com, and
2) The current time on 2 of 5 boxes was incorrect.
So I live 6 miles as the bird flies from the NIST facility, and I'm gonna go to Redmond for my time?
Right.
And M$ made this change without my knowledge or consent for what reason?
FYI, one of the 5 boxes is Win 7, not 10.

Often settings get set to factory default when updates are applied. This isn't new and has been going on since at least XP.
20) Message boards : Number crunching : Linux (ARM processor) app and alternatives (Message 1845830)
Posted 23 days ago by Profile HAL9000
Post:
Seems like armhf might be a better route to go, but would probably need an app compiled to take advantage of it
I would guess the current Linux ARM apps do not. I does look like a new Linux ARM64 app was just pushed out on Beta about an hour ago. So maybe Eric is looking to expand the Linux ARM app base.


Next 20


 
©2017 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.