Posts by Richard Haselgrove

1) Message boards : Cafe Einstein : Seti Refuge Bar & Bistro Wing of Cafe Einstein (Message 136666)
Posted 1 day ago by Richard Haselgrove
Well I looked up online 'OpenReach' ( allegedly BT's fixit brigade ) - for a laugh and to see if I could help with Richard's drooping problem - and found this ( my red emphasis ) :

If you have a fault with your phone or broadband, or wish to arrange an installation, you must call your service provider...

and so, straight off the bat as it were, I perceive a source of grave difficulty.

Cheers, Mike.

Exactly so. Even though both organisations are part of "British Telecommunications PLC, registered no. 1800000, registered office 81 Newgate Street London EC1A 7AJ", there is supposedly a world of difference between BT (my service provider) and Openreach (aka BT wholesale). Officially, I'm not allowed to talk to Openreach - but in reality, they're a friendly enough bunch if you happen to come across one of them down its burrow on your way to the newsagent or sandwich shop.

What they won't do - in reality, can't do, because they're as much in the dark as the rest of us - is give any clue about forward plans. And BT retail are as much use as the proverbial chocolate teapot. I rang up once to try and place an order, and spoke to a lovely lady in Cornwall (300 miles to the south-west) - who was charmingly sympathetic, but had no knowledge at all of the local geography, and could only read 'computer says no' off her screen. I have notes that the service in question (Fibre to the cabinet) was first scheduled to be available by 31 December 2011 - and the south bank of the canal got theirs round about that time. But us northbankers are still waiting.

(mind you, the only alternative - the fully private capitalist cable TV industry - surveyed the village and planned their coax rollout before I moved to this house 21 years ago. We're still waiting for that to happen too)
2) Message boards : Cafe Einstein : Seti Refuge Bar & Bistro Wing of Cafe Einstein (Message 136646)
Posted 1 day ago by Richard Haselgrove
Someone should explain to them what bridges are for...

Canal bridges are for swinging...

Are they? I'll have a go at that :) Sounds fun.

*Pause to consider Richard's response* oh... *remove foot from mouth* ... we don't have many of those round my way to my knowledge. Ours stay put a lot. Might I suggest you don't point out the bridges to them then...?

Our copper lines cross from the tops of two elderly wooden poles, which are starting to stoop and bow towards each other under the strain of years.

Going to make them work till they drop I suppose :/

They've certainly reached 'droop':



Good thing those trunk cables run parallel to the road, not across it - else the local haulage company would pull them down with the first trailer returning to the yard. My nomadic (underground) BT engineer - who has now folded his stool and his knitting equipment, and stolen off into the night before I could direct him towards London - didn't envy his overground colleague who would have to shin up the pole on the far side, past the razor wire defending the haulage depot.
3) Message boards : Cafe Einstein : Seti Refuge Bar & Bistro Wing of Cafe Einstein (Message 136593)
Posted 2 days ago by Richard Haselgrove
Someone should explain to them what bridges are for...

So nice to see you today! :) Hope all is going very pleasantly indeed :)

Canal bridges are for swinging...

Our copper lines cross from the tops of two elderly wooden poles, which are starting to stoop and bow towards each other under the strain of years.
4) Message boards : Cafe Einstein : Seti Refuge Bar & Bistro Wing of Cafe Einstein (Message 136588)
Posted 2 days ago by Richard Haselgrove
Aside : So a 'BT engineer' is some species of troll or orc is it ? You know, evil incarnate .... born underground ... later emerges into the sunlight ... to wreak havoc upon kind & innocent surface dwellers ... eternal plague of the righteous etc ?

[Frances Urquhart voice on]
You might very well think that, but I couldn't possibly comment!
[Frances Urquhart voice off]

Wikipedia - BT Engineer

A rapidly declining species once indigenous to the British Isles, and rapidly dying out particularly in parts of London. Zoologists disagree as to the cause, but some rumours abound of the few survivors found staggering the streets with wild eyes and 3 foot beards, asking what year it is.

If you know of their whereabouts Chris *sweetest and... incidentally... most deranged smile* you will let me know won't you? :)

So - nearly broke the double posting rule in David's thread :))))

*look round seti refuge* might have to install a naughty step somewhere... *pause to glare at install* *ponder whether it should be instal* *decide it shouldn't be and leave*

Some of them have migrated north. Santa's elves have presented my village with a shiny new Infinity cabinet, and earlier this week there was a sighting of the aforementioned, sitting on a camp stool and solemnly knitting its insides. But apparently they still haven't found a way of floating optical fibre across the canal...
5) Message boards : Problems and Bug Reports : BRP4 Intel GPU app feedback thread (Message 136511)
Posted 3 days ago by Richard Haselgrove
Host 11672358 has asked for work:

Host 11672358
http://einstein.phys.uwm.edu/host_sched_logs/11672/11672358

2014-12-17 07:01:19.9178 [PID=23748] [version] Checking plan class 'opencl-intel_gpu'
2014-12-17 07:01:19.9179 [PID=23748] [version] parsed project prefs setting 'gpu_util_brp': 0.000000
2014-12-17 07:01:19.9179 [PID=23748] [version] OpenCL driver version: 10.18.10.3977; platform version: OpenCL 1.2; device version: OpenCL 1.2
2014-12-17 07:01:19.9179 [PID=23748] [version] Peak flops supplied: 6.4e+10
2014-12-17 07:01:19.9179 [PID=23748] [version] plan class ok


Claggy

Ouch. That one's newer than the downloads I found on Monday, and I can't find it on Intel's site - and it still doesn't seem to be working. I hope it's not an OEM special.
6) Message boards : Problems and Bug Reports : BRP4 Intel GPU app feedback thread (Message 136489)
Posted 4 days ago by Richard Haselgrove
The 4th WU is a validation nightmare, but at least contains an example of a Haswell validating against the NEON app. That Haswell might simply use the older driver, though.

MrS

Actually, it uses an even older driver:

2014-12-16 13:49:54.5165 [PID=4684 ] [version] Checking plan class 'opencl-intel_gpu'
2014-12-16 13:49:54.5165 [PID=4684 ] [version] parsed project prefs setting 'gpu_util_brp': 0.000000
2014-12-16 13:49:54.5165 [PID=4684 ] [version] OpenCL driver version: 9.18.10.3186; platform version: OpenCL 1.2; device version: OpenCL 1.2
2014-12-16 13:49:54.5165 [PID=4684 ] [version] Peak flops supplied: 6.4e+10
2014-12-16 13:49:54.5165 [PID=4684 ] [version] plan class ok

That one should be good, judging by Bikeman's remarks about the problems we had at the other end of the driver range when the application was first launched.
7) Message boards : Problems and Bug Reports : BRP4 Intel GPU app feedback thread (Message 136427)
Posted 6 days ago by Richard Haselgrove
Hi Richard!

Good morning Bernd.

Firstly, I speak for Windows only (and in my case, it's 64-bit Windows 7).

I'd guess that the problem affects the entire range of Intel CPUs with integrated GPUs - we know them as HD 2500/4000/4400/4600/5000/5100/5200/5300 graphics chipsets, but there are some indications that newer Pentium and even Celeron CPUs with HD graphics can also run OpenCL on the device.

I find the Intel driver download center hard to navigate, but driver searches tend to end up on a page like

https://downloadcenter.intel.com/SearchResult.aspx?lang=eng&ProdId=3720
(for reference only - please do not use these drivers)

Both the drivers on that page - 10.18.10.3907 dated August 2014 and 10.18.10.3960 dated October 2014 - seem to produce a low validation rate with the current Einstein Intel GPU application (and also the equivalent SETI application).

All we can say for certain is that there was a change in behaviour after the 10.18.10.3621 driver, which produces a high validation rate. Quite what goes wrong - what changes in the result file - I'm afraid I can't say: that might be easier for you or Bikeman to extract from the returned results.

Up to you to judge whether the current error return rate from the Intel GPU app is acceptable to the project (especially considering whether validations arising from two 'new driver' hosts matching are scientifically useful).

As an immediate stopgap, setting a maximum driver version of 10.18.10.3621 should restore consistency of results - but will cut off a large number of hosts, whose users will have difficulty locating the older drivers for download. From that point forward, whether there's anything that Einstein can do to work with the newer drivers, or whether this is properly categorised as a driver bug requiring remedial action by Intel, is beyond my knowledge.

(I do have a HD 4600 'Haswell' which I'm happy to run any tests on if needed)
8) Message boards : Problems and Bug Reports : Conflict with SETI? (Message 136374)
Posted 7 days ago by Richard Haselgrove
Thanks to a Richard Haselgrove reply in another thread, I can provide you a link to design documentation for the system which replaced debt.

credit-driven scheduling

But do be aware that the measure used is Recent Estimated Credit (REC), not actual credit - so all the vagaries of pending tasks and credit rates are ignored.

You will find that over time the two projects (any two, or more, projects) will settle down and co-exist - but any perturbation, such as one project running out of work for a while, will upset the equilibrium for a while until REC equalises again.
9) Message boards : Problems and Bug Reports : BRP4 Intel GPU app feedback thread (Message 136273)
Posted 10 days ago by Richard Haselgrove
A user at SETI (which has similar problems to Einstein with the newer drivers) has found this link on the official Intel download site:

https://downloadcenter.intel.com/Detail_Desc.aspx?DwnldID=23886
'Installs the video graphics driver version 10.18.10.3621 for IntelĀ® NUC products.'

Although the description refers to what seems to be an embedded Celeron development kit, the ReadMe file has the standard compatibility list for third and fourth generation Core processors, and the zip file seems very similar to the one I have saved from my original 10.18.10.3621 download for the Haswell. I'll do a fuller comparison in the morning.

Comparison done for the 64-bit version (I don't have an iGPU running under 32-bit Windows, so I didn't download a reference copy of the original release).

That 'NUC' download is guaranteed 100% identical to the original version 10.18.10.3621 release, which I downloaded as 'win64_153322.zip' back in June - all 352 files have the same CRC and MD5.
10) Message boards : Problems and Bug Reports : BRP4 Intel GPU app feedback thread (Message 136266)
Posted 10 days ago by Richard Haselgrove
@Richard: that would be an easy link to the old driver, or do you have something more in mind? As a reason for posting this, not asking generally ;)

MrS

A small number - a very small number - of users will take sufficient interest in the work their computers are doing after attaching to this project to realise that there is a problem, find their way to this thread, and discover that there is some interim action they can take to reduce the error rate.

No substitute for a properly engineered solution at source, I know, but since there doesn't seem to be any other course of action accessible to end users, I thought it might help a little.


Next 10

Home · Your account · Message boards

This material is based upon work supported by the National Science Foundation (NSF) under Grants PHY-1104902, PHY-1104617 and PHY-1105572 and by the Max Planck Gesellschaft (MPG). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the investigators and do not necessarily reflect the views of the NSF or the MPG.

Copyright © 2014 Bruce Allen