A lot...but again, when a new app is released (newer than both 4.79 and 0.03) will our BOINC client automatically update it?
Now I get it!
No, with participating in this beta test you actually switched the platform from "x86-windows" to "anonymous". The client will never replace an anonymous App by itself. However the machines that are still running the "official" 4.79 App will download and use a newer Windows App when it is available automatically.
However, how do I do that? When I deleted app_info.xml and all the einstein 0.03 stuff the BOINC manager got stuck on 'connecting to localhost'. What do I delete to make it run 4.79 again?
Oops - when the app_info.xml is deleted the Client should just look for the 4.79 and download it again if it isn't there. Of course you must terminate the client (not only close the boinc manager window) before you delete the app_info.xml, but I think you guessed this (though it's not explicitely in the instructions) - did you? If not, it's our fault, if so, there's something wrong with the Client...
It would be cool if you could eventually compile an SSE, SSE2, SSE3 version for Windows.
I once compiled the code with SSE optimization and found it to be about 1% faster than without, so there's not much in the normal code that the compiler found to be suitable for SSE.
One _might_ get further porting the vector code that was (re-)written for the AltiVec to SSE (actually SS2 as I doubt that SSE offers enough precision) but this would need more work again and would end up with yet another code version to be maintained. We are aiming for the "auto-vectorize" feature of the new gcc-4, which should give us generic code as well as code for multiple vector units from the same source code, but it will will also take some more time until it is mature enough to be of use for us.
I couldn't find any other reason for the slowdown compared to the "official" App than a slight change in the priority settings, i.e. I think that people noticing a slowdown have something else running on their machines taking time the Ap has previously gotten. I know that the BOINC people changed this for good reasons, i.e. the computer becoming unresponsive when running BOINC in the background, so I would rather leave it that way.
Same problem, very slow with screensaver:
Installed on old machine with Win 2000, 1 GHZ, ATI Rage Pro.
Regular version of Einstein did run, but very slow once the screensaver graphics went on. In non graphics mode, percentages increased by 0.01 by every few seconds. Once screensaver goes on, this takes about 5-10 minutes.
Please see the Frequently Asked Question page. It sounds as if you are not using a desktop graphics mode that has accelerated OpenGL graphics. Without this the screensaver uses a lot of CPU time. You may be able to enable acceleration with the right graphics driver or desktop settings.
Just curious if any more work on the Windows Test Application is planned for the near future.
Yes.
Quote:
Am getting a bit jealous of our Mac and Linux colleagues, who are recently getting a number of new, updated test applications.... :)
Well, for quite a long time they have been jealous of the speed of the Windows App.
Actually the need to bring out a new App version every week because of some problem that shows up in BOINC absolutely isn't something you need to be jealous to.
We will need to modify our science code for the next round of analysis anyway, so there will be at least one full set of new Apps for you to test.
RE: A lot...but again, when
)
Now I get it!
No, with participating in this beta test you actually switched the platform from "x86-windows" to "anonymous". The client will never replace an anonymous App by itself. However the machines that are still running the "official" 4.79 App will download and use a newer Windows App when it is available automatically.
Does this answer your question?
BM
BM
RE: However, how do I do
)
Oops - when the app_info.xml is deleted the Client should just look for the 4.79 and download it again if it isn't there. Of course you must terminate the client (not only close the boinc manager window) before you delete the app_info.xml, but I think you guessed this (though it's not explicitely in the instructions) - did you? If not, it's our fault, if so, there's something wrong with the Client...
BM
BM
RE: It would be cool if you
)
I once compiled the code with SSE optimization and found it to be about 1% faster than without, so there's not much in the normal code that the compiler found to be suitable for SSE.
One _might_ get further porting the vector code that was (re-)written for the AltiVec to SSE (actually SS2 as I doubt that SSE offers enough precision) but this would need more work again and would end up with yet another code version to be maintained. We are aiming for the "auto-vectorize" feature of the new gcc-4, which should give us generic code as well as code for multiple vector units from the same source code, but it will will also take some more time until it is mature enough to be of use for us.
BM
BM
RE: My first WU done with
)
Oh - faster? What machine (CPU)?
BM
BM
RE: Is there a possebility
)
Please write me an eMail. The AEI.MPG.de webpages may help you.
BM
BM
RE: Seti and Einstein both
)
Which client version are you using?
BM
BM
I couldn't find any other
)
I couldn't find any other reason for the slowdown compared to the "official" App than a slight change in the priority settings, i.e. I think that people noticing a slowdown have something else running on their machines taking time the Ap has previously gotten. I know that the BOINC people changed this for good reasons, i.e. the computer becoming unresponsive when running BOINC in the background, so I would rather leave it that way.
BM
BM
RE: Same problem, very slow
)
Please see the Frequently Asked Question page. It sounds as if you are not using a desktop graphics mode that has accelerated OpenGL graphics. Without this the screensaver uses a lot of CPU time. You may be able to enable acceleration with the right graphics driver or desktop settings.
RE: Just curious if any
)
Yes.
Well, for quite a long time they have been jealous of the speed of the Windows App.
Actually the need to bring out a new App version every week because of some problem that shows up in BOINC absolutely isn't something you need to be jealous to.
We will need to modify our science code for the next round of analysis anyway, so there will be at least one full set of new Apps for you to test.
BM
BM
RE: waiting for a solution
)
What type of installation have you chosen - service, shared or single user?
BM
BM