Windows S5R3 App 4.07 available for Beta Test

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 3,915
Credit: 193,334,072
RAC: 44,723
Topic 193177

A new Windows App is available from our Beta Test Page.

This App addresses two problems that are mentioned in the "Client Errors of S5R2 Apps" thread in the "Problems and Bug Reports" Forum.

It features a new checkpointing mechanism and code that has been implemented from scratch without the legacies from older Apps and their requirements. It is much more simple and thus, I hope, more reliable. However the checkpoint files are incompatible with the previous ones, thus there can't be a "smooth transition" from the current official App to this Beta.

This App also features a somewhat dynamic decision whether to run graphics or not. It will start up without the graphics libraries GDI32.DLL, OPENGL32.DLL and GLU32.DLL, will try to load them on its own and in case of a failure it will run without graphics instead of crashing with a client error.

As a side effect you can now tell the App to run without graphics at all (for testing or performance reasons) by putting a file EAH_NO_GRAPHICS into the BOINC directory.

Both features haven't been tested much internally, so we rely on you testers even more than before.

Note that with the app_info.xml included here you will only procees and get S5R3 "work". The scheduler, however, doesn't ask for the Apps first, so as long as there are still many S5R2 workunits in the pool you might get some "No work from project - remove app_info.xml to get more E@H Work" messages if the scheduler happens to pick some of these for you. Unfortunately the S5R3 Apps aren't able to process S5R2 work (technically they can, but the results will be found invalid when compared to those of S5R2 Apps).

BM

BM

Stick
Stick
Joined: 24 Feb 05
Posts: 790
Credit: 2,476,981
RAC: 3,290

Windows S5R3 App 4.07 available for Beta Test

Got it going a few minutes ago on my Intel(R) Pentium(R) 4 CPU 2.40GHz with this result. It seems to be working OK. (I've already successfully tested "Show graphics".)

Observation: It ran for at least 2 minutes before the "Progress" meter moved off 0.000%. And, now that the "meter is running", it seems to be "jumpier" than in previous versions. That is, instead of ascending smoothly in 0.001% increments, it tends to progress in "jumps" of larger increments. In other words, it could give the false impression being "hung-up" occasionally.

Question: Is the new "checkpointing mechanism" designed to survive the "System Restore" process (as I reported here and here)? If so, would you like me to test it?

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 3,915
Credit: 193,334,072
RAC: 44,723

RE: Got it going a few

Message 73500 in response to message 73499

Quote:
Got it going a few minutes ago on my Intel(R) Pentium(R) 4 CPU 2.40GHz with this result. It seems to be working OK. (I've already successfully tested "Show graphics".)


That's good news, I guess.

Quote:
Observation: It ran for at least 2 minutes before the "Progress" meter moved off 0.000%. And, now that the "meter is running", it seems to be "jumpier" than in previous versions. That is, instead of ascending smoothly in 0.001% increments, it tends to progress in "jumps" of larger increments. In other words, it could give the false impression being "hung-up" occasionally.


Are you sure that this is a difference in the Apps? I think that the setup of S5R3 should in general stretch the startup time compared to S5R2. I'm not aware of any change that should make this differ between S5R3 Apps 4.01 and 4.07.

Quote:
Question: Is the new "checkpointing mechanism" designed to survive the "System Restore" process (as I reported here and here)? If so, would you like me to test it?


It isn't designed for doing so, and I doubt that it will make any difference in that situation from the previous version. Would be nice to know, though.

BM

BM

Stick
Stick
Joined: 24 Feb 05
Posts: 790
Credit: 2,476,981
RAC: 3,290

RE: RE: Observation: It

Message 73501 in response to message 73500

Quote:
Quote:
Observation: It ran for at least 2 minutes before the "Progress" meter moved off 0.000%. And, now that the "meter is running", it seems to be "jumpier" than in previous versions. That is, instead of ascending smoothly in 0.001% increments, it tends to progress in "jumps" of larger increments. In other words, it could give the false impression being "hung-up" occasionally.

Are you sure that this is a difference in the Apps? I think that the setup of S5R3 should in general stretch the startup time compared to S5R2. I'm not aware of any change that should make this differ between S5R3 Apps 4.01 and 4.07.

I was making a more general comparison (to the way pre-S5R3 Einstein apps behaved). I have only had one v4.01 result (still "In progress") and, frankly, I haven't been monitoring it that closely.

Quote:
Quote:
Question: Is the new "checkpointing mechanism" designed to survive the "System Restore" process (as I reported here and here)? If so, would you like me to test it?

It isn't designed for doing so, and I doubt that it will make any difference in that situation from the previous version. Would be nice to know, though.

BM

I'll try it right after I start my next unit. (No sense in wasting the one I have going now.) BTW: I have exited/restarted BOINC to test "regular checkpointing" and that worked OK.

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 247
Credit: 219,266,664
RAC: 0

RE: Note that with the

Quote:
Note that with the app_info.xml included here you will only procees and get S5R3 "work". The scheduler, however, doesn't ask for the Apps first, so as long as there are still many S5R2 workunits in the pool you might get some "No work from project - remove app_info.xml to get more E@H Work" messages if the scheduler happens to pick some of these for you.
BM


The scheduler seems pretty persistent in wanting me to have an S5R2 work unit! No luck in getting an S5R3 unit after several attempts! We will keep trying..... ;)

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 247
Credit: 219,266,664
RAC: 0

RE: The scheduler seems

Message 73503 in response to message 73502

Quote:
The scheduler seems pretty persistent in wanting me to have an S5R2 work unit! No luck in getting an S5R3 unit after several attempts! We will keep trying..... ;)


I finally did get one after about 15 minutes, and wouldn't you know it, it was right after I stopped trying! ;)

Stick
Stick
Joined: 24 Feb 05
Posts: 790
Credit: 2,476,981
RAC: 3,290

Apparently, the server

Apparently, the server doesn't think my Intel(R) Pentium(R) M processor 1.80GHz should be Beta Testing either. I keep getting the following messages:

9/26/2007 10:21:17 PM|Einstein@Home|Sending scheduler request: Requested by user
9/26/2007 10:21:17 PM|Einstein@Home|Requesting 21600 seconds of new work
9/26/2007 10:21:22 PM|Einstein@Home|Scheduler RPC succeeded [server version 601]
9/26/2007 10:21:22 PM|Einstein@Home|Message from server: To get more Einstein@Home work, finish current work, stop BOINC, remove app_info.xml file, and restart.
9/26/2007 10:21:22 PM|Einstein@Home|Message from server: No work sent
9/26/2007 10:21:22 PM|Einstein@Home|Deferring communication 4 hr 0 min 0 sec, because requested by project

AnRM
AnRM
Joined: 9 Feb 05
Posts: 213
Credit: 4,346,941
RAC: 0

RE: Apparently, the server

Message 73505 in response to message 73504

Quote:

Apparently, the server doesn't think my Intel(R) Pentium(R) M processor 1.80GHz should be Beta Testing either. I keep getting the following messages:

9/26/2007 10:21:17 PM|Einstein@Home|Sending scheduler request: Requested by user
9/26/2007 10:21:17 PM|Einstein@Home|Requesting 21600 seconds of new work
9/26/2007 10:21:22 PM|Einstein@Home|Scheduler RPC succeeded [server version 601]
9/26/2007 10:21:22 PM|Einstein@Home|Message from server: To get more Einstein@Home work, finish current work, stop BOINC, remove app_info.xml file, and restart.
9/26/2007 10:21:22 PM|Einstein@Home|Message from server: No work sent
9/26/2007 10:21:22 PM|Einstein@Home|Deferring communication 4 hr 0 min 0 sec, because requested by project


We get the same message....after several tries we gave up on the Beta test.

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 3,915
Credit: 193,334,072
RAC: 44,723

I see that it probably makes

I see that it probably makes more sense to run this Beta if you already got S5R3 work.

BM

BM

Stick
Stick
Joined: 24 Feb 05
Posts: 790
Credit: 2,476,981
RAC: 3,290

Not sure what you mean by

Message 73507 in response to message 73506

Not sure what you mean by this.

Quote:

I see that it probably makes more sense to run this Beta if you already got S5R3 work.

BM

That is, my Intel(R) Pentium(R) M processor 1.80GHz had just only finished this S5R3 Result when I attempted the switch to v4.07 and started getting these messages. But, as you said here, the new Beta app isn't compatible with the official app:

Quote:
. . . features a new checkpointing mechanism and code that has been implemented from scratch without the legacies from older Apps and their requirements. It is much more simple and thus, I hope, more reliable. However the checkpoint files are incompatible with the previous ones, thus there can't be a "smooth transition" from the current official App to this Beta.

However, it really doesn't matter anymore. I was able to download this WU a few minutes ago. And, as I promised here, I will do a "System Restore" in a little while, and see how it survives.

Mikie Tim T
Mikie Tim T
Joined: 22 Jan 05
Posts: 103
Credit: 207,633,459
RAC: 0

I installed this beta on the

I installed this beta on the first of my machines to run out of SR2 work. It started up fine after downloading a couple of SR3 workunits, but when I click on the Show Graphics button, for testing purposes only as I don't run the screensaver, nothing happened, and I didn't find anything in the Messages tab that would indicate why it failed to startup the graphics window. I tried putting the EAH_NO_GRAPHICS file in the BOINC directory and restarted the BOINC service, and the Show Graphics button is now inactive. As I haven't run any of the regular SR3, I don't have any basis to judge definitively whether this runs any faster in this mode, but it sure does seem as though it does. This is host 710154

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.