V6.34 - 2 issues I've come across ( maybe! )

Moderators: Site Moderators, PandeGroup

Re: V6.34 - 2 issues I've come across ( maybe! )

Postby PinHead » Wed Feb 01, 2012 11:36 pm

Thanks for the info Bruce.

The 2 systems I posted logs from both run NTFS. Just as info.

I will say that I was kind of expecting an answer like file chkpoint files are built with X amount of values on normal run and X - Y amount of values on -forceasm runs.
So when it starts with -forceasm and then restarted without -forceasm, it see the chkpoint files as corrupted. But like you said, a hunch of mine based on too few observations. ;)
PinHead
 
Posts: 351
Joined: Tue Jan 24, 2012 3:43 am

Re: V6.34 - 2 issues I've come across ( maybe! )

Postby bruce » Thu Feb 02, 2012 3:06 am

That theory would imply that EVERY time you added or removed -forceasm, the WU would restart from zero. I'm pretty sure that I've added/removed it without the WU restarting. Since this is from memory, not an actual experiment, we can't count it toward an observed fact, but it fits in the hunch category.

Gather some more information and report back. What I would do is wait for a new WU. Let it run at least to the next checkpoint (but not a lot longer than that). Stop/add or remove -forceasm/restart. Did it start from the last checkpoint or from the beginning?

NOTE: Different cores make checkpoints at different times and the checkpoint files have various names. Some cores write a checkpoint at every frame; some do not. Some cores write a checkpoint every 15 minutes after the WU is (re-)started; some do not. Some cores write a checkpoint every 15 minutes after the last frame began; others do not. Some cores follow an entirely different pattern.

Depending on the core being used, I often find several files that are updated regularly including something called *.cpt or *.chk or *CP.arc or thing like that. Something called *.log or *.txt may be updated more frequently and can be ignored. Figure out the pattern and DO NOT interrupt it when it's about to write a checkpoint.
bruce
 
Posts: 20837
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: V6.34 - 2 issues I've come across ( maybe! )

Postby PinHead » Thu Feb 02, 2012 4:14 am

bruce wrote:That theory would imply that EVERY time you added or removed -forceasm, the WU would restart from zero. I'm pretty sure that I've added/removed it without the WU restarting. Since this is from memory, not an actual experiment, we can't count it toward an observed fact, but it fits in the hunch category.

Gather some more information and report back. What I would do is wait for a new WU. Let it run at least to the next checkpoint (but not a lot longer than that). Stop/add or remove -forceasm/restart. Did it start from the last checkpoint or from the beginning?


I'm not 100% on the add -forceasm theory, but the remove theory is 2 for 2 on 2 different systems.

Either way, I'll do some more testing on it.

BTW, did I mention testing code is what pays my bills and vacation each year :lol:
PinHead
 
Posts: 351
Joined: Tue Jan 24, 2012 3:43 am

Re: V6.34 - 2 issues I've come across ( maybe! )

Postby bruce » Thu Feb 02, 2012 4:34 am

PinHead wrote:BTW, did I mention testing code is what pays my bills and vacation each year :lol:


Sorry, but FAH is a volunteer project -- including 99% of the testing (Well, maybe I'm exaggerating -- but nobody outside of the Pande Group is getting any bills paid or any vacations -- and I'm not really sure about a lot of the PG.

TIA for any testing that you do.
bruce
 
Posts: 20837
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: V6.34 - 2 issues I've come across ( maybe! )

Postby PinHead » Thu Feb 02, 2012 4:57 am

Wasn't asking for coin, sometimes I do it for fun!

JUST BECAUSE
PinHead
 
Posts: 351
Joined: Tue Jan 24, 2012 3:43 am

Previous

Return to V6.34Beta SMP2 with passkey [Not Bigadv]

Who is online

Users browsing this forum: No registered users and 1 guest

cron