When I try to generate response spectra, this message: "out of memory" appears at screen. Then a proccess seems to continue but never ends and I can't cancel it.
Can, please, someone tell me what could be the error.
Database I am using has 32001 values of acc.
Thank you very much
out of memory
Re: out of memory
Tavo, could you try a smaller file to see if it will process ok? I have loaded files larger than 32,768 I get a message that such files cannot be processed for Fourier and Power spectra. For such files I can typically process only elastic spectra.
I am not sure if this is a limitation of memory on your computer, but that is a possibility.
One thing I have done to process large files is to strip every other value of acceleration and double the time step using a convenient word processor. Of course, when you do this, you'll need to make some measure of effort to assure that you still have a reasonable depiction of the time history.
Another option is to load the accelerogram and observe portions at the beginning and/or end which you might be able to delete without seriously affecting the input.
The best option is to find a shorter accelerogram which still matches your magnitude, site, and distance criteria as well as any regime (intraplate, interplate, . .) criteria and fault type criteria you might have.
Best of luck Tavo.
I am not sure if this is a limitation of memory on your computer, but that is a possibility.
One thing I have done to process large files is to strip every other value of acceleration and double the time step using a convenient word processor. Of course, when you do this, you'll need to make some measure of effort to assure that you still have a reasonable depiction of the time history.
Another option is to load the accelerogram and observe portions at the beginning and/or end which you might be able to delete without seriously affecting the input.
The best option is to find a shorter accelerogram which still matches your magnitude, site, and distance criteria as well as any regime (intraplate, interplate, . .) criteria and fault type criteria you might have.
Best of luck Tavo.
Tim Huff
Re: out of memory
Ok, i will try using dt=0.01 seconds.
I already cut the acc to show only the strong part of EQ and some of the harmonic post EQ event.
Also de computer could be a bit out of date!
Thank you very much huffte!
I already cut the acc to show only the strong part of EQ and some of the harmonic post EQ event.
Also de computer could be a bit out of date!
Thank you very much huffte!
-
seismosoft
- Posts: 1271
- Joined: 06 Jul 2007, 04:55
Re: out of memory
Hi,
As a complement to huffte's good advice, we add only that there are a couple of other posts in this Forum that discuss out-of-memory issues (you may find them through a search on it), and which may provide useful feedback to you. One non-exhaustive example of such posts is: http://www.seismosoft.com/forum/topic.asp?TOPIC_ID=611
Seismosoft Support
As a complement to huffte's good advice, we add only that there are a couple of other posts in this Forum that discuss out-of-memory issues (you may find them through a search on it), and which may provide useful feedback to you. One non-exhaustive example of such posts is: http://www.seismosoft.com/forum/topic.asp?TOPIC_ID=611
Seismosoft Support