Thanks for your advice, all. I downloaded the Surfer 8 demo software and
found that Surfer 8's results are much closer to mine than Surfer 6.
This, Mr. McCarn's recollection, and a cursory look at Surfer 6's memory
usage during processing all indicate that Surfer 6 uses 4-byte floating
point arithmetic, even though Golden Software tech support claims
otherwise.

I wish I could nail down why my results are still different than Surfer
8 by up to 5%. I think it is probably floating point error that
accumulates over the many calculations required to solve for Kriging
weights, so I plan to look at some numerical methods books and do some
testing etc. If you have experience with floating-point problems and
would like to share, I'm all ears!
 
Eva

-----Original Message-----
From: Bambang Trisasongko [mailto:[EMAIL PROTECTED]] 
Sent: Wednesday, February 20, 2002 2:58 AM
To: Eva Pierce
Cc: [EMAIL PROTECTED]
Subject: Re: AI-GEOSTATS: Surfer 6.04

Hi Eva,

I usually saw the problem in remote sensing image analysis. From my
experience, the difference is based on different use of floating point
data
types. As you may refer on standard textbook on programming languages,
there are slightly difference between languages in handling floating
point
data.
Another problem is coding mathematical/statistical equations. Solving
equations may be constructed in different ways. I usually refer to some
numerical analysis book such as John Rice, Don Knuth or if you using
specific programming languages, there are versions of Numerical Recipes
by
Press et al.
I hope it can help.

Regards,

Bambang




--
* To post a message to the list, send it to [EMAIL PROTECTED]
* As a general service to the users, please remember to post a summary of any useful 
responses to your questions.
* To unsubscribe, send an email to [EMAIL PROTECTED] with no subject and "unsubscribe 
ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND 
Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org

Reply via email to