Thanks Gary and Claus. I got it working. Is it possible to convert
the 3 column data into matrix grid. I tried using INDEX and MATCH as
mentioned here.
http://exceltactics.com/vlookup-mult...g-index-match/
but my data set will have more than 250 unique rows and more than
4000 unique columns or vice-verse. Is there an efficient way to
achieve this? Ultimately I would like to plot the data as a surface
plot.

I'm seeing this data as very much resembling the output log from some
sort of a data recorder, and so most I'm familiar with have their own
software for manipulating that data in various ways. Is this an option?

You could create an array of arrays to build the matrix, then transfer
that to a 2D array for output to a worksheet. I'm inclined to go the
4000 rows by 250 columns route, but either way will result a fairly
large load on system resources. Especially if you go the VLOOKUP route
what with that many formulas (4000*250=1M cells)!

Are you wanting a grid that contains 1 row only for each Retention Time
without duplicate values across the row?

Also, note that the parsing results are 'text' and so would need to be
reprocessed as Variant so they're useable as numeric values when they
land on the grid in order to 'plot' efficiently.

Also, how do you want the value pairs to be displayed in the grid?
Current parse returns each value to its own column and so your 1st
example re-parses to...

0.6:30.92:269:30.99:317:32.59:302:33.26:337...

...and so on.

Is this for '#NPOINTS= 10' only, or for all data? (An example of the
expected result would be nice to work with!)

--
Garry

Free usenet access at http://www.eternal-september.org
Classic VB Users Regroup!
comp.lang.basic.visual.misc
microsoft.public.vb.general.discussion