H-Compress Performance During the Nov 6, 1997 X9.4 Solar Flare

I had a look at the image processing information in the image headers before
and after the flare in the interest of knowing how observing schedules should be modified
the next time this happens.
.
Here is the comparison based on averages of two pre-flare and two post flare images.

The pre-flare images had the following characteristics using H-Compress:

						Uncompressed	Compression
		Scale Factor	Size (bytes)	Size (bytes)	Factor
C1		     32		119697		 835584		6.98
C2		     64		267384 		1972224		7.38
C3		     32		313223		1894400		6.05
				

       		Compression Time Post-Flare
       		---------------------------
Camera       	Compression Time Pre-Flare	

C1			1.17
C2			1.21
C3			1.19


       		Compression Size Post-Flare
       		---------------------------
Camera       	Compression Size Pre-Flare	

C1			1.28
C2			1.41
C3			1.34

After the flare hit SOHO, we switched to a higher scale factor to get a higher cadence
of images.  We used a scale factor of 192 (H8) for all images from C1, C2 and C3.
this actually lead to a higher cadence than usual since the C1 and C3 had higher 
compression than usual.


						Uncompressed	Compression
		Scale Factor	Size (bytes)	Size (bytes)	Factor
C1		    192		 89302		 835584		9.36
C2		    192		319505 		1972224		6.17
C3		    192		203568		1894400		9.30


Conclusion:

Let's try a scale factor of 128 (H7) next time!

Written: 13 Nov 1997 - DW