Sandman
In article <f1ju1ad55jnmcbf58u8t8jvttl94othufv@4ax.com>, Eric Stevens wrote:

1. HPS is, in the example, a pixel-altering process. 2. JPG compression is a pixel-altering process 3. Gaussian blur is a pixel-altering process.

3 can *NOT* reverse 1 with 2 in the middle. It is impossible. It can *counteract* it and the end result may be satisfactory, but number 1 has *NOT* been reversed. Pixels have been altered in the interrim that 3 can not take into account.

Eric Stevens
You haven't understood my diagram above. Once I would have tried to explain it to you but now I know there is no point.

Sandman
Typical Eric, can't really respond to the facts above so he dances his little dance while saying nothing.

Eric Stevens
They are not facts.

Sandman
Yes, they are.

Eric Stevens
You just think they are. It's what lies behind these 'facts' which is important.

Sandman
Again - 3 can not reverse 1 with 2 in the middle. Try to get that through your old thick skull.

If you remove 2, then 3 can reverse 1. Simple mathematics, I'll show it to you:

1. 5 * 5 = 25 2. 25 * 0.99 = 24.75 3. 24.75 / 5 = 4.95

That's what the above steps does. Step one is sharpening, step two is JPG compression - a *LOSSY* process and step 3 is blur. Step 3 has *NOT* reversed the algorithm in step 1, we do NOT have the same value as we started with.

Remove the second step:

1. 5 * 5 = 25 2. 25 / 5 = 5

Now the first algorithm (multiplication) has been reversed by the second algorithm (division) and the result is the same as what we started with.

These are *facts* Eric, not guesses or mere claims. This is how mathematics works.