Help On A 'Spread' Kind Of Ranking...

Hi all,

Just can't figure this out. I'll try not to give a long-winded explanation (I hope), let's say this is the table, for example:
[Table1]
RowNumber | Value
1 | 4
2 | 6
3 | 3
4 |10
5 | 6
6 | 5
7 | 8
8 | 8
9 | 2
10 | 6

I want to write a query that creates a column whose values are similar to (ROW_NUMBER -1), and resets each time a value is present.
Using the above example, 6 will be the test value. So, I want to know how many rows it takes till 6 is repeated.

Output would be:
RowNumber | Value | Spread
1 | 4 | 1
2 | 6 | 0
3 | 3 | 1
4 |10 | 2
5 | 6 | 0
6 | 5 | 1
7 | 8 | 2
8 | 8 | 3
9 | 2 | 4
10 | 6 | 0

I think PARTITION BY can be used somewhere, with the ORDER BY on the [RowNumber] - but I'm just not sure on which ranking function to use, and how to reset the Ranking on the number 6 (above example).

Any help would be appreciated. This has been a mind-teaser for me, and I give

Thanks,
Denvas