When project links change, that source reports "0".

We have all of these old historical devices that aren't active, so the statcounters didn't use 'devices' here.

Well it seems to me that the WCG numbers are possibly wrong, not the Folding ones.
Where are the WCG numbers from?

Elsewhere I see this:

Average floating point operations per second 379,472.8 GigaFLOPS / 379.473 TeraFLOPS (Seti) (active users 204,939)

and this:

Average floating point operations per second 113,622.8 GigaFLOPS / 113.623 TeraFLOPS (WCG boinc clients only) (active users 56,349)

Obviously there are no comparitive numbers for Folding, being non-boinc.

Just thought you'd like to see the Seti comparison.
Is a shame to see that according to Folding stats less than 10% of total historical cpu count is active.....though many of those would be real old rigs that have since died.

Does WCG keep a count of "currently" active processors?
I recall it being something that was going to be added....dunno if it has.

I never said that number was wrong. I just said the page is completely wrong.

The Google numbers make no sense to me - try the Google searches, you get completely different numbers.

Results 1 - 10 of about 1,870,000 English pages for "World Community Grid"
Results 1 - 10 of about 540,000 English pages for "Folding@Home"

Then the SETI number is wrong - clearly the page is out of date.
The UD figure shouldn't be there at all - they're gone.

That just leaves the WCG and F@H figures, and (as presumably you know) they are calculated using totally different methods and simply aren't comparable. The WCG figure is CPU-days, while the F@H figure is an actual device count (a very generous count, including any host that contacted the server in the last 50 days). Chalk and cheese - no comparison possible.

If I thought the page was put there deliberately, then I would accuse Dr. Pande of lying with statistics. However, I think the page was just forgotten, and the data is entirely meaningless.

Well, that's odd.

Anyway, it supports my belief that the numbers are misleading. If different search engines give such different numbers, what use are they for comparison?

On Topic, think i saw a number a few days ago just for BOINC as being like 103 thousand active recent average. Add to that 25-30% for the single core UD clients and you'll come close to 130k machines sweating out reguilarly for WCG.

The particular number is on the wishlist to be added to the stats pages, so until then wet thumb, read the signs and wind direction.