I would say no. I have been running DDD-T on several computers 24/7 with no errors at all. Some projects have had a few errors in the calculation until they were beta tested and cleaned up. DDD-T has been quite clean from what I have seen.

If your workunits error out after only a very short time, that indicates a problem of some kind.

Go to my grid / results status to see whether others are getting errors on the same workunit. If so, the techs would like to know which one it is.

Failing that, errors are usually caused by either bad memory sticks or excess overclocking at your end. I posted a thread here about memtest86+ to validate your memory. There's a page of discussion at wikipedia and a download link.

https://secure.worldcommunitygrid.org/forums/wcg/viewthread?thread=19724#161430

If you have processed a lot of errors, BOINC will reduce your daily quota until you return more clean results. It will then build it up again.

The estimated time is only a guess, but should be fairly close on WCG projects. It is continually adjusted as more elapsed time is accumulated.

If you have a lot of recent errors, the BOINC history may be having a hard time getting an accurate estimate.

I have seen other BOINC projects (not at WCG) with crazy estimates, like 50 hours for something that only runs 15 minutes. That doesn't allow for many workunits in the queue, which is based on adding estimated length jobs until the queue size you have set is reached.