[Insight-developers] Kitware Robots still appear to be dead

Bill Lorensen bill.lorensen at gmail.com
Mon Oct 3 11:19:50 EDT 2011


1 meg is way too large. I already have a patch that BTW found an error
in the 1 meg test.

But I think that 1K is too small.

Let's wait until I compete my analysis. I'm in the measure stage of DMAIC.

On Mon, Oct 3, 2011 at 11:17 AM, Bill Hoffman <bill.hoffman at kitware.com> wrote:
> On 10/3/2011 11:08 AM, Williams, Norman K wrote:
>>
>> On 10/3/11 9:55 AM, "Bill Hoffman"<bill.hoffman at kitware.com>  wrote:
>>
>>> On 10/1/2011 11:55 AM, Bill Lorensen wrote:
>>> Remember, HALF of
>>> the ITK tests are sending more output than ALL the other projects hosted
>>> on the CDash server.
>>
>> Some people have gotten the idea that more output is better. For me the
>> perfect test would produce no output on success, and as much USEFUL output
>> as possible on failure. Otherwise it's a waste of my brain's bandwidth,
>> which I care about more than machine loads and network bandwidth.
>>
> I agree with that as well. Very verbose passing test output is mostly
> useless IMO, which is why ctest caps it at 1K.  More is not better... :)
>  Changing the default to 1meg arbitrarily for all ITK tests seems like over
> kill to me.  Who looks at 1meg of text output...  :)
>
>
>
> -Bill
>
>


More information about the Insight-developers mailing list