Announcing Windows 10 Insider Preview Fast+Skip Build 17046 for PC Insider

OK. Thanks, Tony. Buildfeed is up and its what I predicted. Build 17054 is their candidates. Here's the thing that baffles me about the AMD problem. Clearly, they have AMD machines in Redmond since Dona preface to that release explicitly said it would not boot on AMD system. So why the haste in releasing it. Somebody (not me - I have a life away from this) could research what public feedback they got from Intel owners. Not much was posted here other than justifiable complaints about the AMD glitch. Perhaps Intel itself requested the build?? I myself would be most interested if MS released some of their automated testing programs - perhaps we could find bugs in that system. Someday (probably not in my lifetime) MS could release the version of the update program which dumps the output of the commands during one of all of the 3 boots involves. I always modify the /etc/default/grub file to eliminate quiet and splash parameters so I can see exactly when a boot fails during update (which is almost never).

slicendice said:

Here is a very over simplified Unit-test:

Code:

int valueA = 1;
int valueB = 3;
int expectedResult = 4;

bool SumReturnsCorrectResult(int valueA, int valueB, int expectedResult)
{
   int result = Sum(valueA, valueB);
   
   return Assert(expectedResult, actualResult);
}

And here is the actual code to be tested:

Code:

int Sum(int A, int B)
{
   return A + B;
}

The function SumReturnsCorrectResult will return TRUE in this case.

or it would if there wasn't a bug in SumReturnsCorrectResult.

cereberus said:

I disagree - we were using alpha/beta terms in late 60's, early 70s when main languages were Fortran and Cobol!

Since you must be older than me and having a much better memory than me, I believe you. Those years I was not even thought of yet.

kado897 said:

or it would if there wasn't a bug in SumReturnsCorrectResult.

Oups, correcting...

Here is an interesting article about the history of Alpha and Beta testing. In short it goes back probably to the 50's when IBM was testing hardware.

Alpha, Beta, and Sometimes Gamma

martyfelker said:

Here is an interesting article about the history of Alpha and Beta testing. In short it goes back probably to the 50's when IBM was testing hardware.

Alpha, Beta, and Sometimes Gamma

Nice and short and explains the terms clearly. Thanks. :)

martyfelker said:

Here is an interesting article about the history of Alpha and Beta testing. In short it goes back probably to the 50's when IBM was testing hardware.

Alpha, Beta, and Sometimes Gamma

I didn't know it went back that far but it seems to refer to the state of the software development more than a type of testing.

slicendice said:

Nice and short and explains the terms clearly. Thanks. :)

Your welcome. Thus by definition, we are testing beta software. On rare occasions, I guess alpha code get uploaded to the update
servers.

2/3 host OS's I am testing are in beta although it is more meaningful for Linux to know which repositories are enabled. For example, I have enabled the experimental debian repos and so far the only danger I potentially have is running up against disk space problems since 3 OS's are sharing the SSD. Now I do have a spare 128GB Samsung but I don't know if inserting a 4th disk drive will be stable enough. My other option is to delete the Windows Server 2016 partition. Since I have not purchased that OS I running "slmgr /rearm" to extend the evaluation. I have a Wiindows Server 2016 which I'm going to try and rearm after I install the latest CU for Windows 2016.
.