dotCover vs NCrunch, dotCover is way too slow...

Answered

I am currently evaluating dotCover and comparing it against NCrunch... 

I tested dotCover for about 30 minutes (yes I turned on continuous testing on save and dirty test)... my observation is there is 10 - 20 seconds delay before dotCover start retesting my affected code... which I find appalling... NCrunch however starts retesting my code almost instantaneously... 

Is this normal? Or is there something wrong with my configuration? For me a delay that long breaks the purpose of continuous testing... 

My PC is pretty beefy... (i7, SSD, 16Gb RAM - i know VS only uses 4Gb but still..) besides NCrunch seems to work instantly so I dont believe it is a hardware bottleneck...

Any comment before I give my money to NCrunch instead of jetbrains? 

4 comments

No response. Right now I have given up on continuous testing due to this. Instead I will run the tests manually whenever I want to.

0

Hello,

I'm extremely sorry for missing the request!

Unfortunately, I can confirm that now dotCover's Continuous Testing concedes to NCrunch on performance. Fixing this is one of our top-priority tasks.

However, 20 seconds delay before rerunning affected tests looks too bad. I guess that it can be caused by using VS build instead of ReSharper's one, which is optimized for building only changed projects.

ReSharper Build feature can be enabled here: ReSharper >> Options >> Tools >> Build >> General >> Build Engine.

Sorry for the inconvenience.

0

Good to know this is being looked into... I tried Resharper build once or twice... but always went back to the default microsoft build because I didnt notice any speed improvement with it (I didnt timed it, but there is no perceived speed improvement as far as I can tell), and the Resharper build has issues of its own... it is just not as stable as the visual studio's build.

0

Please sign in to leave a comment.