I've used VNAV Descent under many different conditions, and though in truth I don't believe the NGX lives up to the expectations that PMDG created (there are more issues than have been identified thus far), the VNAV Descent is the best I've ever experienced. In fact, it's been spot on in every normal scenario I've thrown at it. Of course you can't ask it to descend outside the capabilities of the aircraft flight model or compute something that isn't possible (not suggesting the poster did this).

Spencer, Hope this helps and does not confuse..... when you enter a STAR in the FMC, the FMC may or may not enter the correct altitude constraints. Real pilots (I'm a small one, not commercial) have to check the FMC constraints against the charts as sometimes the data in the db for the fmc is wrong or out of date. If the FMC complains about "DES PATH UNACHIEVABLE" it could have many reasons but the main one I've found is that the altitude and/or speed constraints in the STAR are not correct. Lets look at the possible constraint types: /10000 or /FL100 or /100 “ AT FL 100 “ = Bold large text in FMC- /10000B or /FL100B or /100B “ AT FL 100 OR BELOW “- /10000A or /FL100A or /100A “ AT FL 100 OR ABOVE “- /8000A10000B or /FL080AFL100B or /080A100B “ BETWEEN FL80 AND FL100 “ The most inflexible type of constraint here is the AT constraint which means the FMC has to calculate the AC MUST be AT that height at that point.Example:If your first STAR way-point has an AT or ABOVE constraint of 15000 and you have an AT constraint of 5000 just 3 miles on, there is no way the AC can make it and you'll get "DES PATH UNACHIEVABLE" (This was an extreme example but demonstrates the issue) In order to get rid of the " "DES PATH UNACHIEVABLE" you would need to change the constraint to an AT or ABOVE constraint by entering /5000A on the vnav page for the way-point. This gives the FMC enough flexibly to plot a descent path which is achievable inside normal operations from the aircraft.