>I totally disagree with many of the responses here, Eric, it would help the discussion to move along if you could say what you actually disagreed with, and for what reason."Here's one rule of thumb no one should disagree with: The Captain Always has the final say. Take charge of your descent and plan it accordingly to terrain, Minnimum Descent altitude, and obstacles!"Sorry about this but I do disagree, if we're talking about IFR under ATC control, not pure VFR flights. ATC regulations are not rules of thumb, they are law, and international law at that. The PIC is ultimately responsible for the operation and safety of his aircraft, but he/she must follow ATC clearances unless they are illegal or threaten safety, in which case he/she is obliged to request an alternate clearance. (See http://www.faa.gov/ATpubs/AIM/Chap4/aim0404.html#4-4-1 for chapter and verse).In the situation where the PIC wants an early descent, he/she would need to ask ATC, and could only descend when ATC had given the appropriate descent clearance. The PIC certainly does not have "the final say", and would only descend without clearance if he wanted to give his passengers some thrilling near-misses and take early retirement at the end of the flight. ;)

I have found that MSFS ATC correctly matches all of the ILS charts that I have flown, I would suggest downloading a few charts and see if the vectors and altitudes are correct for the approaches you are flying. Some have changed obviously because they have been out of cycle for sometime and FS2004 was released in like september 03 or something.Jeff USAFBTW: 12000 ft isn't a flight level, 18000-60000 are the correct FL's.

One other thing I would note - FS ATC will not clear either up or down until you are with about 300ft of your assigned altitude (for your example, you will not be cleared lower than 8000 until you are within about 8300 ft). Make sure you reset your altimeter when you are crossing 18000. You may be leveling off at what you think is 8000 but is actually higher and therefore ATC isn't clearing you lower.Drew

Paul-Good call on the AIM information, however, what I meant to imply is simply that if there is a problem, the captain is to fly the aircraft safely first and formost, advising ATC of the conflict at hand as soon as practicable. As far as my disagreements, I have see the default ATC wait so late for a proper descent that I had to go missed simply because of being so high still. It had absolutely nothing to do with my being behind the airplane. But then again, it comes down to the fact that we are all discussing the ability of the default ATC to give better altitudes for the given phase of flight in regards to approaches. Someone mentioned that they felt that the default ATC gave the correct clearance for altitudes for the IA but (and here's where I disagree, to be specific) I am constantly vectored lower than the published MDA for the approach. Anyhow, it's not worth an argument.Eric

I looked at the plates for 13/31 at Harrisburg. 3200 is the 25 nm radius MSA. At 14 nm out 2800 or 3000 is the correct altitude to be stepped into depending on your runway assigned with the OM at about 2300.If FS9 has already stepped you down to 8000 then within 25 nm of the OM is should at least take you down to 3400, definitely by 16 nm out (3200 plus 200). If I'm reading the OP correctly, he is not getting the final descent clearance out far enough from the plate altitude entry point. These are straight in ILS approches to 13/ 31.I'm an RC 3.1 user, since the topic was brought up. I have not had any final step down altitude issues. RC samples during stepped descents the MSA specified in regions of surrounding quads of the approach path during vectors. They have an option during preflight setup to overide the final altitude for the ILS merge.The new version four is stated to give much more control over vector leg lengths. Best to read the thread on their support forum here. The release date has not been announced.In one of the 3.3 beta public releases for a very brief time, on go-arounds vector downwinds were too short for all but the smallest GA. This was quickly fixed.