Hi,

 

There is apparently a computational error with the RTA page in that it can't handle times after 2400 hrs.

 

e.g.

 

I had an oceanic entry waypoint that was predicted to be reached at exactly 0000. I attempted to tell the aircraft to arrive at this point at 0005, however, it assumed 0005 was IN THE PAST, and so set the required speed to M0.851 (fastest permissible). I then also got the message RTA ERROR.

 

I tried to enter 2405 to tell it that it is in the future, but that was met with INCORRECT FORMAT.

 

I entered 2359 and it then predicted M0.810.

 

It would seem therefore that RTA is useless for any time after midnight, as it can't handle the fact it is the next day??

 

Best regards,

Robin.

Hi Robin!

 

I had an oceanic entry waypoint that was predicted to be reached at exactly 0000. I attempted to tell the aircraft to arrive at this point at 0005

 

It could just be a mathematical (computer) boundary issue on range. That starts acting up on that very special value 0000.

 

Might not be as bad as you suggest, i.e. that RTA is useless if your target time is past midnight:

 

Let's say the predicted time was 2357, entering 0002 as the RTA might work.

 

Programming this kind of logic when the upper and lower boundaries is essential the same (0000) is a bi*ch. And in essence the exact value of 0000 could be off limits in such a scenario. Depending on the number format used to store time, integer or floating point (together with other rules present in a discrete computer). The underlying computation to determine if a value is inside or outside a range so the code can branch correctly (example: is the value outside of range? then it is the future) is tricky when the value is 0.

 

Reading what I just wrote, I realize the technical overkill. Sorry...

 

And I hardly believe it is worth the effort to check if my claim is correct: that only the exact time of 0000 will cause trouble.

 

Important point: This is not sloppy programming, this is one programming issue among thousands of issues that codewriters needs to consider. And personally I am glad that PMDG did prioritize other more important issues in this magnificent simulation.

 

This one fell through the cracks, it happens...

 

 

Morten

 

Morten

Let's say the predicted time was 2357, entering 0002 as the RTA might work.

No. I was quite explicit - the problem occurs for any time after midnight.

 

Important point: This is not sloppy programming, this is one programming issue among thousands of issues that codewriters needs to consider. And personally I am glad that PMDG did prioritize other more important issues in this magnificent simulation.

I appreciate it may be a hard bug to find, but unless it is how the real thing behaves, it wants fixing.

 

As for selective bug-fixing - there is no excuse. There is quite a serious CTD bug I found that exists in both FSX and FS9, and I find it hard to believe it was missed during testing.

 

Best regards,

Robin.

Hi Robin

 

Hmm! Interesting, but just to be clear:

 

No. I was quite explicit - the problem occurs for any time after midnight.

 

So if if you have a waypoint prediction that happens before midnight, and you enter a RTA time after midnight, the FMC incorrectly treats this time as the past? No matter what value after midnight you plug in?

 

I entered 2359 and it then predicted M0.810.

Was this as expected? Or was this also wrong? Just making sure I understand you correctly.

 

I appreciate it may be a hard bug to find, but unless it is how the real thing behaves, it wants fixing.

No argument there...

 

As for selective bug-fixing - there is no excuse.

That is a term I never heard before. But again, no argument. All bugs known should be fixed. But I'm curious, the MD-11 has been on the market for quite some time now, have you seen any posts on the subject?

 

There is quite a serious CTD bug I found that exists in both FSX and FS9, and I find it hard to believe it was missed during testing.

Please, elaborate.

So if if you have a waypoint prediction that happens before midnight, and you enter a RTA time after midnight, the FMC incorrectly treats this time as the past? No matter what value after midnight you plug in?

Correct. 0000 onwards is treated as being in the past.

 

Was this as expected?

Yes - expected and correct.

 

There is quite a serious CTD bug I found that exists in both FSX and FS9, and I find it hard to believe it was missed during testing.

Please, elaborate.

Check my other thread on CTD. I also posted the crash report there. Same offset every time, suggesting for certain a bug that can be fixed.

 

Best regards,

Robin.

Thanx for the reply Robin!

 

Now I'm really curious, I'll try to test these parameters myself when I get home.

 

This one fell through the cracks, it happens...

 

This statement was made in light of the fact that we probably will not see another SP release for the md-11. A claim that was actually stated in your CTD thread (If it is the right one?, I did a post in that thread about an hour ago. Another interesting discovery from you).

 

I was not trying to be glib. And if your claim turns out to be true, I stand corrected. And it is sloppy programming.

 

Regards

 

Morten

And if your claim turns out to be true, I stand corrected.

I guess we need to continue this in the other thread, but it will be interesting if you can replicate the crash.

 

This statement was made in light of the fact that we probably will not see another SP release for the md-11.

I know the product is a few years old now (apparently 2008!), but we should not be subjected to poorer customer service because it was perceived to be a less popular product (who said this anyway?). IMHO at some point PMDG should at least try and fix some of these issues (at a minimum, those that can cause CTDs).

 

Best regards,

Robin.