Pete Dowson

4.872 is released. No different from the 'h' version you already have -- just the number changed. Pete

Sorry. It was only possible in 32-bit versions by hacking into the code. There is no way we are going to try doing that in the 64-bit P3D4. If there is still a problem (and maybe L-M didn't understand when I pointed it out before P3D4 was released) you need to address your complaint to L-M. Pete

Thanks, but for MakeRwys you are probably more a 'user' than a "customer" as such -- MakeRwys is freeware of course. 😉 Good flying! Pete P.S. got too busy with other things today to make a formal release as 4.872. Maybe tomorrow. It's on my growing list! ;-0

Okay. Good. thanks. It was all due to those 21 copies of the Toronto airport. Only the last set of gates, taxiways and runways will be used, not only by MakeRwys, but also by the Sim. So I suggest you check the documentation for that. There's probably a program they supply to select the one for you based on winds and other preferences, and turn the others off. If it were my installation I'd move the redundant 20 copies out into a backup folder leaving just the one. Note that it looks like there are 42 files -- and ADEX (Exclude file) and ADEP (airport definition file) for all 21 layers. With them all being in the same folder, all the Exclusions are processed first (i.e. same facilities deleted 21 times) before they are all re-added 21 times. MakeRwys emulates the Sim to process in the sme order so it gets the same results. If you aren't sure what is going on, I suggest that you ask Fly-Tampa support. I'll make a version 4.872 of the fixed MakeRwys and release it later today. Thanks for going to all the trouble to help fix it. Pete

Thanks. I obviously made a hash out of my so-called "fix", so both F and G were useless tests. Sorry. Here's the real working one, 4.871h (he says hopefully, fingers crossed!) Same end-of file + F4, F5 please. Pete MakeRwys4871h.zip

I attached TWO version of MakeRwys. What I needed first was to run the F version with that scenery layer disabled (not uninstalled -- no need for that. Just not enabled). Hopefully that would give a successful run with complete F4 and F5.csv files and no error message at the end the the TXT file. So that's what I wanted to see. that would prove I found the correct scenery. Then the G version was supposed to be a fully corrected one, which you could test with that layer reenabled. If those results you posted were from thje G version then I've got something wrong in that version, because now there are NO COMMs entries listed at all ANd there's an odd error message. I just need to know what you've done, whether you ran both tests and which test was the last before the results you sent. Pete

Maybe, but these is still something wrong with it redifining all the runways, the gates, the taxiways, the comms frequences 21 times. Even if the last such definition "wins" (overrides all the earlier ones) it's an immnsely sill waste of time and resources loading it and interpreting, and doing this everytime the airport needs to be drwan. I definitely think there's someting wrong. Show that folder to their support and get them to advise. But that IS an error! Or is that from the "f" version, as I asked? And those f4.csv and f5.csv are empty, zero bytes! What are they from? Things seem to have taken a very strange drastic turn! 😞 Pete

I've just been checking the Taxiway and Gate lists for CYYZ. I know it's a big airport (Toronto) but does it really have 409 taxiways and 268 gates/parking places? I know it is perfectly possible, and I hope that MakeRwys has not included duplicates, but it's a lot to check so I just want to know if those numbers are realistic. The taxiways will probably include the short links too. Anyway, after the above test, could you re-enable that layer (without correcting anything in it for now), and try 4.871g, attached. I've put an extra check in for very large numbers of COMMs and also enlarged the tables to cope. Pete MakeRwys4871g.zip

There's only the throttle positions. Negative numbers are reverse. Positive are forward. You'll onlt get a number, between 0 and -n, where n is a percentage of 16384. The percentage for max reverse,, expressed as a fraction, is a separate value in the AIRCRAFT.CFG and is available also in one of the FSUIPC offsets, as a throttle value. 0.25 in the CFG fgives a max or -4096.. so you can work out 60% A throttle in reversed position will still show its value in the offsets as they refer to the lever position. Pete

Wait! I might have found it. In this folder: C:\Lockheed Martin\Prepar3D v4\FlyTampa\FlyTampa - CYYZ You have many BGLs ALL definiting the same airport data, over and over! Many "ADEX ..." or "ADEP4 ..." files. I count about 21 in a row all defining the same runways, same taxiways, same gates and same COMMS frequencies. I'm pretty sure this is what is causing the problem. Duplicate runways etc. are handled okay (the last ones win), but duplicate frequencies accumulate (as this is an actual possibility, though not completely identical normally). Could you disable that layer and do another test? I think there's something wrong with that install. The assorted AFD BGLs may be trying to provide CHOICES of runways to suit the wind etc (closing one or other). I'm pretty sure you shouldn't have them all enabled like that, but choose one. But if you are supposed to (which would be weird with all the taxiways, gates and runways defined 21 times) then let me know and I'll work out a way of dealing with it -- probably eliminating duplicate identical frequencies. Pete

Okay. One more try: 4.871f. same info please. Can you remember when MakeRwys didn't crash for you? If so, can you remember which scenery you'd added since? (All I'm trying to do is identify the BGL with the data which MakeRwys doesn't like. Then you could send it to me and I can repro the problem here and fix it). Pete MakeRwys4871f.zip

Just caught this before retiring. Could you tell me that that last airport was in RUNWAYS.TXT? Very odd that that file finishes early, as that is well done and closed before the COMMS part, the part which is failing. I'm surprised and other files are made. Are you sure those F4 and F5 files are from that last test? Check the times on them please. They are identical to the last ones. Off to bed now. Pete

Okay. This is my last attempt to get to the bottom of this, tonight. It is late here. Version 4.871e should give me enough information, but it's a lot of code changes I've had to make (only to this COMMS area, mind), so it may not work even as well as previous attempts. But if it does, well ... Attached: 4.871e. Same info as before please, the bit from ***ERROR at the end of RUNWAYS.TXT and the f4.csv and f5.csv files. Pete MakeRwys4871e.zip

Thank you. But no end of RUNWAYS.TXT. does that mean it actually worked, or did you just forget? Ah, no, it didn't work. Those two files end at the same place, no different to before. Sorry, I do always need to see the report at the end of the Log. I can then work out what else to add to find out why. Currently i suspect there's something odd about one of the BGLs, and the answer may already lie within the RUNWAYS.TXT file, but with a file so big it is a needle in a haystack. I know the area of code, but it moves each time i make a change. I might need to add many more Logging lines to the RUNWAYS.TXT file, but i want to make them productive. Pete

Ah, no different to 'a'. That's annoying. The result from 'c' is later, which is interesting. And one more to try, please. Then please let me have the f4.csv and f5.csv files. Thanks, Pete MakeRwys4871d.zip

Yes, the tables are in limited offset space, so are of fixed allocated size. You need a SimConnect interface. John and i will be discussing adding these values as offsets, but in FSUIPC5 only, and not immediately. It'll be listed along with other things. Pete

Okay. That narrows it down quite a bit. i'd just like to be sure so i can make the right changes. Please repeat with 4.871b, attached. Same info please. [LATER] As well as testing with 4.871b, attached, I'm having a guess as to what it might be. I ALSO attach 4.871c. Please try that too. I would like to see the end of RUNWAYS.TXT from both, please. Pete MakeRwys4871b.zip MakeRwys4871c.zip

I managed to have a quick look at the files you supplied. They all appear complete except for F4.csv and F5.csv, which are made separately at the end by scanning the log (Runways.txt). They only contain comms radio frequencies, so they may not even be used by whatever you are running which needs MakeRwys. However, I want to fix this crash and still cannot reproduce it, so I have put some extra information into the crash details. These should allow me to narrow down the precise place where the crash occurs (in my code as opposed to a Windows function, which is all I got from the earlier info). I attach this modified version (4.871a). Could you run that please, and let me have the Runways.txt file (or just the crash data at the end, which you could just copy-and-paste here). Pete MakeRwys4871a.zip

Delay isn't a problem. i'm afraid i probably won't be able to look at this till monday. Pete

Okay. So it did generate these files which, though I did list earlier, you somehow didn't find ...? airports.fsm 448 kb f4.csv 118 kb f5.csv 262 kb FstarRC.rws 2.1 Mb g5.csv 4.9 Mb helipads.csv 130 kb r4.csv 4.6 Mb r5.bin 4.1 Mb r5.csv 4.7 Mb runways.csv 2.5 Mb runways.xml 2.5 Mb t5.bin 13.1 Mb t5.csv 19.7 Mb Now that is ALL the files it is supposed to make, and judging by their size, they are likely all complete. I don't know why you couldn't find even one of them!! I did list them for you, to look for! I suspect MakeRwys has done its job, and completed the files, and for some reason is crashing when it is terminating. I'd like to check that all those files are indeed complete, so could you find them (!) and put them all in one ZIP file, and include the current MakeRys_Scenery.cfg, so I can be sure what I'm looking at is correct. I think you can, in the meanwhile, carry on with whatever it is you are running Makerwys for. I'm not sure when I can fix the crash -- without being able to reproduce it it is quite tricky. Pete

Apparently not. That screen pic shows MAKERWYS.EXE selected, in the middle. Where's RUNWAYS.TXT? The MakeRwys EXE is dated 20 MAR there. All the files made by MAKERWYS will be recent, surely? Are there files missing from the top pf the list.out of the window? Tell you what. But this BAT file into the same folde as the MAKERWYS.EXR:: command /c dir >dir.txt Save it as, dir.bat. Make sure it isn't saved as "dir.bat.txt". Then double click it to make it run. You might need to close the window it shows afterwards. It will make a new file called "dir.txt". Give me that one. it will give me a list of all the files in that folder. No. All the Lorby program does is create a Scenery.CFG file which includes AddOn sceneries which haven't been included in the main SCENERY.CFG folder. Else most newer addon sceneries wouldn't be scanned. I need to know exactly whaich files have been made, or even just started, and which not. Unfortunately the Crash data I showed earlier shows the crash in in a Windows utulity DLL, but it will be for a function given to it with bad parameters. I have no way to track that down without being able to reproduce it, hence the need for more information. Pete

Just one extra file? The program produces several at the same time. Try sorting the Explorer display into Time order (just click on the column heading). All the MakeRunways generated files will be together. It would be very useful to see what is really there rather than be drip-fed. Maybe, as you are so adept at taking and attaching pictures, you could sort as suggested and take a picture of the files near Runways.txt in terms of tim. Checking the Runways.TXT file (thanks for providing that) the really useful and most important information is this: ***ERROR C0000005 at 7752BE24: *** Access violation trying to read address 174D5000 *** EAX 145BAC30 EBX 174D1420 ECX 01D16560 EDX 00000000 EDI 145BE7B0 ESI 174D4FA0 So, it crashed, it didn't actually hang. The screen display was probably a remnant to be cleared. This did occur immediately after processing this file: C:\Lockheed Martin\Prepar3D v4\Flightbeam\Flightbeam - KSFOHD\scenery\KSFOHD.bgl So that I can reproduce the problem and so fix it, could you ZIP that file and send it. I don't have the scenery itself, so it isn't of direct use to me other than solving this, and I promise to delete it immediately afterwards. Pete

Excellent. I thought there must be such a program somewhere! Thank you! Pete

It shouldn't need anything, it is defaulted. But you can stop it by specifically telling WideClient not to scan. Perhaps you should show me the WideClient.INI file from that PC, and which PC is it? In your FSUIPC4.INI you have these listed as having been connected at one time or another: [ClientNames] 1=KURT-INSTRUMENT 2=KURT-INST-2 3=USER-606C6C407F 4=KURT-MAINT 5=INST-3 6=TSR-LAB and you already have programmed button presses from numbers 1 and 5 there. So those WideClient installations must have been operating normally. What was all that earlier about keypresses assigned that you didn't understand? Pete

If it is using the normal SimConnect facilities for the text display, and it looks like it is, then a Lua program can capture such text (using event.textmenu). To do this on the P3D PC you'd need P3D4 and the latest version of FSUIPC5. On another PC you'd use WideFS and the Lua would be run by WideClient. You could do that with P3D3 as well, though P3D4 is far superior. To then send it to the speech synthesis part installed into Microsoft windows is another matter. You'd need to research that as I'm unfamiliar with the interfaces used there. There are facilities to play sounds from WAV files in the Lua library, but nothing specific for Text-to-Speech. It might prove useful for the Lua libraries to include a Text to Speech section, but looking briefly at what's involved it is nothing trivial. I would hope there are other prograns or add-ons which could be called to do this, spearately from withing the Lua plug-in itself. But, for Lua support, sorry but this is a separate research project -- maybe something to be looked at for the future. Pete