Brian17

@Arrowstar So far so good, haven't had any issues with the Porkchop or the LVD yet. As Salajander noted, I did have the same messages come up occasionally in MA using PR8, but it was either when starting very far from an optimal solution or when trying to fine-tune using very tight bounds. In those cases, switching to a different solver helped. Unfortunately I don't have a log or .mat of it occurring, but if it happens again I'll post it. Edit: Also, those errors were with the prior version, before the issues with parallel pool were fixed.

Cool, good to know about the Pattern Search. It helps to know the strengths and weaknesses of each algorithm so they can be used in the right way. The Porkchop issue does seem strange. It works fine on 1.6.5, but not on the pre-release builds 8 or 9. Everything else in the folder is untouched. Anyway, I'll keep an eye out for anything else. Thanks a lot for all the hard work on this, I know it's not easy!

@Arrowstar I think he mentioned an issue with ground objects. I haven't messed around with them yet so I don't have any info. I tried out the new version, doing a quick check of the different solvers and options. As for the parallel pool, I didn't have any problems at all. Started up easily and worked well. Mission Architect didn't have any issues with any of the solvers, doing a quick test of an orbital altitude change. I did run into 3 issues though, so I linked a .zip with a copy of the log file and the .mat file I used for testing in the LVD. #1 - Starting the program and running the Porkchop plotter with the default settings resulted in a ding and an error in the log file. I have had this error occur every time so far, including with other settings (Kerbin - Eve, etc). This happens immediately after the plot is displayed, as soon as the progress bar for "optimal departure/arrival" appears, but before any info is displayed in the output window. #2 - Loading the .mat file in the LVD and trying to run the optimizer using the IPOPT solver results in an immediate ding and error in the log file. I have not been able to run this solver at all, this error happens every time. #3 - Loading the .mat file for the LVD and using the Pattern Search solver results in the optimizer window starting, running what seems to be just 2 iterations, and then just locking up with no ding or anything. Using the Stop button does nothing, so I am forced to close the window. I noticed an entry in the log file for this as well, so I included it with the others. ZIP file - https://drive.google.com/file/d/1A8_Ar76Ct9hpSIqAAx2l-pkFVeC9P9d4/view?usp=sharing That's all so far. The algorithm issues aren't too bad to work around as I still have FMINCON and NOMAD, but the Porkchop plot is pretty necessary for initial planning and trajectory info. Luckily, I still have 1.6.5 so I can use that for the plotter for now.

I'll check the new version since I'm on Windows and see if I get any issues. That error that @vitorboschi had looks pretty similar to what I had with the debug build once the parallel pool got started. I'll let you know if I find anything unusual.

@Arrowstar I checked the new debug build, and it seems to have started the parallel pool correctly in MA, with the message that it connected to 6 workers in the log file. However, I did get a ding and an error about matrix dimensions when the orbit passed outside of Kerbin's SOI. It also dinged and logged an error when trying to actually run the optimizer. This was with a new mission plan made for this test. I've shared the folder link to the log file and the .mat file used for the test below. Parallel Test - https://drive.google.com/drive/folders/1MEVg6vHwwFl8jp7NSS4PL9YZbDPTlyOA?usp=sharing Hopefully the link works, that's the first time I've linked to anything with Drive. I also noticed that when using the Porkchop plotter in any of the pre-release builds (including the debug versions), I would get an error popup about matrix dimensions. I'm not sure if that's just part of pre-release code or another bug, but that log file includes the matrix error from within MA when testing, just in case. @vitorboschi I've tried playing around with the different solvers a little bit. So far I've found that Interior Point and SQP(?) seem to be gradient solvers, so sometimes they can get themselves into a rut and get stuck, but they tend to make big moves in generally the right direction very early on. With Active Set, it doesn't seem to get stuck as easily as the others, but it tends to make small, incremental progress over many iterations so it can take longer overall. That's just my limited experience though.

@Arrowstar Okay, we definitely got it now. Downloaded the debug version, started up MA, and got this in the log file when trying to start the parallel pool: Looks like the code only allows up to 4 workers, but the default is set to the number of local cores, which in my case is 6 (Core i7-4960X). And I assume that plugin code must override the property of the local cluster and force it to start. I'm glad it was simple enough to figure out so quickly.

@Arrowstar Thanks for the quick reply! As for the .mat file, I've tried loading up a couple of the sample files into MA and get the same message. So I opened up the LVD, which I haven't played around with much, and loaded the sample Two Stage to Orbit. It immediately gave the same popup message that parallel mode failed to start. I created the plugin with that little code, and checked the log file and voila! Starting parallel pool (parpool) using the 'local' profile ... Starting parallel pool (parpool) using the 'local' profile ... connected to 4 workers. As a bonus, now when I go into MA and hit Script -> Execution Settings -> Parallelize, I get "Parallel optimization mode enabled" in the output window. While the optimization for interplanetary transfers seems marginally faster, easier ones like planetary orbits are noticeably quicker, and the program as a whole seems much more responsive.

Thanks for the replies. I know there's a lot going on under the hood and it can take some time, I'm just hoping to get it going most efficiently. @Arrowstar I updated to the 1.6.6 PR8, as the version I originally had was 1.6.5. I tried the option for parallel pool, and again it gave the popup message "Parallel mode start failed. Optimization will run in serial." This time around, I left the menu option checked afterward (instead of toggling it off like before) and tried to optimize. It dinged at me, and this is the text that was in ksptot.log: Starting parallel pool (parpool) using the 'local' profile ... Starting parallel pool (parpool) using the 'local' profile ... Error using parallel.Pool/parfevalOnAll (line 23) A scalar parallel pool is required as the first argument to 'parfevalOnAll'. Use gcp to access the current pool, or parpool to start a new pool. Error in executeOptimProblem (line 14) Error in ma_MissionOptimizerGUI>optimizeMissionButton_Callback (line 585) Error in gui_mainfcn (line 95) Error in ma_MissionOptimizerGUI (line 42) Error in matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)ma_MissionOptimizerGUI('optimizeMissionButton_Callback',hObject,eventdata,guidata(hObject)) Error using uiwait (line 81) Error while evaluating UIControl Callback. Thought you might want a look at that, not sure if it's indicative of something on my system, or something in the program. I can submit a bug report if it looks like it might be in the software.

Greetings! I started using TOT a few days ago and am loving all the functionality in this thing. One issue I've been having is that running optimizations in MA for interplanetary transfers is painfully slow, as is any kind of optimization in the LVD. I found the option in the script menu of MA to parallelize optimizations, but when activated it tries to start the worker pool and fails. So far I've tried disabling Windows Firewall, added Incoming Connection rules to the firewall, tried disabling Malwarebytes real-time protection, and running TOT as administrator. Nothing seems to help. Could anyone point me in the right direction of what to look for? I'll admit, while I have some coding experience I've only got a basic understanding of network protocol. Normally Google is my friend in that regard, but I can't seem to find anything on this.