Pete Dowson

I think you'll need to remember what you changed. I'd also appreciate it, please, if you could always make sure you are using the current supported version of FSUIPC, before asking for support. Please check the announcements above. Version 3.50 is three versions and about 6 months out of date now. Version 3.53 hase been current since Jan 1st 2006. However, in this case I doubt that anything has changed in this area. Some actual figures would be useful. Some folks "haze" turns out to be the drawing distance for scenery textures, which isn't actually a visibility option. The graduated visibility option is enabled in your FSUIPC options, but you've not enabled the visibility smoothing, so abrupt changes are still possible between the areas FSUIPC is managing gradually and the FS visibility layer (tops out at various altitudes, could be 8000), and between one local weather area than another. Possibly this is what is happening and it is confusing the issue? Regards Pete

It is still "Dowson", not "Dawson". I think I'd rather have you call me "Pete" than get my name wrong all the time, please. Sensitivity of controls should vary quite a bit from one aircraft to another, and is all about how that aircraft is designed. A jet fighter will be many times more sensitive than a 747. I don't know enough about FS aircraft design to help you make one, but there will be parameters you can place in the Aircraft.CFG file which will change things -- not pitch stability necessarily, but the scalars affecting elevator effectiveness and that sort of thing. There will also be moments of inertia. As far as joysticks are concerned, the main improvment you can make is to use the slopes I provide, with a somewhat flattened response near the centre. Get rid of the message? There's an FS9.CFG file parameter for that I think. Check the FS2004 FAQ. Look in the FS2004 Forum. Ask there if you can't find it. Not sure what you mean by "probably my potmeter is almost dead". Sorry. No idea on that one, sorry. Sounds like you are using a pretty poor aircraft? Look for another, maybe? That's pretty good! Why say "only"? Are the three screens running as one window? If you actually have multiple 3D windows I'm amazed it gets anywhere near as high as that. Two possibilities I can think of -- a memory leak someplace, so your available main memory decreases and you get more disk activity. Or it is something to do with garbage collection on the video cards -- their memory is getting full of old textures and stuff and its taking longer to change them out. However, on both counts I'm guessing. These are definitely the sort of questions you should ask over in the FS2004 forum. Regards, Pete

I should think that would be pretty easy. All the source for the static LIB is provided. You just need to add a do-nothing "DllMain" and export the procedures, i.e. add something like this: extern "C" { __declspec(dllexport) BOOL FSUIPC_Open(DWORD dwFSReq, DWORD *pdwResult); __declspec(dllexport) void FSUIPC_Close(void); __declspec(dllexport) BOOL FSUIPC_Read(DWORD dwOffset, DWORD dwSize, void *pDest, DWORD *pdwResult); __declspec(dllexport) BOOL FSUIPC_ReadSpecial(DWORD dwOffset, DWORD dwSize, void *pDest, DWORD *pdwResult); __declspec(dllexport) BOOL FSUIPC_Write(DWORD dwOffset, DWORD dwSize, void *pSrce, DWORD *pdwResult); __declspec(dllexport) BOOL FSUIPC_Process(DWORD *pdwResult); } BOOL APIENTRY DllMain(HANDLE hModule, DWORD dwReason, LPVOID lpReserved) { return TRUE; } I haven't tried this -- I never thought it worthwhile making the LIB into a DLL as it is so small. It is much more efficient built in, or even have its code merged into the user program. If it cannot handle void pointers, then simply declare the void ones as "char *" or any other sort of pointer, and cast your true data type to that type. If you cannot cast from one type to another you are going to somehow have multiple declarations for the same data (e.g. like C's "unions" which allow one data structure to be utilised in different ways, depending on which union member you reference). Alternatively, alternative procedure declarations, like C++'s overloading might be possible, though that's outside my knowledge I'm afraid. Regards Pete

This is not the "same problem". The one this thread was about was fixed in 3.53 and was simply a registration date thing. Are you registered with a different name and email address to the one you use here, as I cannot find any record of your registration. Of course my records aren't 100% so if you believe you have a legitimate Key, ZIP up your FSUIPC.KEY file and send it to me at petedowson@btconnect.com and I'll check it. The symptoms you are experiencing are exactly those I would expect from bad keys. Regards, Pete

It's Dowson, not Dawson. I'm not sure how it manages a "high speed" and "low speed", as it is only emulating the keys you can use on the Keyboard yourself. Have you tried the keyboard, for comparison? It is most liely relying on FS's own acceleration. Does this happen with default aircraft, or only with the more sophisticated ones? The most likely thing is either that you have enabled the control acceleration fix in FSUIPC's Technical options page when it isn't wanted, or, vice versa, you are using a cockpit which interferes with the acceleration, in which case try the fix in FSUIPC. There really is no way FSUIPC itself interferes one way or the other, but the acceleration fix may be set wrong for what you want. Regards, Pete

Understood. I just thought you might need to find a solution with PaperMaster installed. Ah, okay then. Thanks, Pete

Don't you mean Wideclient 6.51? "DllMain" is simply the entry point to a DLL. All DLLs have them. In this case, as possibly shown by your log, it is presumably a hook installed by that Paper-whatsit program. Your log shows "WINSPOOL.DRV" being loaded. Why do you think that is loaded? There are no print options in WideClient, there's no support whatsoever for spooling. It is something BEFORE this which is being loaded which eventually results in those hooks being installed, along with their supporting DLLs. As I said, it looks like you need to update the PaperMaster software to the latest version, presumably revised to correctly work with WinXP. Regards Pete

Well, I don't have any plans at all to provide any applications, as such. My speciality is in interfacing programs written by clever folks, with more imagination than me, to FS -- to try to provide them with all that they need to realise their aims. There have, for past versions of FS, been programs which do induce failures or emergencies in a posssibly more realistic manner than the default system. For FS2004 I know of FSPassengers, which gets up to all sorts of tricks, and I think there are packages associated with specific aircraft. There may be many others available (sorry, I don't get about so much these days). I'm sure that there will be more packages available in the future which do the sorts of things you propose, and more. Of course, with FSX arriving at the end of the year we may yet be in for surprises from Microsoft themselves in this area! Regards, Pete

Good. Thanks for letting me know. Pete

Okay. I downloaded and installed the Trial version of PaperMaster Pro -- it is currently at Version 7. It doesn't have those two DLLs -- most of its DLLs now begin "J2" or "PM" -- I think the latter are plug-ins for other programs. Running WideClient after installing PaperMaster Pro 7, there's no difference I can see in processor usage. I examined what had been loaded into WideClient's process, and, yes, there were two DLLs from PaperMaster: J2GpfcW.dll J2GRes_Enu.dll I think these must be operating some sort of monitoring of messages on the Network, presumably so it can keep track of what it needs to catalogue. However, certainly in this version, and I am using WinXP, there's no noticeable overhead. I suspect that the version of PaperMaster you have is older than version 7, and was possibly originally written for Win2000, or WinNT, or even maybe Win98? I would think that its poor behaviour on WinXP is due to something which they've now fixed. So, if you want to use that PC with PaperMaster it looks like a version upgrade is in order. Regards, Pete

I assume you must mean Wideclient, because FSUIPC is still only at version 3.53x, and in any case itt runs inside FS which itself soaks up 100% of the processor if it can. They get loaded into WideClient's virtual machine? Uh? Certainly Wideclient itself only loads what it needs, so those DLLs must be hooking into something that it needs. Did you find out what at all? Even stranger --- so those DLLs soak up processor time, but only on WinXP systems? That's good, but it is a very odd puzzle for sure. It would be handy to know what is really loading those DLLs. I've looked for PaperMaster Pro, and it is a very expensive piece of software (199 Euros!). There's a free trial which I've downloaded, so I'll try that and see if I get the same problems. Thanks, Pete

Okay, I attach version 3.539 of FSUIPC, which can apply delays to any axis assigned through its Axis Assignment facilities. The delay is limited to a minimum of 2 x the axis polling interval (which defaults to 10 mSecs) and a maximum of 200 x this interval (i.e. 2 seconds with the default polling interval). Incidentally, the polling interval can be changed by a parameter PollInterval=10 inserted into the main [Axes] section of the INI file (add the section if it isn't there). The units are milliseconds. Delays for axes have to be edited in the INI file. There is no facility to change them or even see them in the option screens. Sorry -- it is just too crowded and my user interface design skills are limited. The format of the axis parameters in the INI files [Axes ...] sections is as follows: For the main axis entry (explanation of values below): n=ja,®delta(/delay) where the parentheses merely show optional parts, and j = joystick # (0 to 18, 16 to 18 being PFC) a = axis (XYZRUV) R is only present when "Raw" mode is selected delta is the delta value (eg 512, or 1 for Raw mode) /delay is the optional delay, in milliseconds When axis controls are assigned (the left part of the options), this is extended by the definition of the controls: n=ja,®delta(/delay),ForD,ctl1,ctl2,ctl3,ctl4 where ForD is an F for "FS control" or D for "Direct to FSUIPC calibration" ctl1 to ctl4 are the control numbers, or zero where unassigned. For Direct mode, these are the calibration indices, 1-4 on Page 1 of calibrations, 5-8 on page 2, etc. Here's an example of an axis assigned to the FSUIPC Spoiler, with a 1 second delay: 0=0Y,256/1000,D,22,0,0,0 If the axis is programmed to send controls based on the axis passing through zones (the right side of the options), there will also be entries for each such assignment, thus: n=ja,UDorB®,low,high,ctl,param where UDorB is U for Up, D for Down or B for Both R optionally specifies Repeat low and high give the axis values for the zone ctl and param are the Control numbers, and Parameter where used. Here's an example for a Gear lever: 1=0Z,256/500 2=0Z,U,6400,16383,66079,0 3=0Z,D,-16384,-13783,66080,0 Note that the delay option (here half a second) still goes on the main axis entry, the one defining the delta (and "Raw" mode if applicable). I hope this is clear. You can edit the INI file whilst FS is running, then simply going to the Axis Assignment options page and clicking the reload button at the bottom of the window. Delays of 200 mSecs or more should be reasonably accurately maintained most of the time, but short ones could vary quite a bit, the smaller you set them, because of the granularity of the polling interval and the sharing of the processor with other things going on in FS. Please let me know how you get on. Regards Pete FSUIPC3539.zip

See if any of the assorted failure flags are being set: search for "Fail" in the programmer's guide Table. There's a batch at 0B64, but they seem to be all panel instrument problems except for the Engine, Another at 3BD6 is similar. Certainly an engine failure should be detectable. The main failure which can easily occur on a hard landing, but doesn't seem to have an indication anywhere, is where the nose wheel strut or even complete gear system collapses. The only indication of that I know of is that the aircraft is flat on the concrete and the gears don't work. I think you can also damage the gear trying to lower it at too high a speed, but again I don't know any indication I can read -- all you find is that the gear system is inoperable when you try to use it. As for body part damage like a broken wing, I would think this all comes under "crash"? Regards, Pete

Sounds like the firewall allows the connection but not the data transfers. Strange. Can't you tell it which programs are friends? I don't have any firewalls between my PCs -- there's only me using them and I trust myself! ;-). I rely on my broadband router to protect me from the wild world outside! Thanks for letting me know. Regards, Pete

I don't have any registration data here. Except for the voluntary phase, before the release of FS2004, I have never dealt with the registration -- that's all in the hands of the sellers. I concentrate on Tech Support and Development. Please check the "sticky" above on "What to do if you lose ...". Regards Pete

Since the SimKits driver interfaces to FSUIPC for its data, and since WideFS is merely an extension of the FSUIPC interface to client computers, and since you appear to need two computers to run two SimKits controllers, thenyes, it seems logical to me. Regards, Pete

There's nothing extra being done with AdvDisplay running. There's no measurable impact on performance whatsoever, and the 15-17 second cycle doesn't seem to relate to anything any of my programs does. One of the commonest cause of stutters with Networks is having the Network set to automatically assign the IP addresses to your PCs. If you set fixed IP addresses manually for each PC (and router if you are using one) then Windows won't periodically go out searching for connections. In any case there is most certainly no need to actually have the AdvDisplay window enabled if you are using ShowText on your other PC. Simply right-click on the Advdisplay window, make sure you have The "multiline" option selected, then also select "hide always". The program will run in the background and simply pass on the data to ShowText. Regards, Pete

If it identifies itself differently to the old one, then none of my GF code will handle it I'm afraid. I have no information on it at all. GoFlight used to send me one of each new unit so I could add support for it, but I've not heard from them for a while. In fact I only noticed there was a new MCP about a week ago when I saw a picture of it in an advert. Regards, Pete

Ahwish I'd known that before Saturday last -- I've only got one PC with an ATI card, and it was a 9800XT, but I upgraded it to an X800 which I purchased cheap from a friend who's gone over to PCI-Express (twin, SLI-connected!). I'm using Catalyst 5.13 drivers (from December last year), and Windows XP. Maybe ATI cards/drivers don't accelerate Windows -- if they don't, they'll be the only make of video card I know of which don't. I'll just go try. I'm using WinXP SP2, by the way. You? [Later] In maximised Windowed mode on my ATI card running 5.13 drivers I'm getting almost the same frame rate as in full screen mode -- if anything, in windowed mode it is measuring very slightly higher, maybe 1 fps or so on average. The only difference which can be noticed at all is the presense of the title bar at the top. Everything else looks and feels identical. I really don't know what's going wrong on your system, but it will be to do with the video card driver settings I should think. If you have stuff like anti-aliasing turned on in FS (Options-Settings-Display-Hardware), turn it off -- go turn that on in the drivers. Same for anisotropic filtering and that sort of stuff. I'm really not the person to give advice about video settings, especially for ATI cards (I' almost 100% nVidia -- the exceptions being one ATI card for testing, and two Matrox Parhelia cards for triple screen connections and displays. Maybe you could pose the question (as to why you have such a discrepancy between full screen and Windowed) over in the FS2004 Forum. Regards, Pete

Not sure how the altitude comes into it -- you should be measuring from the latitude and longitude, the point on the Earth directly below the aircraft. For accurate measurements over long distances you need to use spherical trig and assume great circle routing. Try Googling on "great circle calculation" or similar. For example, I came up immediately with the following reference: http://www.meridianworlddata.com/Distance-Calculation.asp This, like most others you'll find, is based on finding the distance given two known points, so you would have to invert some of the computations to derive one point from the other given distance and bearing. And don't forget to use true north for the bearing! ;-) Regards, Pete

Really Simkits are the folks who should advise on this. Certainly WideFS extends the FSUIPC interface, which Simkit's drivers use, to the other PCs on your network, but you need to be sure that the separate instruments will still work without conflicts even so. I really couldn't advise on that. Really? I'd be surprised if that were the case. Though certainly it is generally easier with glass instruments, surely the information they are processing and displaying are the same? I'd be more concerned over other matters, like separate Altimeter settings (QNH) for each pilot, separate NAV settings, and so on. Stuff like Project Magenta is designed for two pilot operation -- do Simkits actually do separate pilot and copilot controls? Not sure what you mean when you say "what do you mean"? Instruments will work the same on any number of computers, but in a real aircraft the instruments don't necessarily show the same things to both pilot and copilot -- it depends on their own local switch settings. If all you want is duplicate instruments that is easy, if not, you need to check with SimKits. I use Project Magenta for a full 737NG glass cockpit. That works fine. it uses six different computers in the cockpit alone, plus the one driving FS and the external views. Sorry, but you need to check the nitty gritty details with Simkits. Certainly WideFS and FSUIPC can support whatever is needed, but it depends upon their implementation whether you can achieve a truly realistic cockpit environment. Regards, Pete

Good! Well done! Oh, right! Thanks! ;-) Pete

Thanks! Pete

Sorry, I cannot guess from that information. I did not even know any aircraft gauges actually bothered to check FSUIPC versions. Could you do the two tests again, but each time save the FSUIPC.LOG file after you close down FS, then Zip them up and send them to me at petedowson@btconnect.com so I can take a look. It would help if you enabled IPC Read and IPC Write logging (in FSUIPC's options, Logging) before you load the ATR72, so I can see what is happening in detail. It may make the files quite big, so be sure to Zip them, and don't run FS longer than needed. I don't know, but did you check any of the Logs to see if it told you? FSUIPC makes a Log file, as do both WideServer and Wideclient. Please take a look. It is possible that there is an error in your WideServer registration, which could account for both phenomena. If you want me to check that for you, include your FSUIPC.KEY file (also from the FS Modules folder) in the Zip with the Logs. I cannot find any record of your registration for either program here, but my records aren't 100% complete and you may be using a different email address now. Nevertheless I can check the details in your KEY file for you. Regards, Pete

That's very odd. These days, with almost all video cards, there is virtually no difference in performance. In fact on my systems I get slightly higher frame rates in Windowed mode. The last video cards I am aware of that gave the sort of difference you mention were the old 3dfx Voodoo cards, which only accelerated in full screen mode. As a matter of interest, can you tell me the video card model, driver version, FS full screen mode resolution, and Windows resolution? Maybe you have the resolutions wildly different or somehow have 3D acceleration disabled in Windowed mode? Regards, Pete