mentat@walt.cc.utexas.edu (Robert Dorsett) (08/22/89)
In article <3226@cbnewsh.ATT.COM> mfp@cbnewsh.ATT.COM (mark.f.pyfer,ho,) writes: >Here's a new thread -- I don't know if it has been discussed in this group >before or not. Are you kidding?! :-) >tried the pee-cee version, and it's mildly entertaining. What do you >think of the mac version? Are there significant bugs, and have there >been multiple releases? What is the latest release? AS always, thanks >in advance. What it boils down to is: (a) The PC 3.0 version is MUCH MUCH superior; (b) the Mac version has, how shall I put it, "problems." The Mac version was apparently written as a prototype 68000-based flight sim- ulator (there are striking similarities between the Mac version and the Amiga and Atari versions). Many features tested on the Mac version were also im- plemented in FS 3.0 for the PC, three years later. There are problems in the user interface. For example, for portability, the authors did their own, home-grown menu and windowing system. Which is fine, except that it's quite slow, and has a lot of annoying inconsistencies. It doesn't follow the Macintosh User Interface Guidelines. It is not possible to load the simulator on a hard disk. The latest revision number is 1.1, released sometime in 1985. A number of aircraft "features" may be set from the menus. Unfortunately, there's no way to save them, so a good deal of time must be sent "customizing" the airplane after every reboot. The simulator attempts to simulate a Cessna 182RG and a Learjet 55. It also has the "war game" mode, familiar to any user of FS1. The same basic cockpit is used for all these modes, with the exception that the Learjet has a different airspeed indicator, the engine changes to a jet engine, and the RPM gauge (yes, one) gives readouts in % RPM rather than as a tach RPM reading. Unfortunately, Sublogic left a great many features in a buggy, poorly thought out, or unusable state. Among these: * The inability to flare the airplane, i.e., it's impossible to have a rate of descent with any positive pitch. One byproduct of this is that the vertical- speed indicator becomes a critical flight instrument. * Lack of aerodynamic stall. A stall is defined, within the program, as the stick being held all the way back. At any airspeed, holding the stick back to the stops will incur a stall. This is not realistic, by any means. Another byproduct of this is that one gets into the habit of flying the airplane by "stick position." * Cruddy artificial horizon. Very poor pitch information. So-so roll in- formation. * Inoperative ADF (it just doesn't work). * VOR/DME range limited to 80 miles, regardless of altitude. * Very choppy database access. For instance, the principal airport, Oakland Intl, is surrounded by several "overlap" areas. One byproduct is that on an ILS approach to RWY 11, one must pass through numerous disk-access intervals, which completely destroys what little realism that exists. * In some areas of the database, it's impossible to pick up radio aids within a cloud. So much for flying blind. * In some areas of the database, proportion is incredibly poor. 3200x60' runways appear larger than the 10,000' runway at JFK. On many runways, the centerlines are skewed, and located off the runway proper. Over Moffet Field, the refuelling squares are at about 15,000'. * The C-182's performance is, how shall we say, a tad optimistic? :-) * The Learjet's performance is pretty poor. The only thing that can be said for that mode is that it lets one move around the database at a higher speed than normal. It doesn't have an indicated airspeed indicator--rather, it uses TAS, which throws out flying by airspeed. * The wargame algorithm is as predictable as it was in FS1 (just get them lined up and blast away :-)). * There are precious few ILS approaches in the database. Each one has dif- ferent operational characteristics. * The "range" of many VORTACS is limited by the scope of that portion of the database--in some cases, only 15 miles. The Oakland area is particularly confusing in this respect. * The vertical-rate-of-descent indicator doesn't necessarily give vertical rate of descent. * The correlation between the true airspeed indicator and the DME suggests that one of them is not working (by my distance calculations, i.e., area flown, the Learjet can do about Mach 2). * The correlation between the altimeter and the vertical-speed indicator also suggests that one is not working properly. Vertical speed rarely results in the anticipated altimeter performance. * In some areas of the database, it's possible to lose 1000-2000' on the altimeter, inexplicably. * I find the black and white shading schemes incredibly annoying. The com- puter doesn't have color, and patterns are a poor substitute. * The autopilot is poorly designed (pull-down menu item; it should be on the control panel). * The landing gear doesn't do anything (up or down, no difference aero- dynamically, except on landing). * The carb heat control is worthless, except for reducing indicated RPM. * It's virtually impossible to run out of fuel. * The engine instruments don't mean anything. * The trim control doesn't do anything, as far as I can tell. * The ILS approaches are too "sensitive." * "Partial panel" completely obscures instruments--I think it would be much more realistic for the instrument to "freeze" or "break." * It doesn't work on a Mac II or SE/30 (it will, however, work on a Plus equipped with some models of 68020 accelerators). There's more, but that's all I can think of off the top of my head. So what's positive about it? If you provide a liberal dose of imagination, you can have some fun with it. The gyro drifts. It's good for practicing VOR maneuvers (although timed turns, due to various problems, often don't work out). It has "clouds" (periods of zero-visibility). It has winds. It lets the user set the sensitivity of the controls. It lets the user turn on and turn off various "realism" set- tings (some of which don't seem to do anything, such as "light burn"). It lets the user set his position (but not in longitude or latitude). It lets the user save and load previously-saved scenarios (but never from the original disk, even though there's plenty of room, and not using Mac prompts-- I mean, for crying out loud, "Save RAM to disk?") If you like "joyriding," you'd probably find Yeager's AFT (particularly on a Mac II: it has color support), or Fokker Triplane, more fun. If you have the aforementioned imagination, you can probably have quite a bit of fun with the simulator. It has three things to contribute to the "real" pilot: 1. The opportunity to practice rudimentary navigation skills. 2. The opportunity to develop an instrument scan. 3. The opportunity to use reasonably realistic CDI's. However, all of these things are at a very low level. I can see the potential for a great deal of negative training. In past discussions, however, other pilots say it's "great" for practicing instrument flying. I cannot agree: there's no ATC to contend with, there's no convenient way to introduce system faults or problems, and the pilot doesn't have to deal with various other workload items (such as a mixture control, prop control, altimeter setting, etc). The few "system" items that exist are mainly there for show. Having said all this, I will add that MS-Flight is the most realistic "serious" simulator available for the Macintosh. Which says a lot about simulators for the Macintosh. :-) I have, however, had a good time finding all of the above bugs. :-) However, I had an opportunity to play with MS-Flight 3.0 on a PS/2 machine for a few hours, and noted few of the above problems in it. The experience really showed what a shallow effort the Mac version is. Robert Dorsett Internet: mentat@walt.cc.utexas.edu UUCP: ...cs.utexas.edu!walt.cc.utexas.edu!mentat