Conspicuous by Their Absence

Do you mean Hubbabubba's SCASM Tutorial? I would not worry too much if you are confused there. I kinda figured SCASM would lose most of the folks. For that matter, I would not be surprised if my AF99 Tutorial lost a lot of folks also, but the difference is that AF99 is a much simpler tool all around. It is just another Integrated Development Environment. The big difference that I see is that the SCASM TOOL would lose a lot of folks while the AF99 Tool probably would not. The Artistic part with both methods is the REAL value that Hubbabubba and I add and that part is difficult to teach.

BTW, here are a couple detail shots of the Richard Osborne Me109E.
First is a fairly clear view of the Pilot's Face.
Next is an aft view from inside the Cockpit.

- Ivan.
 
Here are a couple detail shots of the Thunderbolts:

First is the corrected Cockpit View Aft for the Bubbletop. Note that the Tip of the Stabiliser is still visible but the rest is not.

Next is the Infamous Flap Bleed from Below. Note the White Triangle inboard of the Inner Gear Doors? Shameful!!!
I don't think I can easily fix it though.

Last is a view from the Cockpit for the Razorback with the expected AF99 Cockpit Bleeds.

- Ivan.
 
Ivan said:
Do you mean Hubbabubba's SCASM Tutorial? I would not worry too much if you are confused there. I kinda figured SCASM would lose most of the folks. For that matter, I would not be surprised if my AF99 Tutorial lost a lot of folks also, but the difference is that AF99 is a much simpler tool all around. It is just another Integrated Development Environment. The big difference that I see is that the SCASM TOOL would lose a lot of folks while the AF99 Tool probably would not. The Artistic part with both methods is the REAL value that Hubbabubba and I add and that part is difficult to teach.

BTW, here are a couple detail shots of the Richard Osborne Me109E.
First is a fairly clear view of the Pilot's Face.
Next is an aft view from inside the Cockpit.

- Ivan.
Click to expand...

Actually, guys, AF99 is a GUI for SCASM. Otherwise, I would not be able to "pry open" their MDL binary model file. The problem, as I see it, with AF99 is that it uses a very odd method to order the showing of groups depending of your POV. It is particularly insane when it comes to internal view. One would have expected that the "Canopy/High wing group" would have been drawn last but, nooo sir, it is the nose or the tail, depending on the position of the internal POV in relation to the model "center" point.

Note that SCASM is not a GUI (General User Interface). MDLDisAs is the GUI that let me use SCASM. Many that have used SCASM stayed pretty close to AF99 method of doing things, simply making minute adjustments of a cosmetic nature. Others have modified their MDL to be compatible with CFS1 stock a/c.

My method is totally original (as far as I know...:kilroy:) in assembling parts together. This will eventually be explained in my thread, but one must learn to crawl before walking and, then, running!:running:

Anyone who has questions regarding SCASM is free to ask them in my tutorial thread. Otherwise, how will I know if you understand what I'm doing?
 
IDEs, GUIs, and Compilers

Hi Hubbabubba,

I don't believe that AF99 really counts as an interface for SCASM. I agree that SCASM is the output, but what you are saying is sorta like saying some C compilers are a GUI for a Macro Assembler because there is an intermediate stage in that language. I have done quite a lot of programming in C on numerous platforms for which I did NOT know the native Assembly language. I never needed to or even wanted to.

The other thing is that there is a serious difference between clicking on a screen to create vertices for your aircraft pieces and being able to edit SCASM code to create a vertex.

There is no question that AF99 does quite a lot of stupid stuff in the translation to SCASM code. I believe that is because it was basically a quick & dirty approach to get something basic out the door without a great deal of consideration for what the inherent limitations would be. I have done this kinda thing before for compilers / translators to invent a language that allows the novice to traverse a date repository without having to know the structure of the data. (XML access to Java and Batch files) It took me just over a week to create the syntax for the language before even starting to implement it. I suspect this kind approach was not done for AF99.

I believe that part of it was also due to the expectations of the time. Who'da thunk we would be trying to create all kinds of complicated shtuff with a tool intended to create FS98 class models.

The goofy "Group Glue" issue you described with Canopy / High Wing being last is because AF99 tries to make its display order choices all before displaying a single group. In other words, you can be Ahead or Behind the CoG. There isn't a conditional check to see if you are Ahead or Behind the division between Body,Main and Tail. What is disappointing is that this kind of thing is partially implemented via some of the Template Parts which I didn't cover in my Tutorial.

Other silly things were a choice of Tags to use such as the "Speed Below" stuff to animate control surfaces. A little thought would have gone places here. With the tags in the range just under 200 mph, They can possibly conflict with the actual operating range of the aircraft. Considering that the design subjects are probably going to be aircraft, it would have made more sense to push the range of numbers up to 3000 mph or at least somewhere supersonic so that it could not conflict with common aircraft speeds.

The choice of 30 Components as a limit and 30 Structures does not appear to have been a good one. Just about every project could use a few more Components but I can't think of one where 30 Structures ever was a limitation.

There are a bunch of other ones that are a bit too esoteric to get into here.

FWIW, I believe that Alain Breton and Pennti Kurkinen both used AF99 for creation of basic pieces of their aeroplanes and used SCASM as a final assembly method which is what I understand your method to be. I plan on doing that for the more recent twin engine projects I have going.

I try to commit as little as possible to the SCASM final processing for my projects because there is always the chance that I will attempt to rebuild from AF99. The P47D27 came very close to getting reprocessed because I found that I had not textured the piece of armour plate behind the cockpit. I didn't do it this time because I looked over the existing texture files and could not find a good place to put in this texture. I know there are problems with this aircraft that I don't really plan on fixing and I have other things that I would rather spend time on. I also got a bit lazy and wanted some instant gratification.

- Ivan.
 
Actually, I was talking about the even easier tutorial that Hubba turned me on to about fixing add ons
to work properly as AI..........

Dave
 
SCASM AI aircraft?

What exactly did that do? I am curious even though I don't really do much with AI aircraft.

- Ivan.
 
Hi No Dice,

Thanks. I know about those. I was debating on using that method whenever I finish up the Stuka. It is probably the only way to get a bomb load to appear appropriately.

- Ivan.
 
Please do Ivan,
Use it on all your projects if you can, No one likes seeing a plane take off with out any undercarraige.

Dave
 
I can tell you I have never seen any of my aircraft take off without landing gear.....

There are a few other things I would like to resolve first though....

- Ivan.
 
Ivan, I added one of yours as an AI to a mission, had to change to the original 47, all addons sit on the runway
without wheels or props ect, unless thay have gone thru the scasm process.

Dave
 
Actually it isn't just the "SCASM Process". It is a matter of putting the 3D drawing code from an AF99 model into a CFS wrapper. It lets the 3D model hook into the variables used by CFS rather than those shared with FS98. If I had chosen to do that with the Macchi C205, I would not have struggled with having duplicated Spinners to allow for animation.

- Ivan.
 
I don't believe that AF99 really counts as an interface for SCASM.
Click to expand...
Are you forgetting the "Produce" button? I'm not saying that AF99 is only an assembler (compiler), but the the AFA files have to be made into a BGL-esque code and inserted in the MDL 610, and I was told a long time ago that Laemming Wheeler opted for SCASM, under license, to do the job. Even if it is not SCASM, it is certainly something very close to it. The "native Assembly language" here is contained in the MDL, and it is not what AF99 is compiling, only the visual BGL-like code is transferred. That's why AF99 contains a "af99simple.mdl", a sort of MDL 610 template for future creations.

What I was trying to explain is that AF99 uses a "fit for all" method of ordering parts, no matter what a specific project demands. As you say, "expectations of the time" were far less demanding than what we expect now. They were in fact based on Flight Shop AF 5, AF99 being only an improvement on precision and maximum number of parts permitted.

As for the "speed below" stuff, I could go on and on and on... Suffice to say that it is a pain in the gludius maximus,if you pardon my kitchen-Latin. An example though; what do you do when you want to turn on a tail position light that happens to be on the rudder? My Taifun had one...:banghead:

FWIW, I believe that Alain Breton and Pennti Kurkinen both used AF99 for creation of basic pieces of their aeroplanes and used SCASM as a final assembly method which is what I understand your method to be.
Click to expand...
Can't say for Pennti, but Alain's method differs from mine in that he basically kept AF99 structures, especially when it comes to showing order. That's what I was saying by "original" method. The idea was already "churning :running: in my head" when we met in Montréal and I remember trying to explain it. The St-Leu church and the jeep were "test beds" for this idea. I never saw any SCX file resembling what I'm doing, and I hope to eventually transmit that method through my tutorial, if I can live that long! LOL!

Actually it isn't just the "SCASM Process". It is a matter of putting the 3D drawing code from an AF99 model into a CFS wrapper. It lets the 3D model hook into the variables used by CFS rather than those shared with FS98. If I had chosen to do that with the Macchi C205, I would not have struggled with having duplicated Spinners to allow for animation.

- Ivan.
Click to expand...

This was kind of a misnomer, like Ivan says. SCASM is only used to modify the internal variables within the "BGL-like" part. The Frankenstein transfer of that code into a CFS1 MDL file is done with an HEX editor. I must admit that it is not always easy to follow, even Corrado had hiccups along the way, but it is the only way I know of getting rid of those legless silly floating AI aircraft.

If Microsoft would be kind enough to release a CFS1 MDL compiler freeware (one can dream, no?).:engel016:
 
Hi Hubbabubba,

Actually I tried to redo the "Group Glue" thing for my F6F-5 Hellcat a few years ago. The idea was that one CANNOT make all decisions about viewing angle before displaying anything. My solution as I explained via email way back was to make each decision as needed. The resulting code was much shorter than the code generated by AF99 but was fairly labour intensive and I had issues with the aircraft displaying at all as an opponent in Quick Combat.

If I understand your method, it seems VERY labour intensive. No doubt it CAN produce a better looking model, but as I mentioned in other discussions, I spend an awful lot of time refining shapes and not being able to see the result visually would not work for me. I don't know how it is possible to use just the coordinates to see how gradual a line curves.

That gets back into the suggestion you were making about breaking the Ju 87B Stuka into two AF99 assemblies to be combined in SCASM. There is no doubt that it can be broken down even further. At what point do we call it good enough instead of breaking things down even more? A Propeller really should be two Components because the Tips and Blades are different colours.....

Personally, I would rather see you put those plans into action and build your Harvard before completing a Tutorial. I have LOTS of data for that plane if you want them.

- Ivan.
 
Hey, Here's a thought: How about building a "Tora Tora Tora" Zero that was converted from a SNJ / AT-6??? Now THAT would be cool!

- Ivan.
 
Ivan said:
Hey, Here's a thought: How about building a "Tora Tora Tora" Zero that was converted from a SNJ / AT-6??? Now THAT would be cool!

- Ivan.
Click to expand...

Unless someone wants to do a remake using CFS1, I don't see the point. I'd rather do a copy of the original than a copy of a copy.:kilroy:

I do remember seeing a French site where the SNJ-T6 was showcased. A former Japanese pilot was fooled for a few minutes, thinking it was the real thing. I think it was Jean Salis.
 
A Copy of the Original would work pretty well also, but I was figuring that you were obsessed with the AT-6 / SNJ platform and I am sure no one has done a conversion to a Japanese Zero before. As for an original, how about the P-64 that was the fighter version of the AT-6?

If you are seriously thinking of doing the A6M Zero, I have a bunch of really good references. I wish I had those before I started this project years ago. At the moment, I am gathering data for a checklist for the A6M2 Model 21 that is shown in these screenshots.

- Ivan.
 
Grumman Hellcats

Just Finished SCASMing these two Ironworks Birds. These are way old projects.

The F6F-5 was my Experimental Aeroplane for "Group Glue".

As an example, AF99 SCASM code will only allow the following sequences: Aft-Middle-Fore if you are Ahead of the CoG or Fore-Middle-Aft if you are BEHIND the CoG. (Remember that the LAST thing displayed will be seen over everything else.) This is why the Tail Pieces always display through the Cockpit Aft Wall in most single engine AF99 planes. (The Cockpit Viewpoint is behind the CoG.)

My idea here was to make each decision separately: If you are ahead of the Nose / Body,Main division, you would see the Aft/Middle-Fore. The idea was to make it possible to show something like Aft-Fore-Middle so the Tail Pieces would not bleed through the Cockpit. It worked pretty well in Free Flight, but didn't work at all for AI controlled stuff in Quick Combat. The aircraft would try to spawn (noted by a hesitation on my old computer) and fail. It would retry about every five to ten seconds.

I suppose I should really try to figure out what exactly I broke when I removed a LOT of code, but just adding the Cockpit Aft Wall as the last thing displayed seems to work well enough that I will not pursue the Group Glue further for now.

The 8.3 file format needed for SCASM textures caused me a little grief over the last couple days.

Next step for these two aeroplanes is to write a Checklist. Test flights show that although these two birds turn well, they are not particularly agile because of some rather mediocre Roll Rates.

- Ivan.
 
It worked pretty well in Free Flight, but didn't work at all for AI controlled stuff in Quick Combat. The aircraft would try to spawn (noted by a hesitation on my old computer) and fail. It would retry about every five to ten seconds.

I suppose I should really try to figure out what exactly I broke when I removed a LOT of code, but just adding the Cockpit Aft Wall as the last thing displayed seems to work well enough that I will not pursue the Group Glue further for now.
Click to expand...

It must be something very mundane, like a Label jumping to the wrong line. If I remember, this is not the first time you have that "AI not showing" issue.
 
Hi Hubbabubba,

Actually this IS the first time I have the AI not showing issue. This is the ONLY Group Glue aircraft I have done to date and it obviously did not work quite as I expected. Considering the amount of labour involved to get there, I believe the minimal modification idea with just repeating the Aft Wall of the Cockpit is a much better approach. There is almost no visual difference between the two methods.

It may seem like a repeat because I never pursued the issue when I first found the problem. Conceptually, the Group Glue changes would make for simpler SCASM coding and "lighter" 3D model, but any changes would need to be done via SCASM at that point. With the repeated Cockpit parts, the SCASM changes are fairly quick. I figure I can complete the changes in about an hour, so the reference copy of the project can remain in AF99 where the edits can be visual.

- Ivan.
 
You must log in or register to reply here.
Back
Top