B17 II Update August 10th by Leonard "Viking1" Hjalmarson
	There were a couple of questions that I neglected to ask in my second interview, and additional information has been released. I've culled a few questions from various forums as well as receiving more information via email. Here is an update on B17 II. Sound APIs A3d gives us access to the Aureal's Wavetracing technology via a standalone API that sits on top of either Aureal's Vortex chips or DirectSound. Wavetracing involves "rendering" geometry that affects the properties of individual sounds according to position. Briefing Room Turbo B17 Quote from the interview: "That means that the engine itself loses no power as altitude increases, but the supercharger will need to be cranked down because as the air gets thinner the relative pressure inside it will increase - effectively reducing the power output." Response from Gunfighter III: Sir, as a trained Airframe and Powerplant mechanic (Former job) I beg to differ with the above statement. The TurboSuperchargers installed on piston engines sole purpose in life was to increase the Manifold Absolute Pressure (MAP) to fool the engine into thinking it was at a lower altitude than it was. You are correct in saying that the air is thinner at alt, therefore the compressor section of the TSC "forced" more air (O2) into the cylinders for better fuel consumption and power. In fact the TSC's on the Wright radials pushed the MAP or in this case Upper Deck Pressure over that of Standard day sea-level pressure of 29.92 in hg. The commonly ran at 40 in.hg Upper Deck pressure under maximum continous power ratings. The problem for pilots was if they left the wast gate closed, TSC "full on" as they descended into the more dense air of lower alt, the upper deck pressure would increase to the point of blowing the jugs right off the crankcase. Gunfighter III Response from Andrew Walrond at Wayward: You are of course correct. That answer was wrong, and I should have noticed it (been rather busy and somebody else wrote that answer.) I wrote the simulation code for the Wright-R-1820-97 used in the B17's. Another problem you didn't mention was turbo overspeeding. The turbo is driven by the pressure difference between the exhaust manifold and the outside atmosphere. At high altitudes (From memory 27,000ft for the wright), the turbo controls must be reduced else the pressure difference will be too great, and the turbo will overspeed and disintegrate. Setting the 'power' of an engine in the B17 actually involved juggling the mixture, throttle, turbo and prop pitch controls. There were tables of maximum permissable manifold pressures posted in the cockpit of the B17's (There is probably a screen shot showing this somewhere) Sorry for the confusion. Rest assured I have got this right, in fact I have become something of an expert. Thanks for pulling us up on that one, Gunfighter. Inertial Starters "Are you going to model the Inertial starters? I hope so... " Already done Gunfighter III. Make sure you get those starters good and wound up before you hit the mesh switch, else they cough, spit, splutter and die. In fact, the whole startup sequence is straight out of the B17 flight operations manual. Every switch, knob and control is modelled. Fuel boosters, cowl flaps, intercoolers, oil dilution, the lot. Andrew. Ed. Note: The AAF manual for the B17 F/G has seven pages of text and diagrams devoted to starting procedures! Bump Mapping Another question regards the details of the graphics engine. On Friday morning the programmer in this area, Dominic Robinson, posted the following note on our forum. Click to continue		Sparks on 109 The bump mapping on the terrain in B17 II is performed *entirely* in software, independantly of DirectX ( I wrote the code for it it ). This enables us to do things that current hardware doesn't support, and to run on a wider range of hardware. In particular there is no hardware accelerator anywhere that will render the self shadowing bump maps that are used in B17 II. The fact that a feature is implemented in hardware doesn't necessarily make its visual quality any higher. Indeed, most rendering techniques exist in software for a considerable time before being implemented in hardware. What hardware usually brings is a performance increase, but often to the detriment of flexibility. The Matrox G400 does indeed implement one form of bump mapping Fireball ( and a number of other cards can produce similar effects ). However, B17 II uses bump maps in ways that are not currently supported on any hardware that I am aware of. In particular it renders bump maps that can cast shadows on themselves. All of this is done in software and will run on any recent card with basic texture mapping capabilities. Bump Map Detail: Click for full image. To help clarify what we've been talking about with regard to B17 II's use of bump mapping, the image above was taken from 1,000ft on a Permedia 2 equipped machine with no fancy multi-texturing or bump-mapping hardware - these are software only self-shadowed, anti-aliased bumpmaps. Notice in particular the subtlety of the shading and shadowing on the hilly grass covered area towards the lower left corner of the images. On the evidence of B17 II so far, I'd say that this shadowing is the single most important factor in rendering terrains that will convince the human visual system. B17 II would use hardware acceleration for this feature if it existed - but it doesn't yet :-( (Note: bump mapping is not used on the aircraft themselves.) The textures are multi-layered, and are dynamically painted with bullet and flak holes, oil leaks spreading out behind the engine, fire buring away the skin... It is likely however that this feature will *not* be available unless you have a video card capable of rendering textures from AGP. Everything about our engine is scalable. It'll run on our minimum spec, but the more horse and pixel-pushing power and memory you give it, the better it will look. As Dominic said elsewhere, it's spooky but the game will look better in future than it does now, automatically.... Fighter Sub-Variants "Will there only be the R6 (additional 2x20mm cannons)subvariant for the Bf109? If that will be the case the german fighters will be easy prey for the escorts. The FW190's will need "clean" Bf109 as topcover. Can you give us the possibility to make a variant a simple loadout selection just as rockets or bombs? And will there be G and or K versions of the Bf 109 and A and D versions of the Fw190?" Wisely, Wayward is not committing to sub-versions for the fighters at this stage. This is something they would like to do, but it may not make it into the initial release. Obviously, there is SO MUCH to this simulation and in order to release this fall the line must be drawn somewhere. Given the importance of this point, however, I expect we will see sub-variants in the first add-on. Personally, I am hoping for a new theater of operations and an additional bomber to fly, perhaps the B24? In such an add on it would be a simple matter to add a couple of variants for the main fighter aircraft. Better still, for all you Spit lovers out there, we could even hope to see a flyable Spitfire! Note: For screen shots of the Norden bombsight see the May Preview For B17 links see the August Interview Summary Page
Copyright © 1997 - 2000 COMBATSIM.COM, INC. All Rights Reserved. Last Updated August 10th, 1999