FEBA Posted November 11, 1999 Share Posted November 11, 1999 I am really impressed with CM, have pre-ordered, and am eagerly awaiting the final version. The game is great. I was playing Reisberg as the German and was on turn 3 when my 88 opened up on a Sherman 515 meters down the road. The target was on the left hand side of the road. The round missed. A second Sherman was closer to the 88 on the right side of the road, but out of LOS. While the 88 was reloading, the second Sherman reversed and rolled out onto the road presenting its broadside to the 88. He was apparently positioning to get an HE shot at the 88 (What an idiot!). The 88 kept the original Sherman targeted. I thought "wow, at 515 meters, I might just send a round through both sides of the intervening Sherman and nail the original target. The 88 fired…and hit nothing. Damn I thought, "How could he miss?". Did he fire high?, did he fire low?, did the round go to the left or right? It sure looked like it passed through the intervening Sherman. But there weren't any hit sounds or any detailed armor hit info. I decided to investigate. (Just can’t believe my 88 boys would miss.) Using rewind, I went back before the 88 fired the second time. I had to replay it several times hitting play, then quickly pause. Finally I caught what I wanted, which was the 88 round, just after it had passed the intervening Sherman. I selected the 88. I then went to camera 1. Looking along the line of flight I lined up the round and the 88 selection rectangle. When I was done the selection rectangle was positioned just above the round. This meant that I needed to get lower than camera 1 would allow to truly line up. However, the 88 selection rectangle was overlayed on the intervening Sherman about 2/3 the way back on the hull with the top edge touching the turret. If I had been able to drop lower than cam1, I estimate it would have been overlayed partially onto the track and at least half onto the hull. I then thought: "Of course!, I’ve got vehicle scaling turned on, that makes the tank model too big. It only looks like it passed through." So I cranked scaling down to "reality" level. However, no such luck, the 88 selection rectangle still was positioned in the same place on the tank. Seems to me this should have been a solid hit, unless the Sherman was moving in reverse at some amazing speed. The Sherman hull is about 6.5 meters long and the shell when I lined it up was about 3 "Sherman lengths" beyond the possible intersection point. The line up intersection point was about 1/3 of the Sherman’s hull length from the end and it was moving in reverse. That means it had to move about 2 meters while the round traveled around 20 meters. That means it had to be moving at a speed about a tenth of the muzzle velocity of the 88 to have the shell pass and then intervene between the 88 and the shell. The CM data box shows the 88 muzzle velocity was 773 m/s. That puts the tank moving at 77 meters per second (in reverse)!! For those metric challenged, I think that’s about 166 miles/hour. I can’t explain it. Anyone have any theories or see where my methodology might be wrong??? I really hope that I have made a mistake. At worst I have uncovered a bug during beta which BTS can fix. Maybe they already have. I took a bunch of snapshots from different angles showing the path of flight and the things I have described. I have the game up now (11/11/99 - 12:50 AM). I will try to leave it up all tomorrow so I can take more pictures if needed. (Windows 98 may decide to die in a spurious manner in the interim.) If anyone at BTS or one of the well respected posters here on the forum has any interest in investigating this I will make the files available. They are big. I have a cable modem limited to 128Kbit upstream. Like I said at the beginning, this is a great game! The amount of information I can gather to record to report this possible problem is a testament to how much detail and work have gone into it. Pford: This is why I didn't send my email turn for our "Last Defense" game last night... Link to comment Share on other sites More sharing options...
Guest Big Time Software Posted November 11, 1999 Share Posted November 11, 1999 There is a bit of abstraction here with "hits" and "misses". We don't have the horse power to completely track the path of the round on its way to the target. While things like houses, walls, trees, etc. can be factored in, variables like other vehicles can not. So when a gun goes to target a tank, it only factors in the terrain and the target. In rare circumstances this could matter (you could never kill 2 tanks with one shot though ), but there is nothing we can do about it with today's hardware The other abstraction you might be seeing is that we don't actually track each polygon of the tank in terms of hit/miss determination. That is also impossible, and doesn't add much realism in any case since it is graphical in nature. What happens is all the complex math is determined to figure out if the shell hit your tank or not. Note that ALL of this math has nothing to do with the visual representation of the vehicle, only its mathematical one. If the shot is a miss it is graphically offset from the target. In some cases the graphical offset can make it seem like the shell actually hit, but didn't. In these cases the miss was probably measurable in a meter or two In other words, you most likely missed the Sherman by a hair! Steve Link to comment Share on other sites More sharing options...
FEBA Posted November 11, 1999 Author Share Posted November 11, 1999 Thanks for the quick reply Steve. Don't you guys at BTS ever sleep? (<Sigh>…to have a job I loved that much…) The shot, although close, was a clear miss on the intended target tank. The target line goes from the 88 to it. The two vehicles were about 40-50 meters separated. I didn’t really expect to nail 2 for one (thru both side plates of a Sherman is a lot of armor.), but I can dream, can’t I? I guess I will have to be satisfied with knowing that the intervening tank commander’s virtual pants went from olive drab to brown. As a software engineer with 20 years experience I understand the choices that must be made for performance. (Remember the days when core memory was counted in kbytes?)Maybe someday when 2ghz processors are the norm you guys can model most of what you’d like. I’ll be in line to pick up CM5 when you do. Meanwhile, I’ll just have to make do with the best state-of-the-art WW2 tactical sims out there, CM1-4. (We want Russians! We want Russians!...Opps, wrong thread.) Lance [This message has been edited by FEBA (edited 11-11-99).] Link to comment Share on other sites More sharing options...
Doug Beman Posted November 11, 1999 Share Posted November 11, 1999 Were you playing with Vehicle Image Magnification set to something higher than realistic? If so, the 88 round would fly through the graphical image of the Sherman, but not contact any of the Sherman as represented in the game engine. DjB Link to comment Share on other sites More sharing options...
Guest Big Time Software Posted November 12, 1999 Share Posted November 12, 1999 Lance, I am not a programmer, but I certainly remember playing games that were restricted to 8k of memory!!! Well, that was the first one I owned, but I also fonldy remember using my first "PC", the PET. I think that had 2.5k or memory. My new computer has 128,000 times that level (if I did my math right ). Pretty unreal! Steve Link to comment Share on other sites More sharing options...
FEBA Posted November 12, 1999 Author Share Posted November 12, 1999 Doug, Thanks for responding. In my post what you called "Vehicle Image Magnification" I called "vehicle scaling". <BLOCKQUOTE>quote:</font><HR>I then thought: "Of course!, I’ve got vehicle scaling turned on, that makes the tank model too big. It only looks like it passed through." So I cranked scaling down to "reality" level. However, no such luck, the 88 selection rectangle still was positioned in the same place on the tank.<HR></BLOCKQUOTE> Still looks like it passed through. I think Steve gave a good explanation of what happened. I'm satisfied and I think we can let this thread sink down the pile. [This message has been edited by FEBA (edited 11-11-99).] Link to comment Share on other sites More sharing options...
Doug Beman Posted November 12, 1999 Share Posted November 12, 1999 D'oh! I missed that part. I refuse to let this thread die. DjB Link to comment Share on other sites More sharing options...
Recommended Posts