Magical Instant Bullets | Hacker News Reader

Magical Instant Bullets

militaryrealism.blog

72 pointsbaud14725858 comments

Do bullets actually fall at exactly the same speed if dropped or fired horizontally? In a vacuum, obviously they do. But my intuition says there would be some impact on the vertical air resistance from travelling 500m/s in the horizontal direction. As the bullet pushes through the air, the air pressure above and below it is increased (you can actually see that in the photo in the post). So wouldn't it effectively be falling through denser air than in the dropped case? Also, of course, it would be falling in a horizontal orientation, whereas the dropped bullet would probably realign itself to vertical.

Anyway, none of that invalidates the broader point, and the effects would be comparatively tiny even if present. Just found it interesting to consider.

Fair warning, you are standing on the precipice of a very deep rabbit hole :-). My Dad was a 're-loader' (loaded his own ammunition) and molded his own bullets and had charts and graphs for an amazing set of things. I even built him a chronograph so that he could "know" the speed of a bullet as it left the gun.

Given that a 'zeroed in' sight will account for the drop and so points the barrel up slightly, some of the energy in a shot goes to lifting it "up" before it starts falling down. You might think that would make it take longer, but it's small. The sin of 0.5 degrees is only .00873 so less than 1% of "heading up" before heading down. The more 'up' you point it, the bigger the difference. But as the author points out, the impact of various factors varies. Dad was happy he could put three rounds from his 30-06 into a 3" target circle at 200 yds.

That's a bit of a mistaken model, the bullet never rises above the boreline of the gun, the scope or sights are just angled down relative to the bore line to intersect the bullet trajectory at a particular point further down range. [0] The effect is larger if your scope is higher above the bore axis, because it requires you to point down more on the sight line to have the same zero, or if your zero is particularly close or far, for most rifle rounds and gun configurations 100 yard zeros tend to be all drop (the combination of realistic height over bore and bullet trajectories just work out that way).

[0] This is a pretty accurate representation: https://s-media-cache-ak0.pinimg.com/originals/40/f3/05/40f3...

I think you misread ChuckMcM. Of course it’s the sight that is angled down. What he wrote was accurate.

Is there really a difference?

GP clarifies "the bullet never rises above the boreline of the gun" which tells me he might have thought that Chuck was saying the bullet literally goes up after coming out then comes down, instead of the boreline being pointed up relative to the pointed-down sightline.

I assumed that's why Chuck put "up" in quotes, because it's not actually going up, but seeming to with respect for the sight line.

Another accurate representation, from the article:

https://militaryrealism.blog/wp-content/uploads/2025/03/figu...

If the sight is angled down relative to the barrel, the barrel is angled up relative to the sight. But which is angled "straight"?

The shooter aims with the sight, so if the target is level with the shooter's eye, the sights are level and the barrel is pointed (ever so slightly) up.

This means you are basically correct, in that the boreline is always pointed higher than where the bullet actually is (setting aside any lift the bullet generates, but that is a lot of detail to get into). However, the bullet does often move upwards relative to the ground (for some distance) since people are often aiming almost level to the ground. So in that sense, the trajectories are not "all drop". That is why I think you are being downvoted.

Yeah, should have said all drop relative to the point of aim/sight line I suppose, wrote for an audience more familiar with guns than HN probably is. Things are usually reference relative to the sight because that's what you're aiming with so movement relative to that is how you'd adjust your aim to actually hit the target.

There's not really any mechanism for lift though, the bullet is not an airfoil and is traveling directly into the airflow so there's not going to be any lifting body effects either while it's in stable flight and after that it's tumbling so there might be some then but it will be chaotic and inconsistent and you're just as likely to have it pushed down if the tail is up.

> As the bullet pushes through the air, the air pressure above and below it is increased (you can actually see that in the photo in the post)

That's just a 2d slice/projection of the air density, the shock wave you see is actually conical and symmetric around the whole bullet it's just not possible to capture with schlieren photography.

As far as I know the pressure above and below are basically identical the bullets are symmetric and until they're well subsonic (the transition between super and sub sonic is a lot of what defines the effective accurate range of a round because the transition is chaotic so it's hard to predict and account for it's path after transitioning) they're traveling so fast forward there's effectively no angle of attack to generate a differential on either side of the bullet. If it does generate a negligible amount of lift it's completely dominated by the force of gravity pulling it down.

Yes, of course the shock wave is conical; I just highlighted above and below because they're the only directions likely to influence how quickly the bullet falls.

I don't think the effect would generate lift per se, assuming a horizontally fired bullet (since, again, it's symmetrical). What it does seem to me though is that it would effectively increase the density of air local to the bullet (in all directions except for behind), and so it would essentially be falling through denser air than it would be if just dropped from stationary. That would result in a small but non-zero increase in the time taken to fall the same distance.

Again though, that's just what intuition tells me; I certainly haven't tried to model it or anything.

The shock wave doesn't actually touch the bullet on the sides though so it doesn't provide meaningful lift. Look closely at the schlieren the high pressure region isn't against the sides of the bullet at all.

Air is already quite dense at sea level equivalent to ~194 dB sound pressure so the sonic boom pictured which is 130-160 dB is not a significant increase in the density/pressure of the air.

The Mythbusters did it for real once. I'd call the results as "within error bars, consistent with same speed drop". Lift does not seem to be a significant factor. Shooters would have noticed. They notice a lot of things.

> exactly the same speed

Within a couple hundred yards within the capabilities of normal riflemen, basically yes. In real life though, there are tons of effects. E.g., there's a minor lift/drop in a crosswind because of the pressure differential from the bullet's rotation. E.g., when you have a gun properly sighted in and shoot at a horizontal target, the barrel will be tilted up slightly, the bullet will be tilted up slightly, the rotation will mostly preserve that upward orientation (like a gyroscope), and you'll have a lower descent rate than you might expect for roughly the same reason an airfoil does (and shooting upward or downward can amplify or have the opposite effect). Beyond a few hundred yards, all kinds of things matter.

There is an excellent MythBusters episode where they demonstrate a dropped and fired bullet hitting the ground simultaneously.

The aerodynamic effects you mention are real, but they are subtleties. Unfortunately, I have not been able to find a high quality, open-source ballistic calculator that actually breaks out all those details. Apparently, "Modern Exterior Ballistics: The Launch and Flight Dynamics of Symmetric Projectiles" is the book to read if you want to know more.

Although considerable basic research is funded by the DoD, and considerable open-source has been generated by them, and of course the DoD has done a tremendous amount of research on ballistics, I suppose it makes sense that they have not actively encouraged the publication of an open-source artillery targeting program.

This is very interesting to consider! I think the point about orientation is a great one. Definitely if the drop was substantial, this would be an issue. Let's say you were standing atop a tall tower in a desert and did the same experiment, then maybe you'd see a slight but noticeable difference. The effects of air resistance would only become significant as the vertical velocities increase with acceleration due to gravity.

No discussion of the history of ballistic trajectory simulation in video games is complete without including GORILLA.BAS

Good memories. My brother and I spent a lot of time trying to hit the sun so that it made a face! And oh, the satisfaction of throwing into the wind perfectly so the it boomerangs back around.

::dancing gorilla:: doo-doo-doo-doo doo-doo-doo doooo doooo

My friend rigged nibbles.bas so that he could beat his brother more often.

The banana was indeed observably unrealistic. With speed above a certain level it could ghost through the opponent without harming them because the collision detection only checked subsequent positions, not the whole line segment between them.

A real banana would not do that.

Really fun piece.

> No computer game will ever have the level of sophistication to model shockwaves, fluid dynamic effects, spin9, the Coriolis effect, and the Magnus effect.

I'm no superfan, but who can forget the CoD4 "escape from Chernobyl/Prypyat" sniper sequence?

https://callofduty.fandom.com/wiki/One_Shot,_One_Kill/Transc...

That was a great mission, but I'd probably guess those "Coriolis effect" sequences were scripted one-off events. Doubt they're simulating that kind of stuff on every bullet fired.

Agreed, I certainly wasn't trying to imply this nuance was present throughout the game. Just feels like a glaring omission for the piece to not at least mention existing attempts to translate these subtleties into engaging gameplay.

(though to be fair, it's been at least a decade since I experienced this one and memory is hazy)

Thank you, that's a great example! Definitely an omission ;-)

people love to talk about a game being realistic or not realistic without asking whether more realism = more fun. Usually the answer is no. Doing collision detection on every bullet on every frame is a cost that may or may not be worth it, depending on the overall design of your game. "Games take shortcuts" is really flattening the discussion and acting like the game developers don't know these things that the very smart blogger knows, when in reality, game developers know this stuff and make conscious design decisions. Most games are not designed to be scientific modeling programs, they're entertainment products, governed by questions about entertainment value, not simulation accuracy. Plus in real life you're not going to receive a network packet that says "a bullet whose trajectory will hit you in the skull will reach you in 0.7 seconds" and give cheaters enough time to auto-duck with a script. Even if you can afford to do collision detection on every bullet on every frame, there are many other reasons why it might not be a good idea. Hitscan is a design option, not something that's "right" or "wrong".

Easy solution: Laser gun, problem is now solved up to 100 miles away (0.54 milliseconds).

GPT says that 'a car 100 miles away would be far too small to see clearly, even with the best sniper scope. It would likely ... not be visible at all due to the limitations of optics [25x to 35x] and atmospheric interference.'

Apparently GPT thinks the earth is flat.

I was gonna say this guy's never heard of a hill, but I don't know how far I can see with my drone, but I'm pretty sure it's not 100 miles. I'm not even sure it's more than 20... It is over 8. That's the level of certainty I have without getting scratch paper and a scientific calculator.

Rainier is visible at least 85 miles away.

But could you see a car on it?

i can see the moon from my driveway, too.

I find the fact that bullets hit instantaneously and precisely way more fun!

In the article they say that parabolic math was not available in 1999 computer games, but the bio rifle seems to do it just fine, even at the time ;)

For the ones that want to have fun with the maths, this demo is awesome https://youtube.com/shorts/akEGxm5rcC4

But this video also means, that you still aim at the player head when he is falling straight, you will just hit him a little lower than expected ;)

In-game sniper that you have to aim inches above the head is just not as fun, especially for an extremely fast-paced game like Unreal Tournament.

Why is it not as fun? It's just a different kind of game. There are plenty of tactical shooter or sniper games that go through a lot of effort to model the ballistics (like the Sniper and Sniper Elite series), as well as pretty much every tank and naval game. The Battlefields weren't as fast-paced as UT, but they made great use of ballistics for everything from sniper rifles to RPGs... added more skill to the game.

Even Unreal Tournament itself had some basic ballistic guns, like for the alt fire of the flak cannon and biorifle.

Indeed it did, although the flak cannon brings another set of issues as the preceding article talks about (basically it combines two very different types of weapons system).

And agree, different types of game need different mechanics. Battlefield games were a very different experience; I think the simple mechanics work in UT.

Funny thing is that Unreal had bullet like projectile weapons that don't used hit scans. And many mods for the original Unreal Tournament had this kind of bullet projectile weapons. Of course, this projectile bullets travels straight and don't try to mimic any real life physics effects, beyond to avoid the instant hit or miss mechanic.

The content was great but the looping animated gifs made actually reading the page nauseating.

If you're still using Chrome you can right click the gif and show controls to get a pause. Sadly not supported in Firefox it seems.

There's a bookmarklet out there I just found trying to address this issue myself but I've not tested it so review and use at your own risk.

https://slbkbs.org/jsgif/

Firefox supports stopping gif animations after one loop. Set image.animation_mode to once in about:config.

Better but not quite as good as Chrome's gif player which lets you seek and pause etc.

Thanks for the feedback! Glad you enjoyed. Point taken on the GIF, and I'll bear in mind. It's a bit of a balance; I don't like using too many embedded YouTube videos because that can spam readers with ads.

The GIFs were great for illustration, you're describing something in the time domain so an animated example is useful. The problem was those GIFs. The towers level with the camera movement and planet moving in the background was just way too much motion alongside text I was trying to read.

Maybe try putting more white space between the images and blocks of text or put several images one after another in a group. Then I could see the examples but read blocks of text with no distractions.

> Compared to the rest of the exotic UT arsenal, these seem unremarkable, boring even. They don’t fire bright bolts of plasma or glowing, exploding goo.

There is also the lightning gun. I recall using this on newer variants of facing worlds.

It seems from the other articles as well that they are reviewing the UT99 arsenal, which didn't yet have the lightning gun.

Author clearly never played any simulator games, like Arma. Or he wouldn't ever posted this BS.

"Games which use hitscan mechanics might be taking a shortcut, but it’s one for which can forgive them. The effects aren’t likely to affect realism too much, and we take these shortcuts in real life as well."

Hitscan is ok-ish for arcade brain-rot games like CoD, where max distance is few dozen meters, def. not OK for any other games.

One time in Arma dayz mod I was trapped on a water tower with an unfriendly player above me for 20m keeping me pinned down. My friend in team speak circled around and took them out with a shot from over 1km away. Good shot!

Author clearly mentions Operation Flashpoint and how games like it do things differently.

I never played Arma, to be fair. I always gravitated more toward the "brain-rot" games! But you're right of course, realistic games use more realistic ballistic models (and everything else).

One reason I never got too into the simulator games was that the line between "real" and "game" is always going to be very arbitrary. No simulator will replicate the experience of slogging over a marshy mountain carrying 100 lbs and a GPMG! The Onion said it best: https://youtu.be/yuTkgi7scKo?feature=shared

A real combat simulator would have you polish boots and cleaning latrines for most of them time, and then wake you in the middle of the night with an artillery barrage.

Oh come on. This article is peak midwit "ackshually."

Like all of the game's weapons, how exactly the sniper rifle works is not described. The bullet's diameter is not described. Its ballistic coefficient is, of course, not described. Most pointedly of all, the bullet's velocity is not described.

There are existing guns that can fire projectiles at hypervelocities. These are mostly "two-stage gas guns" that utilize incompressible gasses to accelerate small metal projectiles for NASA experiments -- sometimes to velocities over 32000 feet per second. (10km/s! Roughly 10-11x faster than conventional rifles, but a common velocity for micrometeoroids.) There's absolutely nothing to suggest that similar mechanisms can't be adapted to smaller arms; it's impractical, but it's an engineering problem, not magic.

There are other possible mechanisms, e.g. a conventional weapon with a magnetic rail that assists in the acceleration of the bullet. Science fiction? Sure. As is the game.

So, anyway, assume the UT gun fires a small iridium slug at 10km/s and there's nothing to complain about. Very fast bullets are very flat-shooting, as any .22-250 appreciator can corroborate.

HM17 as well, there's a reason you can't legally use it to hunt, in some states. (If you miss, it just keeps going for miles. Even if you don't miss, it can still just keep going for a mile.)

Are you talking about .17 HMR? That round does have a very flat trajectory, but it’s illegal to hunt deer in some states because it isn’t lethal and won’t penetrate vital organs (just like .22 rimfire.) .17 HMR is primarily used for shooting small game where precision and minimal damage are preferred.

yes, yes, and yes. my mistake on the name.

Love this. "Akshually" is kinda what I'm going for (maybe not midwit), the point is to take something a bit incongruous and pull it apart pedantically.

I definitely think UT's sniper rifle is "coded" as a regular one familiar to us, with its ammo as shells in a cardboard box. I think adapting gas gun mechanics to a man-portable weapons system with interchangable ammo is a bit more complicated than you allow, but indeed it's not magic. Thanks for giving me more ideas to write about.

There's a lot you missed .. although much of it is bollocks.

FWiW my neighbour shoots targets at 5,000 yards - that's well past the confirmed kill sniper head shot record .. it's worth pondering why, how or if someone is telling porkies.

One Quora level nerd force that many toss on about is earth's rotation, aka the Coriolis effect, aka Bart Simpsons Australian bathtub joke, eg: https://www.quora.com/Does-Earth-s-rotation-affect-bullets

The opinion of an ultra long distance shooter is:

  don't let Coriolis trouble you, you have so much more to deal with, Coriolis is there but means very little to a rifleman, Cheers.

~ https://www.youtube.com/watch?v=KP7IKshdiiY

Cutting to the chase, here's one of several well shot multi camera vlogs of:

5023 yards 24" x 24" target ULR shooting https://www.youtube.com/watch?v=7owwTz7Z0OE

If you wander through the channel there's a lot of practical info re: bullets, barrels, hot and cold firing, spotters, scopes, etc.

> Love this. "Akshually" is kinda what I'm going for (maybe not midwit), the point is to take something a bit incongruous and pull it apart pedantically.

You got the pedantic part down, but I don't think that you took a charitable view of the game's systems, and your assumptions were maximally conventionally-minded.

Rather than assume "this is bullshit and doesn't work," I think you'd do better to try to imagine "how could this possibly work?" That question can lead down interesting paths.

It doesn't even take much, in this case. Very fast projectiles and/or scopes that account for variables like bullet drop, wind, etc. Both of these things already exist, in a sense, or are on the cusp of existing.

Reply on news.ycombinator.com