In response to a suggestion, I overlaid the plotted change-of-speed data as a chart (graph?) over Starlink 9-3: th-cam.com/video/5CEZnVrAhb4/w-d-xo.html
Thank you for sharing, always crazy to remember that the Saturn V got up to 3.8 g just before the S-IC separation. Instead the higher acceleration is relegated to the upper stage. Shows just how dissimilar these two rockets are in their approach with regard to reusability, engine efficiency, staging etc. Cheers!
Are the actual G forces 1G higher? Because on the ground the G meter should be reading 1G, not zero, so when it lifts off, I assume what the vehicle is actually experiencing is 1.5G, not .5 G, and this would carry through the whole flight, correct?
Yes and no! This 'g dial' is just the difference in the displayed speed divided by the time between changes, because that's all the 'data' I have from the video. It's not at all like an accelerometer in a phone or racing car. When the speed is increasing vertically upwards at "1 g indicated" like at 1:40 space-farer-you would experience 2g. Falling tail-first at increasing speed also "1g indicated" like at 6:47 space-farer-you would experience weightlessness. It's complicated by stuff like the orientation of the vessel, and by altitude as the Earth-attraction element decreases the further away ('higher up' for flerfers!) you are, and possibly also by centrifugal forces?
@@Allthegoodhandlesaretakenlmao a bit of to-and-fro this example was by cropping out a tile containing the speed digits, cropping each digit tile out of that, and then comparing to a set of 11 'good' prototype digits. It's code, but with quite a bit of run-check-tweak-run
That display already includes a 'moving average'. It is annoying, isn't it? The method I use produces numbers all over the place, 30 frames per second. I'm considering uploading a comparison video using OCR instead (this is prototype-image-matching). I'm not convinced arbitrary smoothing would satisfy everyone - the 'corners' from motor restarts should probably remain as corners? I am working on improving it...
@@nicepixels8076 they're just 'speed difference'! Not even speed difference, but only the magnitude of speed difference. It's in the description. I'll add a title card next time I do this.
Smoother output would have made a better video! I really want to see if I can improve on the extraction of the actual data - a lot of the 'noise' is incorrect digits being output by my recogniser. There is a 'moving average' in use, but there are corners in the input due to long-lived recognition failures ('5' = '6' errors were common) and thrust changes.
Oh sorry! Didn't see this comment until I came back for some comment-screenshots. I did one for ship reentry - th-cam.com/video/lEs9ubSLZLk/w-d-xo.html
@@hightech9763yeah I think you could be right - there's next to no thrust between adapter jettison and this sighting, it would be following a very similar path?
Why is it showing 0.9gs at 5:30 for example for the booster? Shouldn’t it be 0gs? You do know that it is moving slightly inclined downward as you know its flying engines first and the video shows the angle in which the booster is compared to the ground… shouldn’t that be accountable?
All the dial shows is rate of change of displayed speed. It might be possible to build altitude and rotation into a model to make a display more like an accelerometer. That's a job for someone else!
@@trs4u gotcha! It may also just have been a complicated request! It’s still pretty close and especially when you have that in mind to see that the Booster apparently pulled 6gs even. That is amazing data! :)
@@awesomefeldmanfamily I set out to do exactly that, but it took me so long to get clean enough data out of my image-matching method that I got fed up with it and uploaded short of the target! I should do that next time
@@trs4u I appreciate you spending the time to do it, it must have been so hard how did you do it? You just scanned every second of the video and extracted the data out of the speed and put it into a chart? That seems so hard what kind of software did you even use to do that?
@@awesomefeldmanfamily There's a very long-winded explanation on an old video of mine. It's basically image-comparison, but my old code for doing this wasn't giving great results for this video. This time I generated 'prototype' image tiles using as close as possible a match for the font/size in the video, then took the best match for each frame and digit. There's a few minutes of manual tidy up for aliasing/bright backgrounds breaking the match, and then a final run generates the dials. I use ffmpeg to split/crop the video, and the rest is my own Java code.
The target wasn't splashdown - that's how space vessels landing under parachutes sometimes finish. The target was 7:39 where it stops (0g) with engines still lit at zero metres above the water, to simulate coming to a halt on the chopsticks. The engines-off methane flame and falling into the water was post-success.
because that's what happens when a rocket splashes down? The goal was to softly land on the water using its engines, and it did do that. After engine shutdown, the vehicle was subject to the whims of wind, water currents, and gravity, and so it fell over. When the top impacted, it exploded. It's successful because it did exactly what it was supposed to do. After landing, IF the vehicle survived they would sink it by shooting high velocity armor piercing rounds at it if it couldn't sink itself. It exploded, which was the more likely outcome, which took care of the sinking problem. Did you expect a 270 foot tall tube of air to stay upright in the water or something?
It didn’t explode. At least not before it fell over. What’s it supposed to do after splashdown? Just stand there? I’m sure you’ll find anything to complain about, though. Have fun with that.
@@arielhartung4557 I'm inferring my 'g' from the change of "speed" in the video, so it probably isn't the 'g' you'd feel in your seat in Ship's cockpit, nor as a stowaway in the top of Booster. At liftoff and during landing it's certainly 1g short, it's 'hard to say' accurately anywhere else?
@@thrillchaser9492 It depends on frame of reference. All I've got to go on is 'vehicle speed' in the video. Going straight up unpowered no drag, speed drops at around 1g, but that would 'feel' like weightlessness. Straight down unpowered no drag, speed would rise at 1g, and you'd feel weightless again. It's complicated! The gold standard for how it would 'feel' would probably be load cell data from seat supports?
@@trs4u I'm sorry you are not educated enough to understand gravity. positive gees add weight negative gees make you lighter.. ZERO gees you float.. that is ZERO-G.. The G meter doesn't lie...lmao
@thrillchaser9492 you're thinking of an accelerometer inside a vehicle. Like I said further up the thread, that's not the information I have. The 'g dial' in this video is not from a 'g meter'. It might be possible to combine the orientations of the vehicles and their altitudes with the speeds of the vehicles to get closer to what a 'g meter' would tell us, but that's not what I've done.
Your "G" meter is completely wrong. That sucks because I wish that you actually had the real telemetry showing G forces from the mission. Oh well... :(
the "G" meter values are calculated by measuring the change in the number labeled speed over a period of time, which appears to be per-frame. It's not "wrong" per-se, but it is lacking important data like axis of acceleration, and doesn't take into account gravity or the source of acceleration. It's perfectly fine as a fun thing to look at and go "woah look it was pulling over 5 g's here!" and nothing more.
In response to a suggestion, I overlaid the plotted change-of-speed data as a chart (graph?) over Starlink 9-3: th-cam.com/video/5CEZnVrAhb4/w-d-xo.html
Super Heavy tanking and then casually putting out 5.1 Gs is insane😂 😤
Thank you for sharing, always crazy to remember that the Saturn V got up to 3.8 g just before the S-IC separation. Instead the higher acceleration is relegated to the upper stage. Shows just how dissimilar these two rockets are in their approach with regard to reusability, engine efficiency, staging etc.
Cheers!
3.8 was nothing, Titan II could reach 7 Gs
Still waiting for this impenetrable firmament and flat earth to appear.
exactly lol, they dont even know where it is
Don’t hold your breath
who the hell calls it BFR still?
honestly this is the only valid criticism in this comment section
@@fosstera well i'd also argue that the G meter doesn't account for gravity and is only raw acceleration but I'm too lazy for the math.
Are the actual G forces 1G higher? Because on the ground the G meter should be reading 1G, not zero, so when it lifts off, I assume what the vehicle is actually experiencing is 1.5G, not .5 G, and this would carry through the whole flight, correct?
Yes and no! This 'g dial' is just the difference in the displayed speed divided by the time between changes, because that's all the 'data' I have from the video. It's not at all like an accelerometer in a phone or racing car. When the speed is increasing vertically upwards at "1 g indicated" like at 1:40 space-farer-you would experience 2g. Falling tail-first at increasing speed also "1g indicated" like at 6:47 space-farer-you would experience weightlessness. It's complicated by stuff like the orientation of the vessel, and by altitude as the Earth-attraction element decreases the further away ('higher up' for flerfers!) you are, and possibly also by centrifugal forces?
Essentially, this is from the frame of reference of the earth, rather than the vehicle.
Nice. I’m assuming you did it frame by frame. Did you do it manually or code something?
@@Allthegoodhandlesaretakenlmao a bit of to-and-fro this example was by cropping out a tile containing the speed digits, cropping each digit tile out of that, and then comparing to a set of 11 'good' prototype digits. It's code, but with quite a bit of run-check-tweak-run
“And liftoff of the Big Falcon Rocket”
you probably should've averaged the m/s2 readout because it's really annoying to read flickering numbers
That display already includes a 'moving average'. It is annoying, isn't it? The method I use produces numbers all over the place, 30 frames per second. I'm considering uploading a comparison video using OCR instead (this is prototype-image-matching). I'm not convinced arbitrary smoothing would satisfy everyone - the 'corners' from motor restarts should probably remain as corners? I am working on improving it...
neat!
Hey just a question. Are the dials vectors. Cause I think you are just using speed for acceleration and not vector quantities.
@@nicepixels8076 they're just 'speed difference'! Not even speed difference, but only the magnitude of speed difference. It's in the description. I'll add a title card next time I do this.
Nice work! Can you apply some smoothing to the data so that it doesn't jump around as much?
Smoother output would have made a better video! I really want to see if I can improve on the extraction of the actual data - a lot of the 'noise' is incorrect digits being output by my recogniser. There is a 'moving average' in use, but there are corners in the input due to long-lived recognition failures ('5' = '6' errors were common) and thrust changes.
can you do the same but with ship reentry?
Oh sorry! Didn't see this comment until I came back for some comment-screenshots. I did one for ship reentry - th-cam.com/video/lEs9ubSLZLk/w-d-xo.html
what flew up at 6:32? almost looked like a flying saucer lol
Some frozen condensation maybe? Things move so oddly out of the atmosphere compared to watching them fall or blow away near the ground.
it was the hot stage ring
@@hightech9763yeah I think you could be right - there's next to no thrust between adapter jettison and this sighting, it would be following a very similar path?
@@trs4u yes it is the hot stage ring
@@hightech9763 ok thank you
Why is it showing 0.9gs at 5:30 for example for the booster? Shouldn’t it be 0gs?
You do know that it is moving slightly inclined downward as you know its flying engines first and the video shows the angle in which the booster is compared to the ground… shouldn’t that be accountable?
All the dial shows is rate of change of displayed speed. It might be possible to build altitude and rotation into a model to make a display more like an accelerometer. That's a job for someone else!
@@trs4u gotcha! It may also just have been a complicated request! It’s still pretty close and especially when you have that in mind to see that the Booster apparently pulled 6gs even. That is amazing data! :)
This is totally sick! Can we also get a graph with the time / g force with time stamp in a picture?
@@awesomefeldmanfamily I set out to do exactly that, but it took me so long to get clean enough data out of my image-matching method that I got fed up with it and uploaded short of the target! I should do that next time
@@trs4u I appreciate you spending the time to do it, it must have been so hard how did you do it? You just scanned every second of the video and extracted the data out of the speed and put it into a chart? That seems so hard what kind of software did you even use to do that?
@@awesomefeldmanfamily There's a very long-winded explanation on an old video of mine. It's basically image-comparison, but my old code for doing this wasn't giving great results for this video. This time I generated 'prototype' image tiles using as close as possible a match for the font/size in the video, then took the best match for each frame and digit. There's a few minutes of manual tidy up for aliasing/bright backgrounds breaking the match, and then a final run generates the dials. I use ffmpeg to split/crop the video, and the rest is my own Java code.
Like this? th-cam.com/video/5CEZnVrAhb4/w-d-xo.html
how can it be a successful splash down when it exploded and fell over?
The target wasn't splashdown - that's how space vessels landing under parachutes sometimes finish. The target was 7:39 where it stops (0g) with engines still lit at zero metres above the water, to simulate coming to a halt on the chopsticks. The engines-off methane flame and falling into the water was post-success.
because that's what happens when a rocket splashes down? The goal was to softly land on the water using its engines, and it did do that. After engine shutdown, the vehicle was subject to the whims of wind, water currents, and gravity, and so it fell over. When the top impacted, it exploded. It's successful because it did exactly what it was supposed to do. After landing, IF the vehicle survived they would sink it by shooting high velocity armor piercing rounds at it if it couldn't sink itself. It exploded, which was the more likely outcome, which took care of the sinking problem.
Did you expect a 270 foot tall tube of air to stay upright in the water or something?
It didn’t explode. At least not before it fell over. What’s it supposed to do after splashdown? Just stand there? I’m sure you’ll find anything to complain about, though. Have fun with that.
Shouldn't 0.5g be 1.5g? Weightlessness is -1g?
@@arielhartung4557 I'm inferring my 'g' from the change of "speed" in the video, so it probably isn't the 'g' you'd feel in your seat in Ship's cockpit, nor as a stowaway in the top of Booster. At liftoff and during landing it's certainly 1g short, it's 'hard to say' accurately anywhere else?
WRONG...Zero is weightless... -1 gee would be pulling your face to the dash + gee's mash you into the seat
@@thrillchaser9492 It depends on frame of reference. All I've got to go on is 'vehicle speed' in the video. Going straight up unpowered no drag, speed drops at around 1g, but that would 'feel' like weightlessness. Straight down unpowered no drag, speed would rise at 1g, and you'd feel weightless again. It's complicated! The gold standard for how it would 'feel' would probably be load cell data from seat supports?
@@trs4u I'm sorry you are not educated enough to understand gravity. positive gees add weight negative gees make you lighter.. ZERO gees you float.. that is ZERO-G.. The G meter doesn't lie...lmao
@thrillchaser9492 you're thinking of an accelerometer inside a vehicle. Like I said further up the thread, that's not the information I have. The 'g dial' in this video is not from a 'g meter'. It might be possible to combine the orientations of the vehicles and their altitudes with the speeds of the vehicles to get closer to what a 'g meter' would tell us, but that's not what I've done.
Your "G" meter is completely wrong. That sucks because I wish that you actually had the real telemetry showing G forces from the mission. Oh well... :(
the "G" meter values are calculated by measuring the change in the number labeled speed over a period of time, which appears to be per-frame. It's not "wrong" per-se, but it is lacking important data like axis of acceleration, and doesn't take into account gravity or the source of acceleration. It's perfectly fine as a fun thing to look at and go "woah look it was pulling over 5 g's here!" and nothing more.
That's what I said in the description. If I do this again, would a title card on the video saying that help?