I feel like if a ballistics app could give you your velocity BC and dope chart off of a one shot zero input say like a 500 yard shot. If we could do that it would be just like magic.
Interesting thoughts Wade. I’m all in on this. I’m thinking it could be a good predictor but as you are thinking it would need way more data inputs. I’m thinking this would be similar to using a single bc value to predict bullet trajectory versus using a drag model over the velocity range to predict trajectory. I’m thinking we would need to know the powder temperature which is probably impossible to know. So we could look at the ambient temp and the chamber temp and time in chamber to predict the powder temperature. One could drill and tap a receiver to install and RTD to get it’s temperature. Ya know it’s pretty dark in this here rabbit hole?? 😂
I'll start off with, I've never used AI and not a math expert but your test and Ai "prediction" was nothing more than a guess. Only giving the information of Average and SD, the average is the only thing of value at "predicting" the next velocity (SD giving expected range) because each new shot is a complete new set of variables/indepent of the others. Each component is/can be extremely similar but is not the actually the same component to have true repeatability. If this was Vegas, me betting the average velocity every time would be closer to the result than Ai over the long term with its guess/"prediction". The SD tells you that 68% of the shots are within that +/- from the average. If your SD is 8, guessing the average would be within 8fps 68% of the time and within 16fps 95% (24fps 99.7%)...Ai giving an answer that missed by more than 16 kinda proves its boken and useless when 95% of the time the average would be closer. Now, if you gave it every single velocity, it may have a better chance of "guessing or predicting" because it can evaluate those outliers that may be truly anomolies...?? And give a better/more true average/SD as far as predicting goes. But again, every new shot is a completely independent outcome and has no connection to the previous. Think roulette table, it is impossible to "predict" what the next number will be based off of previous outcomes. As it's a completely independent spin. Each Ai "prediction" that is not the average is merely a coin flip of if the next shot will be higher or lower than the average. (And then a guess in that range) I do think there may be some use for Ai, but its not going to be any better at predicting the next shot as anyone else can. Unless its predicting the change/value of a variable. Such as...What would be the velocity if I increased the current charge by 3%, something along those lines. Or guving it a history of conditions and results and then being able to calculate the shot/hold for the next shot. (Ballistics solver) But again it can really only use current known calculations. Kind of like a speak to text Quick Load, 4dof, shot probability calculations, etc.
Correct assessment, but gotta start somewhere down this rabbit hole. The information I share with it will increase with each series. Plan is to feed it massive amounts of data over all , as well as start introducing externals . May be a big waste of time, or may actually learn something. Either way ,I’m good with it.
Will you please add helmet mounts for us people running cyclops. I don't have the funds to run a double thermal that needs a bridge. Specifically something for a MH25/RH25 . That would be much appreciated.
I feel like if a ballistics app could give you your velocity BC and dope chart off of a one shot zero input say like a 500 yard shot. If we could do that it would be just like magic.
Interesting concept , I bet you could probably do pretty well with a 100 yard zero , 1 shot 500 velocity confirmation, 1000 yard bc confirmation.
Wow, comment section did not like this one. I thought this was was great and I hope you continue this thought and project!
Interesting thoughts Wade. I’m all in on this. I’m thinking it could be a good predictor but as you are thinking it would need way more data inputs. I’m thinking this would be similar to using a single bc value to predict bullet trajectory versus using a drag model over the velocity range to predict trajectory.
I’m thinking we would need to know the powder temperature which is probably impossible to know. So we could look at the ambient temp and the chamber temp and time in chamber to predict the powder temperature. One could drill and tap a receiver to install and RTD to get it’s temperature.
Ya know it’s pretty dark in this here rabbit hole?? 😂
Thanks for watching as always! Make sure you get subscribed !
I'll start off with, I've never used AI and not a math expert but your test and Ai "prediction" was nothing more than a guess. Only giving the information of Average and SD, the average is the only thing of value at "predicting" the next velocity (SD giving expected range) because each new shot is a complete new set of variables/indepent of the others. Each component is/can be extremely similar but is not the actually the same component to have true repeatability.
If this was Vegas, me betting the average velocity every time would be closer to the result than Ai over the long term with its guess/"prediction".
The SD tells you that 68% of the shots are within that +/- from the average. If your SD is 8, guessing the average would be within 8fps 68% of the time and within 16fps 95% (24fps 99.7%)...Ai giving an answer that missed by more than 16 kinda proves its boken and useless when 95% of the time the average would be closer.
Now, if you gave it every single velocity, it may have a better chance of "guessing or predicting" because it can evaluate those outliers that may be truly anomolies...?? And give a better/more true average/SD as far as predicting goes.
But again, every new shot is a completely independent outcome and has no connection to the previous. Think roulette table, it is impossible to "predict" what the next number will be based off of previous outcomes. As it's a completely independent spin.
Each Ai "prediction" that is not the average is merely a coin flip of if the next shot will be higher or lower than the average. (And then a guess in that range)
I do think there may be some use for Ai, but its not going to be any better at predicting the next shot as anyone else can. Unless its predicting the change/value of a variable. Such as...What would be the velocity if I increased the current charge by 3%, something along those lines. Or guving it a history of conditions and results and then being able to calculate the shot/hold for the next shot. (Ballistics solver)
But again it can really only use current known calculations. Kind of like a speak to text Quick Load, 4dof, shot probability calculations, etc.
Correct assessment, but gotta start somewhere down this rabbit hole. The information I share with it will increase with each series.
Plan is to feed it massive amounts of data over all , as well as start introducing externals . May be a big waste of time, or may actually learn something. Either way ,I’m good with it.
Wade you need to come out with a good coyote seat ground level and comfortable
We have been working on something for couple years. Hopefully getting close to release.
😂 bro if you said what the question was one more time I think I was going to rip my hair out 😂😂😂.
Didn’t want you to forget .
@ 🤣 I can’t remember can you repeat the question
Have you tested 6mm Atip , as in expansion on deer
My mv is 3050 fps
Works great
@txpredatorhunting thanks
Will you please add helmet mounts for us people running cyclops. I don't have the funds to run a double thermal that needs a bridge. Specifically something for a MH25/RH25 . That would be much appreciated.
GIVE US WHAT WE WANT MORE 22 ARC
This guy really thinks the heat comes from powder burn? Like obviously friction
I’m so glad I don’t have the “long range shooting” flavor of the tisem
😂