|
|
Post by Marvel4 on Jul 15, 2012 4:41:26 GMT -5
lol CM901. Foxtrotting hopeless. Did you use the correct values? Also, what values did you use for the SCAR-L and M60? |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 15, 2012 7:54:37 GMT -5
lol CM901. Foxtrotting hopeless. Did you use the correct values? Also, what values did you use for the SCAR-L and M60? CM901: 60 left/60 right/45 down/80 up, 1450 CS, 666 RPM SCAR-L: 25 left/ 50 right/10 down/ 40 up, 1100 CS, 750 RPM M60: 80 left/80 right/ 10 down/60 up, 1500 CS, 600 RPM Those seem to match up with the spreadsheet. Do some of the results seem really off to you or is there some controversy as to the true numbers for those weapons? |
|
|
|
Post by Marvel4 on Jul 15, 2012 8:42:41 GMT -5
Yeah, those are wrong.
The CM901's correct pitch is -60 and 65 (the picture in the guide is moved up).
The SCAR-L has a minimum pitch of -5 (no mistake in the guide, I don't know why this is wrong in the spreadsheet).
The M60's recoil in the guide is wrong. I tested it in-game and unlike the MG36, the recoil always goes up, like in CoD4 and Black Ops, so its minimum pitch must be 10, not -10.
I guess you have the correct centerspeed for the handguns (750 for the USP, P99 and Desert Eagle and 1000 for the Five-seven).
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 15, 2012 9:20:33 GMT -5
Yeah, those are wrong. The CM901's correct pitch is -60 and 65 (the picture in the guide is moved up). The SCAR-L has a minimum pitch of -5 (no mistake in the guide, I don't know why this is wrong in the spreadsheet). The M60's recoil in the guide is wrong. I tested it in-game and unlike the MG36, the recoil always goes up, like in CoD4 and Black Ops, so its minimum pitch must be 10, not -10. I guess you have the correct centerspeed for the handguns (750 for the USP, P99 and Desert Eagle and 1000 for the Five-seven). Alright, thanks again for your help. I'll just patch up those results using the correct numbers later today. You should be the official Den Kirson Editor-in-chief  . How the hell did you pick up on the +/- 5 ViewKick error on the CM901 and SCAR-L from my results?! And yes, those match with my CenterSpeed values for the handguns. |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 16, 2012 0:44:18 GMT -5
The eighth draft has been patched to reflect corrections made possible by Marvel4 with respect to the CM901, SCAR-L and M60E4.
Oh, and I'll upload the latest version of my code tomorrow. (Does anyone actually look at the code? I really only upload it for transparency's sake, but I'm curious to know.)
|
|
|
|
Post by Marvel4 on Jul 16, 2012 2:55:58 GMT -5
I'm surprised the M60 is still better than the MG36, even with the correct values.
Also interesting: CM901 is better than the AK-47 on the biased chart.
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 27, 2012 0:37:43 GMT -5
So just to remind everyone that this is still an ongoing project, I'm dropping in to give an update.
I'm still working on incorporating GunKick into my code. I think the hardest part is done and I'm now trying to debug what I've got. (Specifically, I'm trying to figure out why GunKick contributes to recoil even when I set all the Kick values to 0... hmm...). I can't really say how long the debugging will take, but I can't see it taking longer than a couple of days, by which time I will need the GunKick numbers.
I know there are people who frequent these forums with strings to pull. And I've tried getting on Twitter to catch Mark Rubin but he is, understandably, very busy these days. But there got to be some way to get those numbers!
|
|
|
|
Post by Marvel4 on Jul 27, 2012 7:51:27 GMT -5
I noticed you forgot the Black Ops sniper rifles. Also, will you add a CoD4 chart? Would be interesting to compare. Thank you for all your work on this!  I know there are people who frequent these forums with strings to pull. And I've tried getting on Twitter to catch Mark Rubin but he is, understandably, very busy these days. But there got to be some way to get those numbers! So XboxAhoy didn't answer? Did you try to reach him on Twitter? |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 27, 2012 9:02:54 GMT -5
I noticed you forgot the Black Ops sniper rifles. Also, will you add a CoD4 chart? Would be interesting to compare. Thank you for all your work on this! I know there are people who frequent these forums with strings to pull. And I've tried getting on Twitter to catch Mark Rubin but he is, understandably, very busy these days. But there got to be some way to get those numbers! So XboxAhoy didn't answer? Did you try to reach him on Twitter? No, I never tried getting through to him on Twitter. He doesn't seem to use it all that much, except to announce a new video. Not sure how I missed the BO sniper rifles. I might patch that up later today. A CoD4 chart would be nice, since we know the (obviously negligible) GunKick values for that game. I'd be a good test for the code anyway. |
|
eLantern
True Bro
"Oh, cruel fate, to be thusly boned! Ask not for whom the bone bones, it bones for thee!" - Bender
Posts: 10,761
|
Post by eLantern on Jul 27, 2012 9:53:13 GMT -5
This is the best thread I've seen in a long time. Awesome work probaddie and I sure hope you can come upon the GunKick numbers... just reading everything in this thread is making me want to pop my MW3 back in and play.
EDIT: Den, this deserves a sticky
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 27, 2012 11:48:43 GMT -5
This is the best thread I've seen in a long time. Awesome work probaddie and I sure hope you can come upon the GunKick numbers... just reading everything in this thread is making me want to pop my MW3 back in and play. Thanks for the compliment. What can I say, I try . And it's done. My code can now take in GunKick variables and spit out statistics. I've done all the QA testing and it behaves exactly as it should. All I need is to get the GunKick numbers, input them into my weapon list -- this may take a while -- and punch out the ninth draft. |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 28, 2012 13:31:11 GMT -5
I've patched up the BO results in the eighth draft to include the sniper rifles. I also need to mention something about the code I modified to accommodate GunKick. In the "How does GunKick work " thread, we discovered that GunKick allows the irons to cross both the x and y axes during a single kick. As it turns out, this actually makes modelling the behavior of the gun more difficult, not less. This is because I would have to account for what is theoretically an infinite number of cross-axial jumps. Of course that can't happen in a finite amount of time (fireTime), but you can't know just from the fireTime alone how often that could happen. Also, it would ruin the linear relationship between the initial velocity (kick) of the gun and were it lands in a single kick. This would make the math just about gosh darn golly gee whiz impossible to deal with. So, I had to make a simplification: I assumed that the gun will stick to the axes just like it does with ViewKick. This assumption is only a problem if the decay values are very low, the acceleration (recentering) very high and the kick values very big. But I think we would all notice our guns swinging back and forth while firing them if it were a significant problem. I'll know if I can get away with this simplification if/when I get the GunKick numbers and plug 'em into the super-duper recoil chart generator. eLantern: Thanks for the nod. I think I said before that it was just as well that this wasn't stickied due to it being an unfinished work. But if it really is worthy of a sticky I would take it now, just for the fact that more people would see it and, perhaps, be willing to help me get the GunKick numbers from either Mark, Stuart, or whomever is able to help. |
|
|
|
Post by Marvel4 on Jul 28, 2012 14:12:32 GMT -5
EDIT: Den, this deserves a sticky I agree. And since we now know psijaka's recoil plots aren't accurate... |
|
eLantern
True Bro
"Oh, cruel fate, to be thusly boned! Ask not for whom the bone bones, it bones for thee!" - Bender
Posts: 10,761
|
Post by eLantern on Jul 29, 2012 8:46:33 GMT -5
[quote author=probaddie board=general thread=4577 post=91555 time=1343500271I also need to mention something about the code I modified to accommodate GunKick. In the "How does GunKick work " thread, we discovered that GunKick allows the irons to cross both the x and y axes during a single kick. As it turns out, this actually makes modelling the behavior of the gun more difficult, not less. This is because I would have to account for what is theoretically an infinite number of cross-axial jumps. Of course that can't happen in a finite amount of time (fireTime), but you can't know just from the fireTime alone how often that could happen. Also, it would ruin the linear relationship between the initial velocity (kick) of the gun and were it lands in a single kick. This would make the math just about gosh darn golly gee whiz impossible to deal with.
So, I had to make a simplification: I assumed that the gun will stick to the axes just like it does with ViewKick. This assumption is only a problem if the decay values are very low, the acceleration (recentering) very high and the kick values very big. But I think we would all notice our guns swinging back and forth while firing them if it were a significant problem. I'll know if I can get away with this simplification if/when I get the GunKick numbers and plug 'em into the super-duper recoil chart generator.[/quote]
Did you test this out with CoD4's gun numbers?
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Jul 30, 2012 11:08:51 GMT -5
Did you test this out with CoD4's gun numbers? No, not yet. But I suppose that is all I can do at this point, so I'll do it sometime this week. |
|
|
|
Post by demoncdw on Jul 31, 2012 16:12:49 GMT -5
I did some similar things before, and maybe considered some different things.
1. Player knows the average recoil value, and constantly move their crosshair by a reversed velocity.
2. I suppose reaction time is 0.3s, player will aim the gun at the point 0.3s before.
But now I know my previous work will never be accurate because of gunkick.
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 2, 2012 12:44:39 GMT -5
I did some similar things before, and maybe considered some different things. 1. Player knows the average recoil value, and constantly move their crosshair by a reversed velocity. 2. I suppose reaction time is 0.3s, player will aim the gun at the point 0.3s before. But now I know my previous work will never be accurate because of gunkick. I think I remember seeing your work. That was the little app that produced out recoil plots and had a feature to offset the original point of aim, right? It's unfortunate that we don't have those GunKick, numbers, I know . I don't foresee my work doing anything like that, though. By the way, I'm glad someone suggested using the CoD4 numbers as a "dry run". I found a bug in the program, where the code wasn't properly handling the switch between the two decay values. I'm patching it up right now. Expect the CoD4 results sometime tomorrow, if not sooner. |
|
eLantern
True Bro
"Oh, cruel fate, to be thusly boned! Ask not for whom the bone bones, it bones for thee!" - Bender
Posts: 10,761
|
Post by eLantern on Aug 2, 2012 13:02:37 GMT -5
cool.
BTW, your spreadsheet needs updating for the recent hotfix for buffing some weapons.
|
|
|
|
Post by Marvel4 on Aug 4, 2012 6:23:17 GMT -5
The M16 and MP5 were buffed to 895 RPM and the PM9's recoil was reduced by 22%. Maybe keep the old measurements for comparison.
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 4, 2012 9:58:16 GMT -5
The M16 and MP5 were buffed to 895 RPM and the PM9's recoil was reduced by 22%. Maybe keep the old measurements for comparison. Sure, I'll make a ninth draft with all the changes sometime tomorrow. About the CoD4 results: I've run into a bigger bug than I previously thought. Basically, the equations got so complicated that I had to start running a numerical method to try to calculate a value I need for the GunKick probability distributions. But for some reason, it sometimes takes 10-15 seconds to calculate this number (per weapon). I'm trying to figure out why this is happening, so I can't say when those results will be out. |
|
|
|
Post by demoncdw on Aug 4, 2012 19:53:45 GMT -5
I did some similar things before, and maybe considered some different things. 1. Player knows the average recoil value, and constantly move their crosshair by a reversed velocity. 2. I suppose reaction time is 0.3s, player will aim the gun at the point 0.3s before. But now I know my previous work will never be accurate because of gunkick. I think I remember seeing your work. That was the little app that produced out recoil plots and had a feature to offset the original point of aim, right? It's unfortunate that we don't have those GunKick, numbers, I know . I don't foresee my work doing anything like that, though. By the way, I'm glad someone suggested using the CoD4 numbers as a "dry run". I found a bug in the program, where the code wasn't properly handling the switch between the two decay values. I'm patching it up right now. Expect the CoD4 results sometime tomorrow, if not sooner. That's not my work, hasn't released yet. Do you remember what's that tool's name or on which site? |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 5, 2012 19:19:18 GMT -5
I think I remember seeing your work. That was the little app that produced out recoil plots and had a feature to offset the original point of aim, right? It's unfortunate that we don't have those GunKick, numbers, I know . I don't foresee my work doing anything like that, though. By the way, I'm glad someone suggested using the CoD4 numbers as a "dry run". I found a bug in the program, where the code wasn't properly handling the switch between the two decay values. I'm patching it up right now. Expect the CoD4 results sometime tomorrow, if not sooner. That's not my work, hasn't released yet. Do you remember what's that tool's name or on which site? It was created by a fellow Bro. I'm sure it was introduced in a stickied thread but now that I want to find it I can't. Murphy's Law, eh? I think I fixed up the bug in the program, but I've been working on it on a different computer that doesn't have the CoD4 weaponlist I created to run the program on. So I may be able to, when I get home, run the program and have those results up within a couple of hours. |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 5, 2012 23:31:28 GMT -5
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 5, 2012 23:38:41 GMT -5
You have my permission... to die! Sorry, fixing... Edit: Fixed. |
|
mmacola
True Bro
the brazilian guy
Posts: 1,995
|
Post by mmacola on Aug 6, 2012 0:04:32 GMT -5
You should cap the barrett. Nobody shoots it at 625rpm (unless they are noobs). You should make the semi autos the time it takes to recenter them.
|
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 6, 2012 0:15:59 GMT -5
You should cap the barrett. Nobody shoots it at 625rpm (unless they are noobs). You should make the semi autos the time it takes to recenter them. Then the recoil would be zero  . I wanted to capture a realistic worst-case scenario with those weapons, but no one seemed to think 625 was an unreasonable cap (until now). |
|
mmacola
True Bro
the brazilian guy
Posts: 1,995
|
Post by mmacola on Aug 6, 2012 0:19:32 GMT -5
But then you could change the formula somehow to time to shoot again or something. Barrett is around 145rpm, not sure. If the M21 can recenter in 250 (don't know, only an example) it is much better. Just change it so the barrett isn't in the thousands |
|
|
|
Post by Marvel4 on Aug 6, 2012 0:28:24 GMT -5
The MP44 should be in the Assault Rifle category. What rate of fire did you use for the manual action weapons (CoD4 and Black Ops)? |
|
probaddie
True Bro
You're triggering my intelligence
Posts: 11,043
|
Post by probaddie on Aug 6, 2012 0:30:16 GMT -5
But then you could change the formula somehow to time to shoot again or something. Barrett is around 145rpm, not sure. If the M21 can recenter in 250 (don't know, only an example) it is much better. Just change it so the barrett isn't in the thousands Well, you're asking for something completely different now (not that I don't like the idea) but this isn't something this code is designed for. Maybe I'll do something like that while trying to get the GunKick numbers. |
|
mmacola
True Bro
the brazilian guy
Posts: 1,995
|
Post by mmacola on Aug 6, 2012 0:31:21 GMT -5
Ok  It's only the semi-autos that need it anyway. |
|
. How the hell did you pick up on the +/- 5 ViewKick error on the CM901 and SCAR-L from my results?!
. I wanted to capture a realistic worst-case scenario with those weapons, but no one seemed to think 625 was an unreasonable cap (until now).
It's only the semi-autos that need it anyway.