AK47 recoil

[free]

I've noticed that AK47 shoots are a bit more random than they should so I tried to do some testing and below I try to describe what was the outcome of the testing. If you happen to get diffirent results do not hesitate to post them but please remember to provide as much info as possible.

I've just tested shooting 30 bullets at a time btw.

http://img687.imageshack.us/img687/8021/ak4716.png
http://img135.imageshack.us/img135/2529/ak47csp.png

In 1.6 I've noticed that AK47 shoots between 4 and 6 but mainly 5 bullets in a more or less a straight line rotated by an angle between 0 and let's say 5 degrees. After each bullet there is a slight chance the weapon will jump on the X axis a bit to one of the sides and if it does the next bullets shot in line are kept in line with the previous "random" bullet. The "random" bullet is usually the 5th, but rarely it is 2nd, 4th or 6th.

After the first phase there is a slight jump on the Y axis (as seen on the pics between the phases) and then we see a rise on Y axis. The jump however is not obligatory in 1.6 on the contrary to CSP where it happened in each test. Bullets on the X axis then keep to the one side (left or right, that is random). It is most likely for the side to switch in middle of the phase. Rarely it happens that the swich occurs more times or not at all. After all these swiches the bullets' hit pretty much the same place on the Y axis.

In CSP the recoil feels pretty much the same but the random factor at each shot seems larger. I usually don't get a straight line in the first phase and it is most likelly to shot 3 bullets out of line. The bullets randomised in the first phase are randomised on both X axis and Y axis. As we can predict that each bullet hits a higher spot we never know if it will go to the left or to the right in the first phase as these bullets don't seem to form a line most of the time.

The seccond phase feels pretty much the same but also gets a bit too random factor. Much more random on X than on Y though. Also the side switching during the seccond phase may happen up to 6 times.

Again, this is just the outcome of 20 minutes of testing. If you happen to find some more info or if you or the devs find it useless then also please say so. I'd really like to know what kind of info the dev team has gathered and what is needed in order to polish the recoil as much as possible.

Cheers!

P.S.
Each weapon I've tested seems to shoot faster. While in CSP i shoot whole 30 bullets the 1.6 ak is still taking it's 28th shot.

[pRopaaNS]

Don`t know if this analisis is correct, but bullet decals in cs1.6 can`t be relied with. For example, in your sceen bullets decals may show this way, but for others playing on that server bullets decals may show a bit differently, don`t know witch one of them should be taken as the right. To prove that, there are glitches like awp shooting out 2 bullet decals and spraying bullet decals otherwhere than it really hits.

Nature of cs1.6 recoil -
*) first 4 bullets goes out straight vertically, each one a bit further away from first and with 5th bullet they start spraying also horizontally;
*) when beign hit, moving at average speed, recoil increases;
*) when ducking, recoil decreases, even if you duck after you started spraying. If you duck with 4th shot bullet while spraying, then recoil will decrease and it won`t continue to spray horizontally.
If those 3 things works in csp, then it`s fine with me. But I don`t think there is point to try to copy evrything bullet by bullet, since bullet decails aren`t 100% accurate.

[free]

Yes, there is a diffirence between client and server hit in 1.6 and it differs on clients. Obviously client prediction is to blame. I tested it on a listen server so i guess in theory it should have the best accurate hit decals possible as there is no packet loss and delay between client and server.

All in all your short description is afaik a summary of what i wrote above. The problem imo is that these 3 bullets in AK are not always working and each can be affected diffirently by the horisontal spray. That renders the weapon too random even on short bursts.

[yuri]

Each weapon I've tested seems to shoot faster. While in CSP i shoot whole 30 bullets the 1.6 ak is still taking it's 28th shot.ROFs should be exactly the same as in 1.6... there were a lot of tweaking, so it's strange :(

btw thx for the research

[free]

My clanmate made a video to compare recoils. The recoils seem the same on the video (except USP which is way off) but the ROF is diffirent on AK. There is a slight chance the video has wrong timing but he's been making game videos (mostly cs/css) for some time now and i wouldn't suspect such an error of him.

Either way here's the outcome:

http://www.youtube.com/watch?v=obS4mlYd3a8

Thanks for the video sicler!

[yuri]

nice

[free]

Also try to notice the 3 first bullets with AK in CSP on the vid. They seem to go up yet on 1.6 they seem to go pretty much on the same spot.

[yuri]

Also try to notice the 3 first bullets in CSP on the vid. They seem to go up yet on 1.6 they seem to go pretty much on the same spot.Yop, 2-3 bullets go where you aim in 1.6 - in CSP there goes only the 1st bullet. Thats why you can't do much 1.6 typical HSs.

[free]

Also as I described before in 1.6 it swiches left-right side only once or twice but in CSP it swiches sides like mad up to 6 times. That makes it almost impossible to aim all 30 bullets in one spot as you need to adjust horizontally too fast.

[pRopaaNS]

gj for noticing all this.. :)

[tiger]

Nice research. I've told Ben to make a little tweak on the AK the other day, I don't know (yet) how it'll be though.

Btw, could you please use j0lt's test maps (http://cspromod.com/community/threads/3909-j0lt-s-Test-Maps) to compare the recoil? I think using the targets is a better way to do it.

Also, the ROF differences with 1.6 are most likely tickrate related, unfortunately.

[free]

Tickrate related, well probably but the test was performed on a tick100 listen-server. Also faster ROF could the cause of a bigger recoil.

I'll do some more research probably tommorow on j0lts maps.

[yuri]

I also dont believe that 3 bullets diff is caused by tick dispersion. 1 maybe, but 3!?

[Boosen]

reg isnt bad just on the ak

[Improved]

Yop, 2-3 bullets go where you aim in 1.6 - in CSP there goes only the 1st bullet. Thats why you can't do much 1.6 typical HSs.

This is what I miss the most in csp, specially at long range, really hope this gets sorted.

[Geraldsmod]

Yes I also think the AK feels broken. Especially at long range like Improved says. It honestly feels like the bullets are going straight through their head and never regging, but I know it's the recoil because reg is very good in CSP.
Please CSP team, take the advice given in this thread and fix the AK!

[Dare365]

The reason the recoil feels weird when spraying is more because of the feedback you get from shooting them. When you control your spray in 1.6 you look at where the bullets are going and adjust your crosshair but in CSP, it is difficult to see where your bullets are going because of the amount of debris that come off the walls when you shoot them and the amount of blood that comes out from a body shot. Just take a look at the video Free posted, there's simply too many particles to accurately see where your bullets are going in the heat of a battle or in farther distances.

The blood is another thing that needs adjusting while we're on the subject. The blood particle from shooting someone needs to be much smaller. Right now, even body shots look like gooshes because of the big blood splatter and it makes it feel like you're doing much more damage than you actually are.

EDIT: Also, if you look at the picture OP posted, bullet holes in 1.6 are bigger, more distinct and easier to track.

[Improved]

You know you can turn off all of those things to mention right? And not, it has nothing to do with that, btw, its how the weapon behaves, that is what we are on about.

[Dare365]

You know you can turn off all of those things to mention right? And not, it has nothing to do with that, btw, its how the weapon behaves, that is what we are on about.

All what things? You can turn off the debris that comes off the walls, thats it. The blood particles are still too big and bullet holes are still too small. Also, the sound feedback for hitting someone is either too low or non existent. Theres supposed to be an impact sound that is different from hitting the wall that tells you that you're hitting someone. I'm not talking about the "Ugh" sounds the model makes when you hit them, I'm talking about the sound of your bullets hitting someone. You usually notice it when you're walling someone.

Is the recoil different? Yeah, a little bit. Its not the big reason most people are spraying, missing and or even hitting but not doing as much damage as they think they are doing. Again, watch Free's video. The pattern isn't the same but its pretty damn close and the pattern isn't the reason why it feels completely off, its the feedback you get when you're shooting that makes it feel off.

Also, yes you can turn debris off, but the bigger issue is that the CS 1.6 setting doesn't act the same as in CS 1.6. They shouldn't obscure your view as much as they do.

[JinXR]

I brought this issue up a fair while ago many times... Glad you had the time to research it. Well done.

[Improved]

wall of text...

What is the name of this thread? AK47 recoil, meaning how the weapon behaves, not particles, not sound not this and that.

Tell me this, do you hear or even see all that stuff you are mentioning at long range? No, you cannot. Again, if you want to bring other stuff up, make a seperate thread for it.

Weapon behaviour in this thread, thats it. Stick to the topic.

[Dare365]

What is the name of this thread? AK47 recoil, meaning how the weapon behaves, not particles, not sound not this and that.

YES, this thread is about the AK47 recoil and I'm telling you its mostly an illusion. There is no point in having the devs run into a brick wall over and over again, analyzing spray patterns for something that is pretty close to the original when that is not the issue. Making another thread is a waste of space and clutters the forums, its relevant to this topic because it solves the same problem.


Tell me this, do you hear or even see all that stuff you are mentioning at long range? No, you cannot. Again, if you want to bring other stuff up, make a seperate thread for it.

Weapon behaviour in this thread, thats it. Stick to the topic.

The irony of you calling me stupid is painful. If the admins hadn't already warned me about ripping people apart the gloves would be off, champ. I will say this and I don't mean it as an insult but instead because I think you should know that you're probably more disruptive than helpful. The thing that terrifies me about Pro Mod is there are people in the community like yourself who comment and presumably have an impression on the dev team but you have very little of an idea of what you are talking about. Speaking loudly and insulting people who disagree with you doesn't make you correct, in fact it just makes you look more wrong. I highly doubt it impresses anyone either.

[Improved]

Ill start here...

"If the admins hadn't already warned me about ripping people apart the gloves would be off, champ."

Hahahaha, ROFL! I can't do anything but laugh at this talking monkey!

And... "YES, this thread is about the AK47 recoil and I'm telling you ITS ALL AN ILLUSION".

I guess its all an illusion even when the devs say themself that the recoil is not yet 100% spot on, but what do they know, according to you, ITS ALL AN ILLUSION!!

Dude, srsly just do yourself a favour, stfu and sit on the bench and just observe while people get stuff done, you seem to be the one that belong in a different game forum.

"The gloves would be off"

Hahahaahaha...

People like me? Disruptors of CSP! ROFL! Dude, check my threads out, ive been here since the "blue" days always trying to help. Just stfu already you clown.

--- Update ---

And I couldnt care less how the way I express my disgust against morons like you come out or if it impress people, I couldnt care less. So stop the good sameritan act, save it for your mommy will ya, she might believe ya.

[Dare365]

Ill start here...

"If the admins hadn't already warned me about ripping people apart the gloves would be off, champ."

Hahahaha, ROFL! I can't do anything but laugh at this talking monkey!

And... "YES, this thread is about the AK47 recoil and I'm telling you ITS ALL AN ILLUSION".

I guess its all an illusion even when the devs say themself that the recoil is not yet 100% spot on, but what do they know, according to you, ITS ALL AN ILLUSION!!

Dude, srsly just do yourself a favour, stfu and sit on the bench and just observe while people get stuff done, you seem to be the one that belong in a different game forum.

"The gloves would be off"

Hahahaahaha...

People like me? Disruptors of CSP! ROFL! Dude, check my threads out, ive been here since the "blue" days always trying to help. Just stfu already you clown.

--- Update ---

And I couldnt care less how the way I express my disgust against morons like you come out or if it impress people, I couldnt care less. So stop the good sameritan act, save it for your mommy will ya, she might believe ya.

Word. Well, I said what I wanted to say 3 posts ago. I think I'm done talking to someone who is worried about his image in the Counter-Strike Pro Mod forums. We are all super impressed with how long you've been here because its clearly a sign of your intelligence and the amount of respect we should all pay you. Now start talking about the topic again and stop making an ass out of yourself, you can make a thread in the off topic forums if you want to do that but I'm sure you knew that already since you were here since the "blue" days.

[Improved]

Word. Well, I said what I wanted to say 3 posts ago. I think I'm done talking to someone who is worried about his image in the Counter-Strike Pro Mod forums. We are all super impressed with how long you've been here because its clearly a sign of your intelligence and the amount of respect we should all pay you. Now start talking about the topic again and stop making an ass out of yourself, you can make a thread in the off topic forums if you want to do that but I'm sure you knew that already since you were here since the "blue" days.

Hahaha, talk about turning it around you good sameritan you, were are your gloves? Or did you forget you were making threats already? Now you wanna play it nice and wanna be a good sport? Wanna make me like I was the one starting shit, rofl. Ye it's all about my e-penis and all about what people think of me, all should bow down to me and give me respect.

You sound like those people on Fox news, like Bill O'Reilly, when they are in deep shit, the turn stuff around making it look like they are all innocent. Doesnt work on me "champ".

[thomsonic]

Improved, srsly, just stop. Dare365 is right about everything he said in his first+second post whether or not it wasn't the recoil of the ak47 he talked about specifically. It is important stuff that still has relevance in this topic and the problems are indeed something the devs should try and fix.

[Improved]

Improved, srsly, just stop. Dare365 is right about everything he said in his first+second post whether or not it wasn't the recoil of the ak47 he talked about specifically. It is important stuff that still has relevance in this topic and the problems are indeed something the devs should try and fix.

Sure, it is idd important to adress those things too and its good observation from his part, but imo in another thread. In a thread that doesnt mix the two things together, because recoil patterns and particles+sound are two different things.

On another note, my reaction was to the "ITS ALL AN ILLUSION" about the recoil and us pestering the devs with bullshit stuff like he put it, like its all in our heads. Well, then the devs are wrong too, since they mention it also. Now, if he doesnt know exactly what we are on about in this thread and the exact cs1.6 ak47 recoil we are after (not particles, not sound), then wtf? He does not only talk down on people but also mix the two things together. That to me is pure ignorance when hes on about other stuff.

Again, the stuff he brought up is good observation and they need to be adressed too, just dont go out and say like we invent problems and pester the devs when all we want is to help out. He shouldnt talk shit when he's not on about the same thing like the rest of us. It doesnt matter who it is, I'd adress it no matter what. One thing I cant stand is douches like him not even being on the ball and trying to makes others look bad even when others are trying to help. All that is left for people like that is the good old backhand bitchslap.

[n69ky]

okay now stop this :P give your feedback. Don't comment on others opinions/feedback if you don't have to add something or support it. The Devs will "see" whats true and what's just an "illusion"... -.-

gz retired nky xD

[DanLegenD]

Anyone else unable to land a shot in the middle of crosshair long range with crouching 1 shots?

[heyron618]

yea Dan, there's been a few times I crouched, carefully steadied my aim over a long-range target and it took just too many single bursts before my target went down (sometimes the target was even stationary).

As for Dare and Improve's stupid argument (no offense): you both brought up valid pts within the countless insults. Anyway, here's my opinion of free's OP vs. what Dare originally brought up:

Free's test was strictly for the recoil of the AK47, which I also believe has some inconsistencies with its 1.6 counterpart. I really like how free went into depth with his analysis of the bullet distribution in the X and Y direction, noting patterns and deviations from these patterns (randomness). We really need more people to be doing this kind of thing to help get the recoil to where it needs to be with ALL weapons.

And Dare IS right: huge blood spatters, small bullet decals, and sound are all factors that make controlling recoil (or at least observing it in battle) harder.

HOWEVER...Free's test eliminates all of those extra factors that make recoil hard to control by spraying into a wall (no blood splatter) at a relatively close range (bullet decals clearly visible). And even with these factors set aside there are STILL noticeable problems with the recoil. In my opinion, while all the things Dare mentioned make it harder to SEE the recoil, you honestly don't need to SEE the recoil in order to be able to control it effectively. Knowing the distance of the target and how many bullets into a spray you are is more than half the battle. Better sound and visuals only help you 'fine-tune' and gauge how well your muscle memory served you on-the-fly (so you can make adjustments)...but cmon, you DON'T need to see every bullet decal perfectly...it's not like you can say 'I saw the 9th bullet of that particular spray here, so I aimed there' it's more about the overall spray patterns.

[Improved]

...it's more about the overall spray patterns...

If you ever gonna get "perfect" results, then the way free is doing it is the correct way. Meaning the picky way.

Trying to decifer the non random an random calculations, the shooting speed, the kickback of the weapon and singel bullet behaviour takes alot of time and you gotta know what you are looking for and how to proceed.

The devs themselfs a year back or so said that the recoil is a bitch to get right and that they had spent already hours upon hours trying to get the calculations right.

It still needs tweaking, therefor the job free is doing, working solely on the behaviour of the weapon, is plain awsome. I wish I could help out, but it seems he knows what hes doing and I think that he's on the right path and also getting help from others too.

So it's not just about the overall of it all, its the tiny calculations that will make the recoil feel as in CS1.6

[thomsonic]

I have no idea how the coding works, how you guys do it, and if this is a stupid ignorant question, but I hope you (the devs) have considered that recoil/spray patterns may contain more mathmatical elegance than (maybe?) presumed. Much more variable and complex calculations? It seems like you're working on a very linear basis. Like with the tagging and blood gooshes. It seems like it didn't even occure to you that it has to correspond with the damage you take how much you get slowed down/how much blood that has to appear. You just made a constant type of tagging/blood gooshes that slows you down the same/creates the same amount of blood without paying regard to the specific circumstances etc.

[Dare365]

yea Dan, there's been a few times I crouched, carefully steadied my aim over a long-range target and it took just too many single bursts before my target went down (sometimes the target was even stationary).

As for Dare and Improve's stupid argument (no offense): you both brought up valid pts within the countless insults. Anyway, here's my opinion of free's OP vs. what Dare originally brought up:

Free's test was strictly for the recoil of the AK47, which I also believe has some inconsistencies with its 1.6 counterpart. I really like how free went into depth with his analysis of the bullet distribution in the X and Y direction, noting patterns and deviations from these patterns (randomness). We really need more people to be doing this kind of thing to help get the recoil to where it needs to be with ALL weapons.

And Dare IS right: huge blood spatters, small bullet decals, and sound are all factors that make controlling recoil (or at least observing it in battle) harder.

HOWEVER...Free's test eliminates all of those extra factors that make recoil hard to control by spraying into a wall (no blood splatter) at a relatively close range (bullet decals clearly visible). And even with these factors set aside there are STILL noticeable problems with the recoil. In my opinion, while all the things Dare mentioned make it harder to SEE the recoil, you honestly don't need to SEE the recoil in order to be able to control it effectively. Knowing the distance of the target and how many bullets into a spray you are is more than half the battle. Better sound and visuals only help you 'fine-tune' and gauge how well your muscle memory served you on-the-fly (so you can make adjustments)...but cmon, you DON'T need to see every bullet decal perfectly...it's not like you can say 'I saw the 9th bullet of that particular spray here, so I aimed there' it's more about the overall spray patterns.

This is probably going to piss a lot of people off but please bare with me.

I can agree with most of what you said however the stem of my reasoning for being so adamant about the issues I mentioned is because they are more of a core issue with the overall game than just the recoil of the AK. The recoil isn't difficult to control in comparison to the old one, its simply different. I've aced spraying the AK in CSP, its not impossible to control.

Don't get me wrong, I'm a 1.6 player and the closer we make it to 1.6 the happier I am but if Valve released CS 1.7 and the recoil worked like it does in CSP, we would hear a large outcry from the community, as we should, but in less than a year the community would adapt and forget about it. I mean, just think about how different guns are from beta to 1.6 or even the AWP from 1.5 to 1.6. We cried and bitched about the AWP delay and acted like Valve broke the game but before we knew it, we had embraced it.

The recoil obviously needs to be addressed sometime in the future simply because the CSP dev team isn't Valve and they cant get away with such a change, but I think blaming the control of the AK mostly on the recoil is a cop out. The AK has always been a difficult gun to spray with and I think people would be much more satisfied and understanding once they understand where their bullets are going instead of crying foul on server reg or messed up hitboxes. In other words, I think recoil is a beast that needs to be tackled after the core game mechanics are working properly. Especially with the AK, precise recoil is a very difficult monster to approach when everyone has their own ideas on how the AK works.

Anyone who has played high level Counter-Strike knows that a lot of the problems players face are in their head. Again, I'm not saying the AK recoil isn't messed up, but I'm saying that from what I've seen in my decade of playing this game competitively, your worst enemy can be yourself and I think its very easy to blame the game before you blame yourself when you don't have evidence of your own fuck up. Its not a personal insult, I think literally every Counter-Strike player ever has blamed the server or the hit boxes or interp when it was their own aim that was at fault, and its 10 times easier to blame the game and transitively the recoil when you and everyone spectating you can't clearly see where bullets are going.

[deefop]

fyi to test the recoil and see where the bullets are REALLY going you can type cl_lc 0 and cl_lw 0.
that removes lag compensation.

--- Update ---

also #33

its not even just about recoil... like other people have said, if you crouch and fire single shots at a stationary targets head and chest, they should drop. Theres no excuse for that. Playing csp i can't even count how many times I would tap someone in the chest or head over and over and over and have them not die. It's obvious enough that tons of people are complaining about it. Recoil only really takes effect AFTER you've shot 4 or 5 consecutive bullets to the point where its no longer easy to control where your rounds are hitting. But on the first 2 or 3 single shot taps there basically no recoil to speak of, and therefore there is no reason why those shots shouldn't be hitting the target.

[wHEElEr]

"but I think blaming the control of the AK mostly on the recoil is a cop out"

did he really just say that?

get a clue dare.

[Dare365]

fyi to test the recoil and see where the bullets are REALLY going you can type cl_lc 0 and cl_lw 0.
that removes lag compensation.

--- Update ---

also #33

its not even just about recoil... like other people have said, if you crouch and fire single shots at a stationary targets head and chest, they should drop. Theres no excuse for that. Playing csp i can't even count how many times I would tap someone in the chest or head over and over and over and have them not die. It's obvious enough that tons of people are complaining about it. Recoil only really takes effect AFTER you've shot 4 or 5 consecutive bullets to the point where its no longer easy to control where your rounds are hitting. But on the first 2 or 3 single shot taps there basically no recoil to speak of, and therefore there is no reason why those shots shouldn't be hitting the target.

Yes! That's exactly it. There are things in the core game that don't work and needs fixing and to blame it on the recoil of the AK is mostly false. Shit, a few pages ago we have first hand footage of the AK acting relatively the same as the one in 1.6. Its not perfect but its not the biggest issue right now.

The recoil IS different but a lot of people think they are missing because its different when in fact their shots aren't connecting, players occasionally have invincibility and their is a lack of feedback as to where bullets are going. I've had just as many instances with the glock, USP and M4 as I had with an AK where I feel like I should be killing them and it doesn't happen. It sounded like you were disagreeing with me at first but what you typed is exactly what I'm talking about.


"but I think blaming the control of the AK mostly on the recoil is a cop out"

did he really just say that?

get a clue dare.

Such constructive criticism. I give you 4 paragraphs and you give me 2 lines of BS text. Stay classy.

[free]

If you happen to like the CSP blood, recoil etc then please stop arguing here and leave the topic alone as it is meant to describe the diffirences in both games. It's up to the devs if the AK will be more 1.6 or not.

If you find the particle of CSP disturbing then please make screenshots or movies with all the blood particles you can get in 1.6.

If you happen to dislike the recoil as I do then try to record your own demos on the dev map on both csp and 1.6, then slow the demos 10 times and upload them or better try to compare them.

[heyron618]

Improved: do you realize that you took that comment completely out-of-context? I was praising Free’s hard work and attention to detail just as you were. When I said, “It’s more about the overall spray pattern” I was referring to Dare’s mention of the sound and visual clues you use to track your recoil. You don’t need to clearly see every bullet to be effective at spraying just the overall pattern.

Free is definitely on the right track with what he’s doing. Only suggestion I’d have for him is to ‘crunch numbers’ quantitatively IN ADDITION to the qualitative observations he’s making. In short, using statistical analysis by utilizing the exact coordinates of each bullet revealed by commands like ‘viewpunch’. Just remember, I’m in no way putting down ANYTHING Free has done thus far. In fact, after seeing other projects he’s worked on (ex: particle editor), I’m simply amazed at his modesty for not already being a developer or official contributor to the CSP team. Maybe I’ll take a crack at the statistical stuff when I’m not too busy or elaborate on what I mean so someone else can do it.


Dare: just so you know, I NEVER said that recoil is the only issue that needs to be addressed, nor did I say it is most important one. You’re talking about the ‘big picture’ of CSP and are absolutely dead-on with comments like, ‘If this was CS 1.7, people would bitch but then adapt to the recoil’ and ‘the exact recoil of 1.6 is not important to make this game competitive’ and 'people blame the server more than themselves'. However, you’re trying to convince people more stubborn than me that in a thread that is FOR ‘AK47 recoil’(aka: unrelated to what you brought up).

I mean, I UNDERSTAND what you’re doing…you probably assume that most of the people commenting and viewing this thread think exact 1.6 recoil is MORE important than it actually is, and yes…that is a fair assumption, but it is also a stereotype.

People want 1.6 recoil is not just because of nostalgia and elitism, but because it is something that HAS ALREADY worked in past competitive environments. And while recoil that is different is not necessarily worse competitively, in this case it IS worse.

For example: CSS’s recoil is thought to be inferior to CS 1.x’s and for good reason. For one, recoil is a lot lower (which allows less skilled players to control spray more easily). Second, recoil is more ‘random’ meaning if you shoot a particular way one time and kill a guy, there's a good chance doing the same thing later could fail…I can provide more in-depth examples if you'd like, but to avoid writing a novel I’m just gonna stop here.

Either way Dare, the CSP dev team is going to continue to work on recoil AND other things that you’ve mentioned, just as they were beforehand. All I was saying is recoil's impact shouldn't be downplayed, even though the reason people bitch about servers is about MUCH MORE than just recoil (netcode, animations, movement, sound, other visual cues, etc.)

[JinXR]

The AK recoil - tap bullet pattern is horrible and needs fixing, lesser of the two evils is the actual spray pattern.

Enough said, no need to write 1000 words on something you can sum up in 1 sentence.

[Lordearon]

listen to free and start posting screenshots/movies and less walls of text please - also take your sissy fights out of this thread.

end. it. now.

[JinXR]

How come the majority of players (especially in Australia and part of this forum) know the AK has an issue. Yet, i've heard many people on these forums deny it exists.

With the amount of negative feedback been given on the AK (and it seems to be the one issue).
EDIT: I've just gone into a lan server of my own and tested the recoil and the tapping. Yes it does look exactly the same, if any thing, the CSP recoil is slightly wider at the end of the clip than the 1.6 AK.

I'd suggest looking at far shots v enemy where the AK lacks registration, anyone else agree?

[relentless]

yea agree, for me its not the recoil, its the sniper rifle like single shots you could do with it in 1.6 that is lacking.

[drixa]

Definitely agree that the sniper 1 tap, two tap, is off. It should be deadly accurate but its slightly off. Just something to tweak.

[Quexel]

push

[DanLegenD]

this came from a bug in the engine they have discovered currently and rectified in the CS:S beta. Interesting enough Valve actually did port "MOST" of the gun code to CS:S ( aside from the Deagle) the reason its so much more easier is because of the gigantic models, which i feel there should be a hardcore mode in CS:S just like there is in other modern FPS' that makes the models smaller to CS 1.6 size.. but i digress. The bug I'm talking about is first shot accuracy after you switch to AK, the first shot has the smallest deviation in the game, equal to that of the awp, after the first shot it goes to the recoil we have in CSP now for the AK, very hard to aim with. CS:S beta has changed this so that the first shot now is at a half way point between the old deviation and where the deviation settled after the first shot, and now it always go back there.

EXPAMLE: Current first shot deviation with AK47 (and m4 but its not as precise) in CS 1.6 and retail CS:S is equal to the awp, almost non-existent deviation. After first shot, no matter how long you wait or if the round restarts deviation is 3 times larger than the first shot, unless you quickswitch weapons (double-tap q) or reload.

Current CS:S Beta AK47 first shot deviation is only 2 times larger than the awp deviation, they understand that it shouldn't be as extreme as it is after that first shot because of people having the habit to quick switch to get good recoil but they know first shot can't be as precise as an awp shot. The new system also returns to that deviation 2x awp deviation after recoil settles, which is a timely matter

Currect CSP AK47 first shot deviation is 3x larger than awp deviation, because when they practice it they dismiss the first shot accuracy and assume its lumped in. if they do a 2-shot 15 times in a single clip (2 x 15 = 30) then that means the deviation set by them is about 2.98 times the size of awp deviation. Not very fun when most pro's know to quick switch instinctively (even though valve just confirmed this issue, pro's have always intermittently quick switched with AK and other rifles because of instincts) in cs 1.6 and they come to csp and cant hit that first shot no matter what its frustrating.

[tiger]

DanLegenD is right (except it's not an engine bug, it's just how the weapons were coded). The tweak I asked Ben to do is related to that.

Also, it's not 3x less accurate than the awp, it's actually 1.75x. And the first shot in 1.6 isn't exactly equal to the awp, it's a little bit less accurate if I'm not mistaken but it's hardly noticeable.

[nki1]

This is long known and was brought up in a lot of other threads. However the team never stated if they try to fix the bug for CSP and closed the threads...

[Patrickk]

yeah realy nice job, and i hope it will be recognised!(sry for my english, hope you understand)
i have to agree to any point free said, and i also feel the CS 1.6 weapons are "faster" shooting. And thats a main problem to create the feeling of CS 1.6.

in my opinion in phase 1 the increase on the y axis after the 3th bullet isnt enough. That makes it feel very like source....
i realy think that needs to be fixed. But if you achieve to manage this problem, another big step is done!

[majkool]

I studied some files and I know how to slow down shooting and I also have analyzed appearance of this weapon recoil.

Full auto 30 ammo.

Distance from wall (4*64)
http://www.mediafire.com/imgbnc.php/1e933eb809dea54b55294e8d1b38d16c5dcb002cf32ab6ff41 c1dd67423d00396g.jpg

Distance from wall (6*64)
http://www.mediafire.com/imgbnc.php/0631d4c7cf6f9f8e684ee87cb813a5a7e41fe2098aa32eb515 3b6948baf461806g.jpg

Distance from wall (8*64)
http://www.mediafire.com/imgbnc.php/5250dccac9b521588f07a0136bdab5798f394656b0030a8a26 073a10140940a46g.jpg

Distance from wall (10*64)
http://www.mediafire.com/imgbnc.php/4de2caffcf182a74ad3f304dc0677e1f230091c002aa1dd970 760fc0f53a28786g.jpg

Distance from wall to wall
http://www.mediafire.com/imgbnc.php/7ef62ccd0700a3e76e2b606589fd8f761694e2269fc7383578 a50e4743c458306g.jpg

[n69ky]

your first and second pictures are the same, but they are titled with diferent distances.

[majkool]

your first and second pictures are the same, but they are titled with diferent distances.

O rly?!

That was one of the point of this analysis.

[n3krO]

majkool, i don't remember very well but i think at cs1.6 recoild is more like 3rd and last picture (for ak47). Am i right?

[majkool]

From these images, we can conclude that ak47:


have random recoil side (left & right)
height and width increases with distance from the target (this is normally for CS game, but not if these values are higher than like in 1.6)
and maybe some more which I haven't noticed yet...

[free]

Nice research. I'd suggest making the side recoil slightly weaker than it is in csp 1.05 and the bullet spread side should switch just like 2-3 times per clip. It does now twice as much.

We need to catch up and exchange some data i guess majkool. Ill be on the old teamspeak at 8 probably. I'd appreciate if you were online then. Thanks

--- Update ---

Editing ctx files was on my todo list since 1.04 hotfix but I never actually felt like doing so. Good someone from the community finally got his hands on these files ^^.

[n69ky]

all i wa saying is, the first two pictures were the same. which you have fix'd now. nothing about the text or something...

[free]

Actually it reminds me of one thing about "promods". The first "promod" i remember of was on Quake3 and it was named Quake 3 Challenge Pro Mod Arena.

The q3cpma changed q3's random shotgun recoil into a predictable version making it shoot always the same, just like you did with ak while your test. Few years ago (when csp was at 1.01 as far as i remember) i thought it would be good to have 100% predictable recoil in csp. I don't think it's a good idea today but i just wanted to share that with you guys.

[n3krO]

Removing the predictable recoil of cs1.6 we'll lose another skill factor which is bad and can make the mod fail...

[free]

Removing the predictable recoil of cs1.6 we'll lose another skill factor which is bad and can make the mod fail...
I was talking about the UNpredictable factor, not the predictable one. The UNpredictable factor can be called luck and relying on luck != skill.

But then remeber this mod is about recreating the recoil of 1.6 before any changes happen.

[n69ky]

even if the code contains some randomness, its predictable :P

But then remeber this mod is about failing.

[tiger]

There's really no point in doing that atm. The issue with the AK comes from the minimum value for m_flAccuracy which did not match 1.6, thus making the first bullet and those that follow not as accurate as 1.6.

Also, please do not try to edit the weapon scripts, Redstar has access to those files (so he can edit them any time) and there's nothing related to the spread/recoil/accuracy in there (aside from the ROF).

[majkool]

But we can slower down of spead shoting in there, maybe when we do it recoil of weapons will be corrected or look like in 1.6

[yuri]

If you really want to try mimic the 1.6 rate of fire -> CSP AK empties it's clips in the same time as 1.6's AK... However Red spent a lot of time doing this, so I dunno.

Also wait for the variable tweak which tiger mentioned.

[free]

You need to keep a closer look at the usp, its also off while fast shooting.

[majkool]

There's really no point in doing that atm. The issue with the AK comes from the minimum value for m_flAccuracy which did not match 1.6, thus making the first bullet and those that follow not as accurate as 1.6.

Also, please do not try to edit the weapon scripts, Redstar has access to those files (so he can edit them any time) and there's nothing related to the spread/recoil/accuracy in there (aside from the ROF).

Simple he edited "AccuracyOffset" and "CycleTime" in .ctx and .cpp but why he left the other weapons?

[yuri]

Simple he edited "AccuracyOffset" and "CycleTime" in .ctx and .cpp but why he left the other weapons?Because there are many others issues in the code. Pistols has their own system, etc.

[free]

Simple he edited "AccuracyOffset" and "CycleTime" in .ctx and .cpp but why he left the other weapons?
It's not as simple as that. AK and many other rifles need diffirent AccuracyOffset for the first bullet and a diffirent one for each next bullet and that is just an example.

[sseyler]

My clanmate made a video to compare recoils. The recoils seem the same on the video (except USP which is way off) but the ROF is diffirent on AK. There is a slight chance the video has wrong timing but he's been making game videos (mostly cs/css) for some time now and i wouldn't suspect such an error of him.

Either way here's the outcome:

http://www.youtube.com/watch?v=obS4mlYd3a8

Thanks for the video sicler!

This video seems to corroborate the feelings I've had about the spray recoil for the AK and Colt. From looks alone, the recoil patterns of the AK and Colt in CSP appear to be very similar to those in CS 1.6, although they seem slightly more random. The main difference is in the feel of the recoil, which I liken to the fact that in the "second phase" of the spray (where the gun moves horizontally), the bullets are above the top vertical crosshair in CS 1.6, whereas in CSP, the bullets are closer to the middle of the top vertical crosshair. This tendency is very apparent in the way one must move his crosshair when spraying (with an AK, for exmaple) at a target; i.e., in CS 1.6, one must place the crosshair closer to the stomach and in CSP, one must place the crosshair near the middle of the chest.

[free]

The video is from 1.05. AK and dealge were changed in each patch so it's hard to believe this old video describes 1.07.

[tiger]

By the way, if you guys think the AK was better in 1.06 we can revert the changes (in this case it's just a script value so very easy).