With cordite in the air, splintered steel, shell casings and powder burns, there’s only one explanation...
Post feedback & meet new friends. General SG discussion.

Moderators: Pardner, TheDoctor

Postby Sucalakafufu » Tue Dec 29, 2009 12:45 am

moRtem wrote:Apart from that: Wow, you had a lucky spread and did 2/3 headshots with a rifle on a non-moving-distance, while you are not moving yourself, on medium distance. That's a hell of a rifle!

lol well im able to do this. if i have a low ping. its more of knowing the feel of the rifles (especially since the feel in 1.0 isnt so good :P)

moRtem wrote:I aimed at the middle of the head (where the nose is supposed to be)

in my experience this usually does end up with a headshot... though aiming very slightly higher probably would give u headshots more often

moRtem wrote:Yes -- aiming at head equals to hardly ever land a hit, even when you are aiming accurate.

well im not sure for u. but its not too hard to hit head for me (at least when challenging players with less skill). i havent experienced the "i cant hit head at all!" except with packetloss and bad ping

Joe Kari wrote:I agree, there is something wrong with the arm hitzone.
The best way to fix that IMO, should be that arm, hand and foot hitzone should not stop the bullet, allowing it to follow it's line and try to touch another hitzone (with something like 50% of remaining damage)...

this seems like a good idea and i would like to test this sometime :)

mLy wrote:I gave a quick example, Shoulder good be in the 'body' hitbox.

ah ok. even though i dont like that idea too much i see your point
SG Name: Sucalakafufu
Clan: [CWNN] - Clan With No Name
User avatar
Sucalakafufu
Gunslinger
 
Posts: 239
Joined: Thu Apr 16, 2009 8:59 am



Postby moRtem » Tue Dec 29, 2009 1:11 am

No offense, Sucalakafufu. But i even provided footage now which clearly shows what i keep writing in this forum.

Watch the vids, and stop that blabla about bad ping, packetloss, "works for me", etc..


Face the facts that are clearly shown by the vids i posted and ask yourself if it's cool that the game randomizes every shot you make.




/quit
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



Postby Sucalakafufu » Tue Dec 29, 2009 1:53 am

@moRtem: i did watch your vids.
i was just saying that i do not get the same inaccuracies as in your footage.

i just tested those same weapons. first: i dont think the pistols in 1.1 have the same accuracy over that range as in 1.0. (not sure but that is my understanding). i also have to say that the feel of the rifles to me is that the crosshairs on the rifles are not entirely accurate. this could add to the feeling of randomness. as long as i ignored exactly where the crosshairs were, i got more accurate shots. as for the pistols, the remington seemed to shoot fine in a closer range than in your videos. the S&W seemed about right (assuming that its not supposed to be as accurate as the remington and peacemakers)

i also shot off shots with the peacemakers and a sharps rifle. the sharps was razor accurate and the peacies were just a tiny bit less (in their respective ranges of course)

needless to say, debugging mode did prove to show slightly more accurate spread than in your vids. but when i added bots and didnt rely too much on the crosshairs, i was able to hit where i wanted at least 60% of the time. sometimes the spread made it go slightly off but not too much (i.e. hitting neck instead of head). i dont think any of the guns are supposed to shoot like a railgun except for the sharps

so maybe my suggestion is for us to make sure the crosshairs are also accurate to where the gun will be shooting.
SG Name: Sucalakafufu
Clan: [CWNN] - Clan With No Name
User avatar
Sucalakafufu
Gunslinger
 
Posts: 239
Joined: Thu Apr 16, 2009 8:59 am



Postby moRtem » Tue Dec 29, 2009 2:05 am

Sucalakafufu wrote:so maybe my suggestion is for us to make sure the crosshairs are also accurate to where the gun will be shooting.


When i shoot at the same spot 12 times (akimbo-mode), and i have a close look at the wall i shot afterwards, where it clearly shows 12 different bullet-marks, then where does this have anything to do with a crosshair "probably being off" ? (considering that neither mouse was moved, nor the move/strafe-buttons where used)

It's simply random spread.. nothing more, nothing less.

Same counts for rifles and anything else. (yes, also the sharps is not totally accurate.. but accurate enough, even on mid-to-long-distance).

this could add to the feeling of randomness.


it's not a feeling, it's a fact.


/quit
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



Postby Coup'George » Tue Dec 29, 2009 2:35 am

@moRtem :

Same test, but my spread seems different than your :roll:

http://www.dailymotion.com/video/xbo6me ... videogames



Maybe you change some cvars about spread ?
User avatar
Coup'George
Smokin' Amigo!
 
Posts: 7
Joined: Tue Oct 16, 2007 5:30 am



Postby Sucalakafufu » Tue Dec 29, 2009 2:44 am

moRtem wrote:it's not a feeling, it's a fact.

i didnt mean in the tests. the tests show inaccuracy yes. i meant more when testing on bots. thats where the "feeling" of the gun comes in. thats where the crosshairs feel inaccurate to me.

i also did not say anything about pistols' crosshairs. just rifles' crosshairs

btw with my tests i get about the same spread as in Coup'Geourge's vid. i also get about the same headshot rate. even from pretty far distances
SG Name: Sucalakafufu
Clan: [CWNN] - Clan With No Name
User avatar
Sucalakafufu
Gunslinger
 
Posts: 239
Joined: Thu Apr 16, 2009 8:59 am



Postby moRtem » Tue Dec 29, 2009 3:03 am

Coup'George wrote:Maybe you change some cvars about spread ?


Did not, like stated a couple of times already.

My tests are based on 1.1b3 like it was delivered by the downloadable package. I did my tests offline on a local server (simply by typing \devmap br_alamo_tiny).

Is it possible that you tested on one of the official 1.1b3-servers? - according to this topic, the spread got modified there a bit:
viewtopic.php?t=2005

--

The core-problem stays the same though, even when spread is less, there is randomness within the given spread. Which often decides about whether you're hitting one hitzone, or the other. And therefor may decide about win or lose in a battle.


/quit
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



Postby Joe Kari » Tue Dec 29, 2009 3:21 am

The only fact there: you don't have a standard SG version ;)
I play this game since Western Q3 2.0 (probably 4 or 5 years), and what I saw in your video has *NEVER* happened, not in WQ3 series, and not in SG series including current SG 1.1beta3.

Coup' George's video demonstrate that the spread is not that bad, that there is not so much randomness and that hitboxes should be kept.
And during my test, I have had better result than him.

You can speak about randomness at long / very long range, but at mid-range and close range, all is about aiming skills.
User avatar
Joe Kari
SG Team
 
Posts: 879
Joined: Sun Sep 16, 2007 8:44 pm
Location: France



Postby moRtem » Tue Dec 29, 2009 3:48 am

Joe Kari wrote:The only fact there: you don't have a standard SG version ;)


This is the very last time that i repeat myself: I use the standard 1.1b3 package provided by tequila. The only change i made, is that i use the SDL.dll which got patched by ioq3-team, because of this bug:
viewtopic.php?t=1987

I aswell didn't apply any changes on weaponstats.


/quit
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



Postby dowoshek » Tue Dec 29, 2009 12:27 pm

Maybe it's some kind of problem with system/hardware configuration? I didn't try to test with debug mode, but I clearly get similar results as Mortem got - check my post sent 2 weeks ago:
http://forum.smokin-guns.org/viewtopic.php?p=16051#16051
User avatar
dowoshek
Gunslinger
 
Posts: 134
Joined: Tue May 19, 2009 12:29 pm
Location: Poland



Postby moRtem » Tue Dec 29, 2009 3:06 pm

Another episode. This time with a 100% clean and fresh SmokinGuns 1.1b3 installation. Tested locally with default settings (except /cg_brasstime 25000 , /g_debugweapon 1 and /bot_pause 1 .. aswell i /bind n +attack and /bind m +button6 to be able to shoot without touching the mouse):
http://www.youtube.com/watch?v=-MYMeZm7pcA
http://www.youtube.com/watch?v=V3sbiwIPWLo
http://www.youtube.com/watch?v=bXk9_7q4MwM
http://www.youtube.com/watch?v=xmIoDn5rNxo
http://www.youtube.com/watch?v=0yGB1UjuN4U


Demos can be downloaded here:
http://www.megaupload.com/?d=0JN7NF4P


--

Now it would be constructive if devs would realise that there's a massive problem, instead of accusing me of providing falsified footage.


/quit
Last edited by moRtem on Tue Dec 29, 2009 4:55 pm, edited 1 time in total.
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



Postby Tequila » Tue Dec 29, 2009 4:08 pm

:lol:
You start to be too nervous to continue a fair discussion.

Seems something still may be wrong on your part. I check my "spread" settings by just typing "spread" command on my console. I can reproduce your shot pattern when I double the spread for the Winchester 1866.

Your videos are not good proof. When submitting clean bug report, you should describe precisely each action we need to do to reproduce. But just be patient and show you're patient if you really think you are pointing us to a very dirty bug.

Can you check if you don't have any pollution in your home folder ?
Did you remove your q3config file when installed a fresh copy ? We need to know that too.
Can you start again your case from the fresh install, open the console, type "spread" command, and even "cvarlist", then "condump dump.log" and submit the "dump.log" file on pastebin ? This will help us to verify your settings and eventually to identify where is the problem on maybe our part by just comparing the same kind of stuff.
8)
User avatar
Tequila
SG Team
 
Posts: 1095
Joined: Thu Nov 15, 2007 11:33 pm
Location: Montpellier, France



Postby moRtem » Tue Dec 29, 2009 4:23 pm

Tequila wrote:Can you check if you don't have any pollution in your home folder ?


when i said fresh install, then i meant a fresh install

Did you remove your q3config file when installed a fresh copy ?


When i said a default q3config, then i meant a default q3config

Can you start again your case from the fresh install, open the console, type "spread" command, and even "cvarlist", then "condump dump.log" and submit the "dump.log" file on pastebin ?


Anything to serve you right:
http://pastebin.com/m31f6ac0

You start to be too nervous to continue a fair discussion.


not really

Seems something still may be wrong on your part. I check my "spread" settings by just typing "spread" command on my console. I can reproduce your shot pattern when I double the spread for the Winchester 1866.


i don't even have that command.

Apart from that, i believe that you are not testing with default settings, since you got altered weapon-stats on all your 1.1b3 servers.

--

Anyhow -- to keep repeating myself: I am not trying to find a nasty bug here. I simply want to show that the existance of spread is pure nonsense, especially together with having different hitzones.

It's a game, and not a gunfight-simulator.


/quit
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



Postby Tequila » Tue Dec 29, 2009 5:07 pm

moRtem wrote:
Seems something still may be wrong on your part. I check my "spread" settings by just typing "spread" command on my console. I can reproduce your shot pattern when I double the spread for the Winchester 1866.


i don't even have that command.

You must start a map to try this command. So start the a map, type "spread", "cvarlist" must be also run when a map is started. Did you understand we are debugging qagame QVM ? So it must be started, and it is only started when a map is launched.
Can you try to provide a more complete console dump ? Easy on linux, but I really don't know other windows. Do you have an "stdout.txt" file somewhere ? This one should come from SDL and can be the complete out put of your client.
8)
User avatar
Tequila
SG Team
 
Posts: 1095
Joined: Thu Nov 15, 2007 11:33 pm
Location: Montpellier, France



Postby moRtem » Tue Dec 29, 2009 5:20 pm

Tequila wrote:You must start a map to try this command.


Blubb -- my mistake - i really did it without having a map started.

the condump is here:
http://pastebin.com/m44a22e2d

So basically it's the spread-settings which are also noted down by torhu (with the exception that i seem to have sharps 100 and he wrote, it's set to 200 - i guess torhu's topic lacks an update there):
viewtopic.php?t=451

So start the a map, type "spread", "cvarlist" must be also run when a map is started. Did you understand we are debugging qagame QVM ? So it must be started, and it is only started when a map is launched.
Can you try to provide a more complete console dump ?


The cvarlist is the full one (made, after a map was launched), otherwise you wouldn't see g_debugweapon and so on.

Do you have an "stdout.txt" file somewhere ? This one should come from SDL and can be the complete out put of your client.


Such a file doesn't exist on my PC.


/quit
User avatar
moRtem
Gunslinger
 
Posts: 284
Joined: Sun Mar 15, 2009 5:56 pm



PreviousNext

Return to Saloon

Show Sidebar
Show Sidebar

User Control Panel