[karnute]

[update: correction not sure the dmaHD disabling can improve no-hits, see my later reply in this thread]

Short: In my case, disabling dmaHD sound makes the diference to remove some subjective no-hits.

Long:
Since version 4.2.017 (or maybe earlier), i had the subjective feeling that there was something wrong in the way i view some players in crowded servers (i had pings around 50). It was like if they were warping but with small movements. When i was aiming to them, they suddenly jumped side to side almost randomly like warping. Something as if they were not where i see them, and it was different from the usual straffing to avoid bullets. That strange movement does not appear in demos autorecorded. Reviewing my demos in slowed down time, i was really shooting to the sides of where the player was, even from short distances. The rest of rendering seem to be normal and smooth (except very small fps drops in uptown, etc.), so my general impression during gameplay was simply like "no-hits", and after viewing my demos the sensation was like i was failing to aim.

I tried to change some cvars, tweaking my mouse, sensitivity, etc. with no results, until I disabled dmaHD sound (seta dmaHD_enable "0"). I need to do more tests, not just a pair of matchs, but for the moment it feels better now. Maybe, these are not the same "no-hits" that others perceive, but just in case it could be of any help to somebody...

I use official binary for Linux 64bits, clean installation (no 3rd party .pk3 maps, etc.), locked to one processor. The computer is a bit old (2008), core 2 duo and geforce 9800-gtx, but it worked ok before. Just disabling dmaHD sound, it started to work similar to other previous versions. Maybe the computing for dmaHD sound is too much for my old CPU. It is a pity, because i prefer the dmaHD sound by p5yc0runn3r, it seems more natural and i could distinguish with more precission from where the sounds came, but rendering of players is more smooth with it disabled.

This post has been edited by karnute: 13 February 2014 - 10:03 PM

[vareul]

Karnute, i will give that option a shot. If the dmaHD fix is related to hardware performance, it probably wont help me.

My computer specs:
AMD FX-8150@4.2GHZ(no turbo)
8GB G Skill Ripjawz X 1333MHZ Dual Channel RAM
Nvidia GTX 760
Gigabyte GA-78LMT-USB3 motherboard
160GB WD HDD from 2006.
1 TB Segate External HDD from 2013

All of my games are installed on the external. I can play BF3 on ultra at 60 FPS stable, but i get FPS drops in Urban Terror. Maybe the FPS drops and hit issues are combined in some situations. I was at 120FPS stable during the crouch and fire situations. I rarely have above 30FPS in heavy combat.

My primary install has 3RD party maps in it. My secondary install is getting the "No server list from master server" error.

Update: The dmaHD fix did absolutely nothing to help the FPS drops. Guess i cant check to see if thats part of my issue.

This post has been edited by vareul: 02 February 2014 - 04:49 AM

[FS Frankie V]

vareul, on 02 February 2014 - 03:48 AM, said:

Frankie, i wasnt accusing anyone of hiding anything. i was trying to make it clear that i dont suspect anyone of wrongdoing. I know that noone is hiding things, and i want you to know that i know that. Figured it might help if you were at ease about the interaction between us. I am presenting a controversial issue, after all.

Sure no harm and no foul and if there is a problem that does not pass the practical test side of things we would want to know about it. If you say you know the hell what you are doing as to setting up a server I'll take your word for it.

However.

I am unclear if you are referring to the problems you are having as a player or a server operator.

As for how things work as to server responsibilities here is link as to how the Source engine/server works.

NOTE: It is not identical to the idtech3 engine but is easier to understand and most docs on idtech3 are difficult to understand the basics and if anyone has a link to idtech3 netcode as to how it works in everyday language please send it my way. It is however the same in principle.

https://developer.va...ayer_Networking

The bottom line is even on a 14 year old engine the service it needs to supply is not as fragile as to ping or network conditions that are not obvious and if something does go south then needs to be corrected by the numbers.

If it's a case of you having troubles setting up a server then the standard operating procedure is to PM contact(at)urbanterror.info and give us as “much” information as to can think of as to what you have set up as to relative information.

[vareul]

Im having the issues as a player. Ive been playing Urban Terror for about 4 years. Im not the most knowledgeable, of course. However, i have enough knowledge of Urban Terror, to know when i should at least hit my target with one bullet. I wouldnt see this as a problem if i was only encountering these problems during mobile combat.

I reread the thread. Two other people have mentioned having similar issues. One said they fire off entire magazines without to much as a single hit. The other person said they have trouble with hits when the target is right in their face. Ive had both of those problems, and consistently at that.

Im aware that its certainly possible to miss 30 shots if youre jumping around and sliding and using full auto. Im not bothered by that at all. Im bothered by being unable to land a hit or kill on someone who isnt even paying attention to me, and im using common sense when trying to hit them.

Even if all of this is figured out and fixed, im not going to end up some pro league prodigy. Im of average tactical skill, and i accept that. My aim isnt bad, just not 100% consistent in mobile combat. I just would like it if i could help my teammates when i get a good shot lined up.

This post has been edited by vareul: 02 February 2014 - 05:52 AM

[FS Frankie V]

OK as a player if you are having these issues connected to a server then we would need information as to the server(s) you are referring to.

As mentioned the server manages all things collision and hits wise and the client (player) does not even require a hit mesh except to ensure that it is loaded as part of the common package. The test process I described will however prove simple enough if the server has a control file miss-match and to restate it's the servers responsibility to ensure hits occur and “not” the client.

Factors like ping and fps do come into play but not at the levels as described.

P.S. A 2-3 second of burst with anything is spray-n-pray territory by the way.

[vareul]

Frankie V, on 02 February 2014 - 06:08 AM, said:

OK as a player if you are having these issues connected to a server then we would need information as to the server(s) you are referring to.

As mentioned the server manages all things collision and hits wise and the client (player) does not even require a hit mesh except to ensure that it is loaded as part of the common package. The test process I described will however prove simple enough if the server has a control file miss-match and to restate it's the servers responsibility to ensure hits occur and “not” the client.

Factors like ping and fps do come into play but not at the levels as described.

P.S. A 2-3 second of burst with anything is spray-n-pray territory by the way.

Im referring to Fallin Angels CTF, Canadian Carnage, and some unbranded 4.2 CTF thats usually pretty full around 4PM EST.

I know that holding the trigger down causes spread. I know that theres some spread during 3 round burst. I know that still applies when stationary and crouched, and your target is stationary. However, the spread isnt (shouldnt) be enough to cause a triple miss in that circumstance. I also know that semi auto shots can be off center if you are moving, or during scope penalty.

I dont have an issue with missing cause of spread. If i miss during pray and spray, thats how the game goes. Ive tried to narrow the problematic examples down to ones that dont include situations of pray and spray, or possible lag.

Im more worried about the stationary, crouching shots, and the ones that miss targets literally less than a body length from me, than the ones that miss when moving. Mobile combat can be hit or miss depending on ping as well as bad aim because of FPS drops or spikes. I dont know how helpful it would be to even focus on the mobile combat aspects of this. I personally think the problem can be found, more easily, if the focus is put on stationary and/or crouched shots.

Ive been playing for at least 6 hours the last 4 days. Other than that i play off and on. For some reason my profile says ive only played for 11 hours. It said i had 10 hours played time last night, and i played for 8 hours today. Im having these issues between 0 and 4 times a map. Usually varies depending on the longevity of the match.

I started using my auth key some months ago. I think i have about 80 hours played while using my auth. I have at least a months worth of play time without logging in, and before the auth system existed. Thats at least 720 hours.

This post has been edited by vareul: 02 February 2014 - 06:45 AM

[p5yc0runn3r]

karnute, on 02 February 2014 - 03:58 AM, said:

Short: In my case, disabling dmaHD sound makes the diference to remove some subjective no-hits.

Since 4.2.018, dmaHD has been optimized for speed and thus you can re-enable it without getting no hits :)

[karnute]

p5yc0runn3r, on 02 February 2014 - 09:46 AM, said:

Since 4.2.018, dmaHD has been optimized for speed and thus you can re-enable it without getting no hits :)

[update: correction not sure the dmaHD disabling can improve no-hits, see my later reply in this thread]

Sorry, i have not explained well. I perceived the strange micro-warpings in other players (leading to my subjective no-hits feeling) also in 4.2.018 with dmaHD, so now (after some days playing 4.2.018) i have disabled dmaHD to feel it a lot better. Disabling dmaHD in 4.2.018 reduces a lot the strange effect (but not completely), my hits improved, so the subjective feeling of no-hits is reduced.

I can only speak about my case, the no-hits of other players could be different. The problem is that the strange micro-warpings i feel (much less with dmaHD disabled) do not appear in the demos, so i can only speculate about them. There could be some coupling or (de)sinchronization between sounds triggering (steps, shoots, explosions, etc), network packets, prediction, rendering, etc. that could be agravated with older hardware.
Is it possible that dmaHD tries to render sounds from events more distant than the standard sound? This could explain that in crowded servers with a lot of action there are much more computations to render a lot of sounds that could not be heard in any case (just expeculating again).

P.S.: I had only changed sound cvars: com_soundmegs=16, s_khz=44, s_mixahead=0.8, s_useopenal=1. Maybe any one of those was causing problems with dmaHD?

This post has been edited by karnute: 13 February 2014 - 10:04 PM

[FS Frankie V]

Mico-warps can be caused by a few different things.

A hitch in prediction is one of them but is software based so the number of occurrences would not be ongoing during game play and at it's worst a player could see themselves in two different positions.

What could be a factor is attempting to maintain a high 125 frame rate over the course of hours of playing that over time degrades due to factors like heat.

Here is how it works.

If you set the max cap of the game to 125 the game “will” under all conditions output rendered frames at that rate. Over time the video card will loss top end performance as it heats up (ie heat = performance loss always) and will tend to “drop” frames in an attempt to keep up even though the number states you have a fixed output of 125.

This would translate to a warping effect due to a frame or two being lost between one position and another.

If you are experiencing the detached feeling try setting you max cap to 80 and see if that makes a difference and if so then I would be interested hearing about it so I can add it to my notes.

[karnute]

Frankie V, on 02 February 2014 - 07:54 PM, said:

If you set the max cap of the game to 125 the game “will” under all conditions output rendered frames at that rate. Over time the video card will loss top end performance as it heats up (ie heat = performance loss always) and will tend to “drop” frames in an attempt to keep up even though the number states you have a fixed output of 125.
This would translate to a warping effect due to a frame or two being lost between one position and another.
If you are experiencing the detached feeling try setting you max cap to 80 and see if that makes a difference and if so then I would be interested hearing about it so I can add it to my notes.

In my case, the micro-warp effect was there from the start of playing and does not agravate with playing time. The nvidia-settings show the gpu temperature is just a few degrees above the normal (64ºC) after a lot of playing (the variable speed fan makes a good job). So probably, GPU heat is not affecting here.

In any case, I tried with com_maxfps=100 and 80, but i feel no improvement over my usual 125, maybe a tiny bit worse. With dmaHD disabled the no-hits almost dissapear and probably the only remains are my own fails to aim.

But I just found that a similar micro-warp effect can be feel with latency differences [update: see my later reply]. I tried now in a distant server (Canadian-Carnage) where i have a very high ping (250 fluctuating up) and there are a lot of players with similar high pings, and the feeling of no-hits and micro-warps is very notable, even with the dmaHD disabled. So maybe it is something related to the latency compensation or networking sinchronization with frames, but it could be a different effect with similar appearance.

This post has been edited by karnute: 13 February 2014 - 10:11 PM