ARMA 2: OA beta build 77007

[ast65 10]

The white pixels in certain LOD´s in OA vegetation are really annoying, they´re around since about 1.57 or short before.

I have those only with ingame AA on, though, so I´m using ATI´s MLAA postfilter instead.

With adapted settings picture quality is very good, too but I´d love to use ingame AA.

Please try´n fix BIS :)

[-=seany=- 5]

The white pixels in certain LOD´s in OA vegetation are really annoying, they´re around since about 1.57 or short before.

I have those only with ingame AA on, though, so I´m using ATI´s MLAA postfilter instead.

With adapted settings picture quality is very good, too but I´d love to use ingame AA.

Please try´n fix BIS :)

Yep its bloody irritating all right. Vote to have it fixed and if you have anything to add to the discussion please do :) :

http://dev-heaven.net/issues/15668

[ast65 10]

voted ;)

[soul_assassin 1745]

hmm first beta ive tried for a while and it seems that I started to see alot more z-fighting in takistan building ruins. Been playing almost every night with 1.56 before and never noticed it.

[Tankbuster 1735]

Yeah, Z fighting quite bad, especially on ruined buildings, also tree LODS never loading up. Seeing very poor tree graphics.

I also saw the white pixels Alex72 mentioned.

[grillob3 11]

PogMoThoin

Leave the localVRAM=; nonlocalVRAM=; that gives you best performance in my (case localVRAM=1063515776; nonlocalVRAM=2147483647; gives me 2 to 3 fps more than localVRAM=1073741824; on a trifire 5970+5870. i dont know why!lol) and leave "read only" so the game does not change that for you anymore!

[PogMoThoin 10]

PogMoThoin

Leave the localVRAM=; nonlocalVRAM=; that gives you best performance in my (case localVRAM=1063515776; nonlocalVRAM=2147483647; gives me 2 to 3 fps more than localVRAM=1073741824; on a trifire 5970+5870. i dont know why!lol) and leave "read only" so the game does not change that for you anymore!

Whats the point in read only? the game only writes it there, never reads

http://forums.bistudio.com/showpost.php?p=1632766&postcount=71

[kklownboy 43]

Well the "read only" is only good to keep a Refreshrate that Win7/Vista may have issues with on a CRT....I never found any benefit to Read Only, for the RAM amounts.

[[kh]jman 49]

Seeing very bad tree and foliage LODs here too. I have to get very close (maybe 1 metre) before they will switch to the correct LOD and quite often they don't even do that. For me this issue has been around for a while now, since around 1.54 but much worse now. Up until then not had the issue at all. My hardware setup has remained constant and defrag once a month or so. Used to be worse on Takistan over Chenarus but Chenarus is quite bad now since the latest offical patch and latest beta. Reducing viewdistance makes no difference.

Edited January 6, 2011 by [KH]Jman

[suma 8]

Gnat;1827997']Similar to metalcraze comments' date=' the AI pilots don't seem to be doing anything differently, except;

- They tend to do a lot more "dry runs". ie An attack run, but no bomb drop or missile launch.

- They now at times do a huge grab for air / pitch up, then almost straight away do a steep dive. Seems to happen around actual ammo launch.

The main problem is all AI pilots nose-down to attack their target, even if the weapon doesn't require nose down.[/quote']

Please, provide an update repro for this in the http://dev-heaven.net/issues/16219#change-81135 so that I can check it.

I am sorry, I am unable to provide technical details, the fix is based on many minor fixed and adjustments and the description would be too long. If you will provide an accurate feedback including repro steps how to see what you do not like this time, I will revisit the issue.

[metalcraze 290]

Jman;1828564']Seeing very bad tree and foliage LODs here too. I have to get very close (maybe 1 metre) before they will switch to the correct LOD and quite often they don't even do that. For me this issue has been around for a while now' date=' since around 1.54 but much worse now. Up until then not had the issue at all. My hardware setup has remained constant and defrag once a month or so. Used to be worse on Takistan over Chenarus but Chenarus is quite bad now since the latest offical patch and latest beta. Reducing viewdistance makes no difference.[/quote']

Try setting Video Memory to Default. If that doesn't help - try setting it to Very High.

Please, provide an update repro for this in the http://dev-heaven.net/issues/16219#change-81135 so that I can check it.

I am sorry, I am unable to provide technical details, the fix is based on many minor fixed and adjustments and the description would be too long. If you will provide an accurate feedback including repro steps how to see what you do not like this time, I will revisit the issue.

I'll try to post an example mission in the evening that should showcase the issue with dry runs.

[Muahaha 10]

Hi Suma, don't mean to rush... are you guys planning to resolve this bug with official missions?

http://dev-heaven.net/issues/13530

[WillaCHilla 10]

Jman;1828564']Seeing very bad tree and foliage LODs here too. I have to get very close (maybe 1 metre) before they will switch to the correct LOD and quite often they don't even do that. For me this issue has been around for a while now' date=' since around 1.54 but much worse now. Up until then not had the issue at all. My hardware setup has remained constant and defrag once a month or so. Used to be worse on Takistan over Chenarus but Chenarus is quite bad now since the latest offical patch and latest beta. Reducing viewdistance makes no difference.[/quote']

Can confirm this. Also for me with the newest beta, some trees at Takistan show this behaviour. Bushes, etc seem to be okay like in the previous ones.

Try setting Video Memory to Default. If that doesn't help - try setting it to Very High.

Normally the 'Default-setting' is the one I'm using with my current rig without any kind of those problems since a long time. 'Very High-Setting' will give me some stuttering after some time. Will test the behaviour with it though when I'm back from work.

[SaOk 112]

Good beta again, but I am seeing AI shooting much empty enemy (light) vehicles. Also the units with (commandstop this;)-enabled may start moving after some minutes in firefight. This didnt happen few betas back. I have some AI groups positioned on roof and now some of them commiting suicides. Making them all have their own 1 man groups could fix that, but it would make the mission clumsy in other areas.

Edited January 6, 2011 by SaOk

[[aps]gnat 28]

Please, provide an update repro for this in the http://dev-heaven.net/issues/16219#change-81135 so that I can check it.

I am sorry, I am unable to provide technical details, the fix is based on many minor fixed and adjustments and the description would be too long. If you will provide an accurate feedback including repro steps how to see what you do not like this time, I will revisit the issue.

Thanks Suma. Below added to BugTracker, with demo missions.

Edit: I should have clarified why "AI diving behaviour is still unsuitable". One simple reason is any Laser or Free Fall munitions larger than a FAB250 WILL damage or kill a plane on a pass.

Another; AI continuely wait to launch as much ammo as possible all the way on the dive to the target. They will miss on targeting (because of "last-second" launches) and also kill the same target twice.

After more structured testing, I can conclude it has been improved with the 77007, but the AI diving behaviour is still unsuitable.

Attached ArmA2/OA test missions on fairly flat desert_E;

Test1 - Su25 vs APCs - Armed with only 8 FAB250

Test2 - Su34 vs APCs - Armed with only 8 FAB250

Test3 - Su25 vs APCs - Armed with only 8 Ch29

Test4 - Su34 vs APCs - Armed with only 8 Ch29

AI never really do the same each time (which is a good thing really). ie only average repeatable tests.

LP= 1.57.76815 Beta= 77007

Observations;

Height of aircraft over targets on pass (after first);

- LP range between 30 and 80

- Beta range between 45 and 90

- Seems Beta planes approx 10-20 meters higher on pass over target

Death from own Bombs (at impact zone);

- LP happened several times during tests

- Beta, no occurances

"Dry Runs" or attack runs with no launch;

- LP did have a few

- Beta seems to have a few more occurances (but not a big problem at this stage)

Bombs dropped too early and fall short

- Low occurance. Seems to be same both LP and Beta

Missiles launched too early and fall short

- LP, most Su34 launchs failed to reach target

- Beta, no issues.

Dont remember observing in LP, but with Beta did see just a few time AI climbing BEFORE reaching the target, after launch.

Would like to see MORE of this.

Cant explain very well, but the beta seems to have the AI adjusting their altitude a lot more.

Observed;

- Sometimes a big climb when AI first see target

- Sometimes a climb and then a dive for no reason after passing the target.

- Sometimes a climb and then a dive straight after a bomb is launched.

These adjustments seem to not always be logical and not always be an advantage for the AI.

NOTE: These were flat landscape tests, to guage better true altitude.

Its likely the same tests should be done "in valley" to see differences (on my long To-Do list)

My impressions. Others can comment.

I know the BIS team *might* consider this issue closed. But I'd ask them to consider more work in this area because we have plenty of aircraft in this game that are not "direct ground attack" aircraft, and modern guided weapons and so would benefit from the more "modern" attack envelope of guided weapons.

I guess that may lead to a need to dramaticily increase of the angles in the target acquisition cone.

Edited January 6, 2011 by [APS]Gnat

[NoRailgunner 0]

Can someone confirm that AI refuses to shoot at empty vehicles despite a given fire command? Imho the player could order his AI team to fire at any targets and they did it... (even ArmA1 had this as tutorial mission) :confused:

Would it be possible to get the AI use some stuff via direct action command eg:

select AI soldier - select from action menu "use Laserdesignator" or "use AT weapon" or "use AA weapon" etc ?

So in result the AI would follow the orders of human team leader more consequent.

Other example:

order the AI to switch off their lights on weapon - within a second they switch it back to - lights on.

Possible solution - let all orders given by player override the AI fsm at least in stealth/danger behaviour. Plus let these orders generally last longer than few seconds. :)

[eddie247 10]

3 times in 2 days , climbing into btr60 has crashed my game while using beta patch... I will look at devheaven later

[Caillin 0]

After getting very frustrated with the amount of stuttering on OA from texture loading, this beta patch has almost completely fixed it (particularly zooming with scopes). I can only assume that I was definitely getting affected by the misreporting of the vram item.

[Coffeecat 10]

I just tested Eagle Wing, its also broken like many other scenarios.

Killed Ikarus and waited for the Nuke Cutscene then your chopper is slowly gliding too the ground, crashing. Your dead and can suceed the mission:(

[ast65 10]

Strange, I just replayed EW from beginning to end with this beta without a glitch.

Gotta admit, I had that bug you´re talkin about, too once a time but with another beta... :j:

[[aps]gnat 28]

...... not sure, but when did the AN-2 start bobing and stall&crash when activating ILS landing?