Anyone with problems with ATI drivers read this.

[Flanker35M]

S!

Using Effects=2 is not recommended even by Oleg as he said it causes slow performance and have artifacts especially in smoke. Can be flashing etc. I use Water=0 because of online play. I am not there to stare at wave tops but shoot down, or evading others most of the time

Blue hue is tied to UseVErtexArrays. If not using it, then no hue either. The VertexArraysEXT has no effect whatsoever.

DB605, getting back to you at Hades site.

[=PF=Coastie]

Hi guys,

Terry Makedon(head of the ATI Catalyst driver program) replied to my email and said this problem will be looked into.

As for the atiogl.dll, you guys should be using the 9.10 file (nine point ten) not the 9.1. Especially if you are using a 5xxx series card.

I am currently still tweaking my settings with my brand new rig (5770/amd Phenom II x 4 955BE/win7). I plan on creating a full tweak thread about my findings soon.

Real quick though, I am running 1920x1080 with 4x SSAA(Supersampling AA gets rid of the crawly rivers in the distance!), max Aniso, perfect, with water=0 and I am getting 62fps avg, 27fps min in TBD track.(V-synch off).

I have also found a work around for the Blocky text issue if you do not have the 9.10 atiogl.dll file.

Start the game/go to tracks/choose The black Death track/let play until the text clears at exactly the :21 second mark. Close track and you can play all you want with no more blocky text as long as you do NOT close the game. You can go online using direct IP too. But if you use HL, this will not work.

More later guys, Thanks!

[maclean525]

You are correct about the driver version, I looked at the version number which is: 6.14.10.9026 this is associated with the 9.10 catalyst release.

[dduff442]

9.1 and 9.10 are (how you say) the same number?!

Flanker's tip about disabling UseVertexArrays is spot on. Checking back, I find my original settings were based on a post of his at the ATI forums. Small world, eh?

Water=1 or higher still kills my system. I was able to improve performance a bit by setting the priority for the Il2fb process to 'high'. If you try this and it messes up your sound, just open Hardware>Sound Settings and click 'Apply and Back' which will reset your sound without changing anything else.

I actually overclocked my i7 to 3.2GHz and got a small improvement but decided it wasn't worth it and reset it to the default values. Fast core-duo machines will beat i7s in Il2 because Il2 only uses one core anyway.

[=PF=Coastie]

If you think of it as a decimal, yes, it is same number. But when describing ATI's driver versions, that is a big nono and can cause big problems. Think of the decimal as a simple separator for two whole numbers. The first number signifies the YEAR and the second number signifies the MONTH. The 9.10 drivers were released in OCT 2009 and the 9.1 drivers were released in January 2009, ten months before the 9.10 drivers.

Hope this helps you understand the ATI driver numbers now.

[dduff442]

Originally Posted by =PF=Coastie

If you think of it as a decimal, yes, it is same number. But when describing ATI's driver versions, that is a big nono and can cause big problems. Think of the decimal as a simple separator for two whole numbers. The first number signifies the YEAR and the second number signifies the MONTH. The 9.10 drivers were released in OCT 2009 and the 9.1 drivers were released in January 2009, ten months before the 9.10 drivers.

Hope this helps you understand the ATI driver numbers now.

Fair enough. The file attached to the earlier post is the correct 9.10 version, though.

[Flanker35M]

S!

Coastie, I did file a Feedback Form for Catalyst team explaining the issues. But better now that you got Terry's attention. Thanks a ton! Hopefully ATI will take a look in this. At the annual virtual pilot con we discussed the IL-2 and how it performs. We did extensive track running etc with logging and checked results. It seems code of IL-2 has errors in some values in it, but hard to tell what without more troubleshooting capability.

Even IL-2 age is showin a small bit of attention could be given, to at least address the support for newer gfx cards. IL2 is so far the only game I run that has problems. Others I play, like STALKER, Rise Of Flight, Aces High 2, Mass Effect 2 etc. Have NO issues at all..go figure.

I am planning to go for AMD cpu just for fun, using Intel i7 920 atm with the Intel DX58SO mobo. Just need to see if the AMD can use DDR3 yet

[Qpassa]

good news,thanks coastie

[Flanker35M]

S!

Did some testing last night with Catalyst 10.1 Hotfix drivers and different DLL-files of IL-2 floating around. To see which ones would give best image quality and performance. Quite surprising results as average framerates were within 1-2fps between ALL of them with same settings applied to the graphics card. The most recent dll-files caused terrain to flicker and corrupt, but setting Catalyst AI ON solved that. Catalyst AI just has a negative impact on FPS, it drops and noticeable in heavier scenes of The Black Death. But average FPS was with ALL 53-54FPS at 1920x1200 32-bit, Perfect etc. Will post my test setup later today for reference.

A few pointers though. If using AntiAliasing EdgeDetect with 4X and AA Mode at Performance, there is only a very small hit on FPS, maybe 1-4fps. So a very well implemented thing. With Adaptive or SuperSampling almost the same, maybe slight stutter. But need to run more tests to verify this. Anisotropic filtering, if application controlled, gives worse image quality than overriding it with CCC. IL-2 seems to use 2x Anisotropic only. I ran tests with 16x and there was a difference in the water shimmering.

Catalyst AI is a bit complicated issue. Some DLL's required it to be at Standard at least to get normal graphics(terrain only). Some dll's worked normally and faster without it. Catalyst AI at Advanced caused water to look like it was "pixelated" or covered with small squares. So I settled with a set of dll's that gave best image and performance without Cat AI.

The variation of error messages in console with different dll's was wide. Some gave a lot of them, most only the clamped error now and then. So a set of new dll's for IL-2 would clearly help users to get more out of IL-2, but that is up to TD/MG to decide Test results with settings and conf.ini later today..still at work..meh!
Coastie, can you keep us posted on ATI Team's responses on the driver issue? Would be great to see these issues rectified.

[=PF=Coastie]

I rcvd an update from Terry and they are working on it. We will need to be VERY patient with this guys. As long as e have the 9.10 dll's we are ok. A lot of testing will be going into this as well as making sure nothing else is broken in the driver after making adjustments.

I will test some of your findings Flanker. I like what you have come up with so far. But I disagree with your assesment on the in game aniso.