householddog

I'm not sure if it can be done via free look. You can always do a similar action, though, by using both the forward and strafe buttons at the same time.

I mean it should change, either for the better or worse, when you look at it. As opposed to strafing it out of view. If there were more polys, there should be less fps, if there are less poly then more fps. As it is, there is no difference whatsoever, with a lot of buildings/objects. If its rendering things behind the object that would explain why the FPS would not change. Its not an entirely conclusive test, I agree, but interesting, non the less.

Agreed razorman, but I did this in an empty mission. Thee should have been nothing going on at all. OK so long as we are all in agreement with what I said, (its not really finished yet). I am happy. Finally, don't most flight sims use huge draw distances? Or am I being a "pseudo 3D-engines expert" by making such an observation? Comparing an ID engine game to ARMA is an apples an oranges test. QUAKE maps are drawn in small areas and crammed with objects. The maps are designed to be small with a lot of details. Arma maps have much less detail. To do a propper test, on anything, you need to change one thing and see the result. IE you need to put an ARMA2 map in quake or a quake map in ARMA2. Not so easy. My main argument is to do with hardware scaling, not distance scaling btw, IE By the way. I am not the only person who thinks it doesn't scale well. Its the equal 3rd problem on the bug tracker. http://dev-heaven.net/issues/show/1770

Quick question, for the devs. What is going to be fixed/improved in the next patch? I am just wondering what is being worked on, generally.

I think I forgot myself. :) Just wondering what we know about the next patch?

Never harms to ask. Lock away then. ;)

Ahmed I think it might actually be ram speed that is the big determinant as well. The I7s all run 1300mhz+ ram speeds. There are a few people with poor specs but fast ram, that are getting good scores. I also achieved my highest rating by using fast RAM. Because the dividers on my mobo don't support it though, I could not push it very far. I got a 7% increase going from 800 - 940. I just need someone who can really declock their ram to test it better. I upgraded my processor from a dual to a quad and only got a 10% increase in performance. Sorry nate it was me, not you. Not reading straight tonight.

When you mentioned scale, it go me thinking about a test I forgot to do. I dropped my view distance from 1700 down to 500 in the settings. This netted me 12 extra fps. From 40-52. This is a massive reduction in the amount of objects it has to draw, but nowhere near a corresponding increase in FPS. Try it yourself. It shows that it is not the size of the environment that is the problem. There is something, fundamentally holding back its performance. If you reread my original post on this topic. My advice is to wait for the patch. Maybe see what OPFP2 is like. I didn't say buy it. I just gave my opinion on what seems to be wrong with ArmaII and what ArmaII Also, if you guys are correct, about the game being optimised, then I would expect no performance tweaks in the next patch. Correct? If I see one single performance increase, from the next patch. I am going to come back and rub you nose in it. ;) :) Of course, if it doesn't, this post will mysteriously disappear. :lol:

Sounds good. Can you do me a favour? Can you lower the clock on your memory and run the bench again? Just your ram not your cpu. I am wondering what effect ram speed has on performance. Cheers.

Yes but then you would expect the fps to change if it were the polys. It pretty well stays exactly the same, when you look at it. At the very least there should be sweet FA rendering going on. Obviously the game is culling renders out of your field of view. You need only look up at the sky to see that. Its simply not using occlusion culling, in a lot of places. I wonder if it still renders everything on the other side of a hill too?

I don't really think the ballistics are going to be a major issue. They are certainly not difficult to calculate. They have been calculating them for over 100 years, with pen and paper. Any coder could program it themselves with little difficulty. Thats not really what an engine is designed to do. An engine is a shortcut on the environment, physics and modelling that speed up development. It helps the developer create software quicker by having routines that provide a basis for their project. It means that the coder does not have to write, from scratch things like gravity, shadowing, occlusion, animation, etc etc. Generally the designer can easily change the settings. They could put in gravity similar to the moon, for instance. These are reasonably easy things to change. They can always code the things that the engine lacks. Or adapt its routines for a different project. The developer is not entirely limited to the engine. The efficiency of the graphics engine would be of most importance, to arma2 and ofp2. Mainly due to the large draw distances. All I know is that holding onto legacy systems for development, hardware, networking or anything computer related, often hampers progress and performance.

Its bizarre. I am having an argument in another thread with somebody who thinks Arma2 scales well to hardware. I really don't see it myself.

Yeah I have been doing that, but there is still a lot of HDD activity.

What you are saying is that you don't need performance? So playing at 20fps is better than playing at 35fps? All else being equal. That is a fairly poor argument. If somebody is reading this, wanting to know if they can buy the game. If the only argument, for poor performance, is that you don't need good performance, you are going to turn people off. You are also telling them, that Arma2 is not a game that performs well. Crysis is a good example of software that scaled well. Pretty punishing on hardware, but you could easily tone down the settings and have a decent game. In fact, the game detected the hardware for you and made the settings itself. Most games these day, do. In Arma 2 its not that easy. For instance, people report, not getting any performance increase, from lowering the resolution. Its hard, to find a game, where, lowering the res, does not increase the fps.

Is Virtual memory just referring to the size of the swap file on the hard disk? Or is it something else? I am not sure, how it helps, if its just the swap file. Is there any way of disabling the sound completely. I mean totally killing it, so the game doesn't play it (ie not just turning it down to zero)? I just want to see how it affects FPS.

Yeah, as I said before, I tried it. No noticeable difference. ATI Tray tools does it for ATI cards.

I really think the engine is pretty poor. An engine is not about how good a game looks, per say. Rather how much performance you get with those visuals. It's not particularly efficient IMHO. It scales very poorly, most of the main settings seem to have effect on performance. The only that seems to have any effect is the fill rate but its impossible to fine tune it to any degree. There is little to no culling in cities, which would really help with fps. I am not having a go at the game as such. Just, so far, I really can't recommend it, in its current incarnation. Quite likely, by the 505 release these things will be fixed and optimised. I hear, a lot of posters, saying Arma2 can't be optimised further. If this is true I couldn't really recommend it to anyone, without a top notch, computer ever. I would say to these people, you are really not doing the game any service by saying this. In effect you are telling people their computer's will never handle it. Not good for promoting a game. But if the question is; what is the difference, between a game, that isn't finished, as opposed to a game that hasn't been released? My answer is exactly the same as yours eric. Let's wait and see how they deliver... Honestly I like Arma2. I wouldn't have just forked out A$700 to upgrade my computer to play it, if I didn't like it. It's just that its not finished, something that I hear, second hand, BI agrees with. Roll on 1.02 ;)

ArmaII seems to have a fair bit of the modding community behind it.Which is a plus. Arma 1 and 2 are sort of cult games. They have a small, but very strong following. To be honest the actual engine on ArmaII pretty well sucks. People are calling on more improvements to physics etc. I just don't know where the processing power is going to come from. The thing already punishes the best PC setups. ARMA 2 has a higher theoretical MP limit. Arma 2 comes with the editor shipped. Arma 2 is a fairly easy port (from what i hear) from old Arma I OFP DR claims to have a more destructible environment. OFP DR, claims to, allow you to enter all the buildings. You can put attachments on weapons etc. I think the OPFP II is going to have much better physics, for a start. Its probably going to have better performance too. That simply comes from having a much newer engine, that has been proven on other games. In particular the physics should be top notch, as the engine has been used in car racing games. If you are wanting to see which one to buy. I would wait for OFP DR to come out. Hopefully ArmaII, will get some better performance, and you can see what the alternative is like. PS BI is thinking of releasing Arma II on console aren't they. The whole console argument is a little bunk, if this is the case.

Just changed it to faster ram. The old ram was 800 the new one 940. In fact the faster was smaller than the old. The faster ram was 2GB the slower ram 4GB. PS Bags credit for finding out about ram. ---------- Post added at 02:54 AM ---------- Previous post was at 01:36 AM ---------- I really wonder how accurate this benchmark is. The HDD activity is immense. possibly it could be slowing it down?

Hi guys, have done 2 tests one with my old CPU e6750 and one with my new CPU q9400. My specs . Texture Detail - Normal Anisotropic Filtering - Normal Terrain Detail - Normal Objects Detail - Normal Shadow Detail - Normal PostProcess Effects- Low Cpu - E6750 OC (3.2ghz) and Q9400 OC(3.2ghz) Ram - Kingston 900 GPU - Gigabyte HD4890 OS - XP Resolution - 1280x1024 Normal Score E6750 OC (3.2ghz) : 3375 Normal Score Q9400 OC(3.2ghz) : 3696 So much for it being CPU bound and/or supporting quad core well. I will test the quads now by setting the ArmaII affinity to only 2 cores, see if it drops the performance much. I have a theory that the memory is really the thing holding it back. It would explain why I7s get such high scores, yet the lower quads don't scale up. I have some 1024 memory lying around- might give it a go - its faulty though and only 2GB. Tried disabling 2 cores - 3232. ---------- Post added at 02:03 PM ---------- Previous post was at 01:00 PM ---------- OK I put in 2 GB of memory clocked to 940. Normal score : 3800 Might be onto something here. A fair increase for Going to push it to 1066 and drop the clock (Mobo doesn't have sufficient dividers) Might be able to push it to 4000?

I tried out the same experiment in COD4. It doesn't seem to give any frame rate increase when stepping out from a wall. Its possible it could be due to the insane frame rates I am getting (250fps) I am not really sure. I played around with zbuffer settings using atitray, it really didn't seem to make much difference. ---------- Post added at 07:51 AM ---------- Previous post was at 07:48 AM ---------- Dunno if its really updated either. You can pump huge renders, polys etc even into very old 3d engines. You could even try replicating this Arma on the old DoomI 2D engine. It just a matter of getting very inefficient result. IE, what we are seeing now. ;)

I had a bit of a play around with this in the editor. The method I used was to hide behind corner, then strafe into the open slightly. Keeping my orientation the same way. What I found was interesting. Some buildings' walls do actually raise my fps higher by around 10-15 frames. Other ones do not. Particularly wooden buildings, I believe do not increase your fps when they block your sight. Concrete ones do not, as far as I can tell. Some sort of penetration/los madness? Not sure yet. Just an early observation. You guys/gals, might want to try it, and see if you get the same results.

If you went for a driving test with that kind of eyesight, I wonder if you would get your license. Seriously though. I agree the blur should be turned off. If I am having framerate issues, its a lot easier to diagnose, if the blur isn't there.

No worries. Good luck!

Blood I was just wondering whether you have run a memory checker like memtest86+ on your machine. I had problems like this a few weeks ago with Steam. Turned out to be a bad stick of ram. It was fine on small downloads, but anything over 2 GB and would get corruption. Generally speaking the error correction on FTP downloads is pretty good. Its very rare to get corruption from your connection. When it comes to error correction on (non server) PC ram, its non existent. I particularly ask this because you say you formatted a couple of weeks ago. This sounds like you had problems, then fixed them. Bad ram can cause corruption in XP as well. Possibly this could explain your need to format XP? Just something you might want to check before further frustration sets in. Heres the link to memtest86+ http://memtest.org/ Cheers