Jump to content

Cameroon

Members
  • Posts

    889
  • Joined

  • Last visited

Posts posted by Cameroon

  1. This is at the launch of the game? For me, that has to do with setting AA/AF in the Nvidia control panel and is cosmetic only; clicking brings up the menu screen.

    Maybe try setting the Triple Buffer option in your Nvidia control panel? I thought I saw someone mention that somewhere as a fix to the intro movie not showing up, maybe it would help your problem as well?

  2. Originally posted by Battlefront.com:

    Cameroon,

    It would not surprise me to find out that whatever the problem is with G80 is found in the G70. Hmmm... actually, do you have Vista? From what I was reading earlier today Vista really brings out the worst in this problem.

    Steve

    Nope, not Vista - WinXP SP2. For awhile I was using the 94.24 drivers since they gave the best performance (going to try them again tonight if I can), but the 163.67 drivers appear to work about as well for me. Inbetween versions suffered from massive FPS penalties.
  3. Originally posted by Battlefront.com:

    Yes, the Alt-tab thing shows conclusively that there is a bug in nVidia's drivers specific to memory handling as has been reported on other websites. Hukka is spot on when he suspected there is no way one should get a FPS boost by Alt-tabbing if the nVidia stuff was doing its job correctly.

    There are other problems beyond this, though. We're pretty sure of that. Some may even be ours, or at least there might be things we're doing that make things works. But fundamentally this memory handling thing is probably at the root of all 8800 problems. We're also wondering if the people that have "no problems" with 8800 cards may find that their performance goes up once this problem is fixed.

    Steve

    But does the same apply for those of us with 7950s and similar? I had started to write a different post and ended up with the reply to Hukka and accidentally left off that I've got a 7950 GT w/ 512 MB of RAM.

    I get playable framerates, but they don't change no matter what settings I turn on/off in the control panel. Campaign mission #2 and Allah's Fist both get ~20 FPS pre-alt-tab and 24-27 after with just about any combination of AF/AA, screen rez, etc.

    I wouldn't be surprised if the G80 memory bug is affecting the G70 series as well and I figure there's a good chance that any improvements for the G80 series will trickle down for me. I also figure it's good to get information indicating that it isn't just G80 owners that are getting performance changes from alt-tab or similar tweaks.

  4. Originally posted by Hukka:

    Yes it appears that the 8800 GTS texture memory bug has a big part on this whole mess. I tried this at Allah's Fist: I moved the camera right and left until the fps dropped to around 15 when the camera was still, then I alt+tabbed and voila! fps raised to around 32.

    So I guess that the fps shouldn't drop to around 15 in this kind of test at all without the bug.

    edit: I did some more testing and now I'm quite sure about this. Altough it's kind of hard to explain...

    I go to the blue edge of the map. I am at the center of the map. I'm at the zoom level 4. Then I tilt the camera to the far right and far left, my fps drops to around 15 and stop my camera and look straight across the map. fps is still 15.

    The I alt+tab and boom! my fps raise to 32 and if I tilt the camera a bit to right and left, I still got the 32 fps. But if I tilt to the right extreme and left extreme and look straight after that, my fps is 15.

    I've been poking around with alt-tab and it appears that it raises the FPS because CM ends up shortening the draw distance for 'doodads' (2d graphics of grass and such). Try this:

    Do the alt-tab and record your FPS

    Now, use shift-] to increase from Balanced to Better and record your FPS

    Use shift-[ to go back to Balanced and check your FPS again

    If it's like my experience (7950 GT w/ 512 MB), you'll be back at yours starting FPS.

    Run through the procedure again and you'll probably notice after the alt-tab that the grass isn't drawn as far out (again, if it's like my experience).

  5. Grrr - reading this here at work makes me really wish I were home playing CM smile.gif

    We're all getting hit by the QB troop purchasing (Mishga and I were discussing it in the tech support forum) and it's really unfortunate since it does make a good QB that much harder to come by. Still, I'm looking forward to getting home sometime tonight and trying out the first volume. Thanks for all the effort everyone! smile.gif

  6. Sadly I just ran through a bunch of impromptu testing with hotseat QBs in 1.03 and here's another classic:

    still_goofy.jpg

    The attackers have 43 regular conscripts, the defenders have 131. That was with a 20% force reduction for the attackers.

    The force picks were ok a couple of times, but would also come up with force selections like the above. I'm ok with not purchasing my units if the computer doesn't do stuff like that (note: both forces were Syrian Light Infantry and were conscripts)!

    [Edit]

    To be clear; I don't mind difficult battles that are not "even", but in setups like those illustrated here there's not even a remote chance for one side.

  7. Yeah, it seems a lot more reasonable with a -30% modifier. Probably something that needs looked at though, I should probably go find one of the QB threads over in the general forum or here and add that bit in (if it isn't already).

    [Edit]

    Hmm... actually -30% might be a little too drastic, I'll have to try -20% and see how it goes.

    [ September 08, 2007, 04:20 PM: Message edited by: Cameroon ]

  8. I just finished playing a Syrian v Syrian QB and attached is the screenshot at the end (I was on the defense):

    too_many.jpg

    My question; does it seem to you like that's WAY too many attacking infantry? 256 attacking v 52 defending or just about 5 to 1. The scenario was (supposed to be) a probe and both sides ended up being conscript Syrian infantry.

    Anyone else seen these kinds of really outrageously mismatched numbers?

  9. Originally posted by c3k:

    Allow me to ask a question.

    In the movement panel shown in the screenshot, the various movement options have certain colored tabs, yet the movement line as plotted in the overhead view only seems to match one of the colored tabs: is this correct?

    I would rather see the movement orders plotted on ground to match the colored tabs.

    Perhaps because they were all "MOVE" orders? :D
  10. Originally posted by sandy:

    I think that the problems highlighted in this thread indicate just one more example of why, so far, and IMHO, CMSF seems to be a step backwards from CM1 in actual gameplay.

    If these issues are bugs which can be fixed by patch 1.12, or whatever, all well and good.

    My worry is that they may be the result of ill-chosen but fundamental design decisions...

    I certainly never saw anything like this in many years of playing CM1

    No comment by BF I note.

    Then you never had your tanks shot through a building in CMx1? There were plenty of times where an AFV had LOS (and thus LOF) through the corner of a building in my games, so that engine had its LOS/LOF issues as well.
  11. Originally posted by handihoc:

    PS I installed the new Nvidia drivers, but sadly they've not improved the mouse lag/camera jerk.

    Have you tried the 94.24 drivers? I don't know what card you have, but I get a massive, positive framerate difference with 94.24 compared to any of the recent releases (even the beta 163.44). As in 10-15 FPS.

    Also, it seems that it now actually is good to have the affinity set to both cores on my C2D :D

  12. To throw more info into the mix:

    Intel C2D @ 2.23 Ghz

    2 GB RAM

    Nvidia 7950 GT w/ 512 MB

    Windows XP SP2

    Here's a framerate comparison, just with driver changes:

    Campaign Airport Attack (2nd or 3rd battle)

    163.44: 20-21 FPS w/ shadows off, no AA/AF from starting viewpoint

    94.24: 35+ FPS w/ shadows off, no AA/AF from starting viewpoint

    I'll be sticking with 94.24 for the time being, NVIDIA definitely broke some part of their OpenGL path.

    Battlefront: I'm more than willing to help out with tweaking settings, changing drivers, testing builds, etc. but will be out this weekend.

  13. Originally posted by Becket:

    Do you guys have enough data about the bug in Quick Battles where units are placed outside of their setup zones?

    Is that the one that's caused by not having a setup area for the AI plan(s) painted (which is not a bug so much as a needed QA check on the published QBs) or something else entirely?

    I'm not referring to creating the "Setup Zone", but actually painting in the setup area for the AI's plan(s).

  14. Originally posted by PrezCartman:

    Hukka,

    It is very easy to punish for destroying buildings. Generally the US is going to have more firepower, not to mention the Javelin, and be more willing to bring down buildings.

    I am actually developing a scenario right now with a limited US attack against a mixed terrain area. The major parts include a small airport, family farm, boarding school, and former Syrian base. Because at some point after the war is over we want the civilians to be able to move back into their old lives without trouble the US commander is penalized for any building damaged except the Syrian military complex. What you can actually do is select a group of buildings, assign them a total point value for preservation for the US side and each building damaged will take away a certain amount of that overall point total. The Syrian don't have enough firepower to really damage any of the buildings so there is no reason to set it up to penalize them for destruction, plus they are largely irregular troops and hence care more about inflicting pain on Americans then losing a few buildings.

    I think that route works well, unless you're playing a scenario against a human who knows that it's worth his or her while to drop a particular building or buildings. That's why I'd like the enhancement I described above - it's side specific and so can be asymmetrical even if the other side "doesn't play fair".
  15. Originally posted by Cpl Steiner:

    </font><blockquote>quote:</font><hr />Originally posted by Raptor 2101:

    is there no ability to give both side the simelar objectiv.

    if both sides has the object "dont destroy building" all is fine...

    Raptor,

    I'm not sure if this is possible because I don't think the game considers who actually caused the destruction. If both sides get negative points for the objective, destruction will just result in a draw. </font>

  16. Originally posted by Martyr:

    A quick note:

    Back in the CMx1 days we plotted exact steps for a vehicle to follow. In CMx2 things seem to be different--waypoints are more about telling the unit where to go than about how to get there. You see this when you plot a straight line path for infantry, but they follow a meandering course in order to stay in cover on the way. Naturally this takes some control out of our hands.

    I think we'll get used to the idea, although I'm not saying it's perfect.

    Agreed, except that the problem is that we do have to tell the AI where we want it to go pretty much exactly. That's even been flat out stated by BF that if we want the AI to go a certain path, we have to plot it.

    That's fine, except that then the movement between two waypoints needs to be refined so that the vehicle makes an actual curve rather than just a turn. In CMx1, if I wanted to make the vehicle retain a high rate of speed, I would plot out a nice wide arc. In CMx2, because the AI doesn't drive a straight path between two points (because it looks for cover, etc.) that will end up backfiring.

    Getting a smoothly interpolated arc through a waypoint would, it seems, pretty much fix the problem with units overshooting waypoints and/or making ridiculously slow turns.

×
×
  • Create New...