Jump to content
COMBATSIM Forum

Tigershark

Members
  • Content Count

    146
  • Joined

  • Last visited

Community Reputation

0 Neutral

About Tigershark

  • Rank
    Captain
  1. Tigershark

    CTD

    Braking works better if you don't have to sip fuel. I have a hold-down switch for airbrake on the throttle, which makes it easier to use. You can get buildings from a decent distance if you keep a straight trajectory and remember to aim low, not high. High diving angle helps.
  2. "Vector X for Y" means go to X degrees clockwise from North, for Y miles. Look up your heading on the map or the HUD. To maximize your range, fly high, 30-35 thousand feet is the typical real-life cruise altitude, it also works well in-game, and throttle back to about 80%-90%. Actually, the game will even say "cruise" when you're in the range. In-game this cuts down on fuel consumption a lot, allowing the plane to travel more distance than it ever could. In campaigns, edit your missions for best results.
  3. In-game's AIM-120R is loosely based on the same prototype that IRL will be known as AIM-120D. Note that in-game 120C is also not quite the same as its IRL version.
  4. I've got a new cable for my Saitek, and it seems to be working better. Waiting in Air Combat on SimHQ in case someone decides to show up.
  5. That information is very old. Fortunately it does suggest to refer to the product page. Maximum temperature for GTX 570 is 97C: http://www.geforce.c.../specifications The last card that had 105C was the GTX 400 series. For GTX 570-580 they downrated it to 97C (though it's actually a bit higher up to 102C for low-end cards). Notably, the silicon didn't even change between 480 and 580 - a temperature of 105C is safe for the chip itself, so it's not the reason for downrating. Like I said on the first page, all high-end Nvidias suffer from solder cracking, have been since 6800. The problem is delayed but very common, up to half the cards eventually go that way, and caused by repeated thermal stress, eventually breaking some of the solder balls connecting the chip to the PCB away from one of the surfaces. This loss of contact results in visual artifacts, instability, overheating or outright failure to start. Cards with this problem can actually be fixed with about 2/3 success rate by removing electrolytic capacitors and plastic parts and heating the card with a hot air iron and an oven to resolder the chip. Nvidia is well aware of the issue, but apparently it isn't easy for them to fix (they don't own any manufacturing themselves) and they aren't concerned enough to change their contractor (it usually happens after the warranty period and is easily repaired anyway), but enough to lower the throttling temperatures. The incidence of this problem depends on operating temperatures, and the common experience of the overclocking community to date has been that cards running at GPU temperatures within 80C are not likely to experience it, e.g. it's unheard of with water-cooled cards. Overclocked cards running above 90C add instability to the mix of issues. 85C is about in between. Now, that's the scoop, what to do with that is up to you. If your warranty is longer than you plan to use the card, you might not bother, send it back when it breaks. Otherwise you may want to take some steps. Also, since I mentioned optimized fan speed curves, here's how one should look: Fan speed itself will vary. The 50-66% range here is where you feel comfortable gaming, if you're OK with more, extend it up, or you can lower it if you need silence. The cliff at 55C allows the fan to spin up about as soon as you start a game, rather than waiting for the GPU to heat up. Finally 86C is where you want to cool it down.
  6. Update: After making the post above, I recalled that my own cards were a bit overdue for cleaning. So I turned my PC off and did it. The result is minus 5 degrees idle (probably even more at load). And it's not like I don't clean it often enough as it is, usually monthly, delayed a bit more recently. I'm done by now, so all it took was a little over an hour, and I took everything out of the case to do it thoroughly. If yours hasn't been cleaned for long enough, even 8C-10C would not be surprising. You won't believe how much dust can accumulate there and how much effect it has.
  7. 75C is fine, I'm surprised you managed to get it down to 64C under load. It's past 80C when you have a reason to lower it. Also, from the temperatures you're getting, I take it that your CPU cooler is a sealed one like Corsair H100, right? (These aren't what is usually meant by water cooling). If so, make sure that it blows outside the case - the manual may suggest to place it to suck air in, but that shouldn't be done (raises case temperature).
  8. Try lowering the speed then. The key is getting an acceptable noise level speed that can still cool it reasonably and near-flatlining around there. Also, how long have you had this card, have you tried cleaning its heatsink? They are very prone to dust accumulation. Using a vacuum cleaner, without the brush, from both ends (fan intake and exhaust) should do a lot without disassembling.
  9. Advanced fan control (presumably you're using it) in Afterburner works through the program - video cards don't normally use multi-step curves. 85C is pretty normal in the sense that it won't kill the card or that it's not unusual for a 570 with stock cooling, you can live with it. But it's still probably worth getting it lower if noise doesn't change significantly. Just prolongs the card's life, nothing critical. I like to first find out 3 fan speeds by noise: 1) can't hear it, 2) very comfortable, 3) not comfortable but still can play with. Then set up the curve such that it's speed 1 until say 55C (light gaming load), very sharply rising to speed 2 by 60C, then very slow rise for a while, then a moderate notch up to speed 3, at say about 75C-80C, continuing, and at 85C-90C another very sharp rise all the way to 100%. This keeps it quiet in the entire range where it can stay quiet, while preventing overheating.
  10. After saving, don't forget to select the number of profile to save (1-5). Don't close Afterburner, just minimize it to tray. It only takes 10MB of RAM with all features on. With proper settings it will start with the system and stay in tray.
  11. Probably just how the stock profile is set up. Try using Afterburner or a similar program to make a more aggressive profile, with earlier spin-up and hysteresis.
  12. The limit for GF110 (used in 580 and 570) is specified as 97C, slightly less than 100. In practice 95C when overclocking (at stock clocks even 90 shouldn't happen!), even in stress tests, is the point at which you want to take your voltages and clocks down a notch. Games will sometimes load the GPU to that temperature, it will start throttling and/or become unstable at 95-97C. Plus, temps on mid-high to high-end Geforce cards should be kept a little lower to avoid solder cracking (manifesting as visual artifacts worsening with age), so for long-term use 90C should be the never-exceed temperature. Low-mid Geforces and all Radeons are not susceptible, so a little higher like 95C is an acceptable never-exceed threshold. Ah, didn't expect that. Just wondering, do you have your GPU(s) in the loop as well, and what waterblock and radiator do you use, EK?
  13. 70C is on the other hand too low... unless your sensor underreports the temps, an air-cooled CPU will normally hit 70C under full load in Prime95. What I use is a custom fan curve. For GPU set in MSI Afterburner, for CPU it's harder. You can set the fan to spin up to 100% past 84C (add 2C or more hysteresis) - it's audible if you want a warning, but 100% fan speed will keep the temperature down on its own if you don't reduce the load.
  14. It matters if it's 32-bit or 64-bit. Only 64-bit will have the CPU run at full power. Special fan profile isn't needed. That's way too high. Tjunction is specified at 105C, and it's temperature that puts it at risk of quick destruction. Don't let your CPU go above 90C, if it does, drop the voltage and the clocks. GPU can go up to 95C, that's also it, at 95C with Geforce you are very likely to suffer GPU solder cracking in 1-2 years of part-time use, so still try and keep it under 90C.
×
×
  • Create New...