Jump to content

ratinox

Registered User
  • Posts

    128
  • Joined

  • Last visited

Everything posted by ratinox

  1. The best TIM in the world is useless if there is no coolant flowing through the heat exchanger. In other words, there are no "secondary methods" in a cooling system. Either everything works or the whole fails. Ignore cleaning an airflow at your own risk.
  2. MSI is good for the price but the build quality isn't as good as Dell/Alienware. They don't offer any kind of on-site service; it's either take it to the vendor or pay for RMA shipping. Step up to Clevo/Saeger and you get the build quality at a higher price point but you still don't get on-site service.
  3. Which is 10C lower than Intel's maximum temperature rating of 105C so you're operating within the manufacturer's specifications.
  4. Cleaning out the dust should always be your first activity. Nothing else will help if the physical bits are clogged. Prop up the back with anything that opens up airflow under the fan. I use a couple of cheap rubber door wedges. That alone is good for a 5C to 10C temperature reduction on the CPU under peak load. I don't know when it was added but I know that the A13 firmware has a toggle to disable CPU Turbo. Turning CPU Turbo off is good for a 10C to 20C peak temperature reduction on the CPU.
  5. I've never had a problem with ISRT that was not of my own making. For example, trying to dual-boot Linux.
  6. Then you have (had) a faulty main board. Dell will (should) do something about that if it's under warranty or you pay for the repairs.
  7. I doubt it. The 67C throttle point is nVidia's spec. Dell can't change that without violating whatever agreements they have with nVidia. At the very least it would void the GPU warranty. The problem disappears if you prop up the back of the unit to open up airflow under the fan. But I repeat myself yet again: http://forum.techinferno.com/alienware-m14x-aw-14/1755-m14x-cooling-3.html#post28820
  8. Optimus is the mechanism that switches between the on-board Intel IGP and the discrete nVidia GPU. Optimus is part of the nVidia graphics driver, not the system or video firmware.
  9. Mathematical accuracy vs. speed. The Quadro line is for professional use where consistent mathematical precision is required. The GT line is for games where speed is required and mathematical accuracy beyond 2-3 decimal places is irrelevant. In practical terms, you won't see a difference between the two in actual game renders other than frame rates. Where a GT might render a game at 30 frames per second, a Quadro of the same vintage might render it at 5-10 frames per second. On the flip, if you're doing CAD work, the GT might leave you with machined parts that don't quite fit together properly. This might be acceptable for Ikea but not so much for Boeing.
  10. You don't need an unlocked BIOS to disable BDPROCHOT. c.f. every Throttlestop guide ever written.
  11. Factory defaults. Those offer the best balance of heat, performance, power consumption, stability and longevity. For anything else? Ask Google about overclocking guides.
  12. Welcome to the real world where refreshes and new models arrive every 12-18 months. You had the choice of waiting or buying when you did. You choose to buy and you got your R2 several months before those who waited for the new models. That's not a matter of fairness. That was your choice, the same choice that anyone who buys computers has to make. GTX 765m is not "much better" than GT 650m. It's incrementally better. The SPU and clock specs look good on paper but in actual use, with real games, 765m functions as an overclocked 650m with more SPUs -- because it is an overclocked 650m with more SPUs. We won't see "much better" until nVidia starts shipping the Maxwell GPUs next year. PS4 and Xbone will have about as much processing power as a mid-range Core i3 which means what you have is already several times better. AMD's Jaguar FPUs? They're for netbooks. They're made for low price and low power consumption, not for performance. Don't get me wrong. I love AMD's Fusion architectures of which Jaguar is the latest iteration. It's just that Fusion is terrible for high-end gaming. But then, neither Xbone nor PS4 are for high-end gaming. They're for the iPhone generation of social networking and media sharing. If you're having problems with BF3 then it's not the GPU. It's something you've done or haven't done. Install the latest nVidia drivers. Set the game to use the nVidia GPU instead of the Intel IGP.
  13. I listed them because they're games that run both CPU and GPU to full capacity.
  14. So. I've been running the unlocked A13 firmware for about a week: CPU Turbo disabled in firmware, BD-PROCHOT disabled in firmware, Core Clock and Memory Clock set to +135/+400 with MSI Afterburner. It's been solid for Guild Wars 2 and XCOM: Enemy Unknown.
  15. Yeah, because a whole 180 scan lines is a deal-breaker. Hint: that's about half of a NTSC TV screen's vertical resolution squeezed down into about an inch of screen real estate.
  16. Note: nVidia Inspector resets the GPU and memory clocks to their default values on launch.
  17. svl7 put flash.bat in the archive with the bin file. It prevents the GPU from throttling the CPU. Read the Throttlestop guide for details.
  18. Minor nit: you forgot to include the flash.bat file in the firmware zip. Edit 1: Some interesting things in this firmware. I don't recall if it if was in the unlocked A11 but the unlocked A13 has a setting for toggling BD PROCHOT. Nice having this as a firmware setting near the CPU Turbo setting.
  19. The original M14x has higher-end Sandy Bridge CPUs. The current generation Haswell CPUs are only about 15-20% "more" than their Sandy Bridge predecessors. Similarly, nVidia's Kepler GPU architecture in the R2 and not-R3 is only about 20-30% "more" than the Fermi GPUs in the R1. You won't find "over double the performance" for any price point. Such electronics don't exist. The original M14x probably isn't worth it today. On the other hand, I'm not convinced that AW14 is worth it compared to R2. Haswell + Kepler isn't significantly better than Ivy Bridge + Kepler. Haswell + Maxwell, nVidia's next generation GPU architecture, will be but that's not happening until next year.
  20. The only reason to put a Quadro 4KM in anything is OpenCL/CUDA processing and certain kinds of CAD work. Alienware kit does not lend itself to either of these tasks. Notably, CAD work is hampered by the display options. The screens are beautiful for gaming but terrible for serious work. Try doing a color calibration if you want to see just how bad they are. There are better options for professional notebooks than Alienware. Go with one of those if you really need the Quadro.
  21. EEVEE! Shippers are notorious for not handling with care. A couple of weeks ago I got a parcel, a box about twice the size of a typical notebook shipping carton, with one side crushed by about 1/4 of the box's full size. Fortunately the contents was a pillow -- really, a down pillow I'd ordered -- so no actual damage to the merchandise. Had it been a computer with foam packing around it then the box wouldn't have been crushed; the foam would have absorbed the impact energy. That's what it's for. Shippers do far worse than what the guy in the video was doing.
  22. It's not a closed system. Every time the jet "inhales" it draws in cool air from outside the chassis. Every time it "exhales" it pushes air across the heat pipe and out of the chassis. It's more like a bellows than a jet, but "jet" sounds smexy while "bellows" sounds dull.
  23. Starting yet another thread to rant about this isn't necessary. But to answer your questions: 1. Yes. Several, in fact. They're detailed in the many threads here on T|I, at Notebook Review, Alienware Arena, Tom's Hardware and others so I won't bother reiterating them. 2. See Answer #1.
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.