Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

straight_stewie

Member
  • Content Count

    1,554
  • Joined

  • Last visited

Awards


5 Followers

About straight_stewie

  • Title
    Veteran
  • Birthday 1994-05-10

Profile Information

  • Gender
    Male
  • Location
    North Mississippi
  • Interests
    Audio, Programming, Engineering. Just a hobbyist now, unfortunately.

Recent Profile Visitors

2,943 profile views
  1. straight_stewie

    Help with Memory Overclocking, doing an oopsie

    What motherboard do you have? I'd like to look up the manual and see what "performance enhancer" is.
  2. straight_stewie

    Help with Memory Overclocking, doing an oopsie

    There is a possibility that the conversion from XMP to DOCP is not working properly. The first thing you may want to try is to make sure that it is setting all your timings and voltages correctly. If that doesn't yield a solution, set the RAM to auto and boot into memtest86. Run a full standard test at those speeds and see if everything is working properly. If memtest86 gives errors, then we can go from there. Otherwise, the problem may lie elsewhere. Have you already overclocked the processor?
  3. straight_stewie

    Best Config for ML + virtualization + Gaming

    CUDA applications ignore SLI. This is why I take every opportunity to say that I think Nvidia should push CUDA as an open specification, then we could build graphics libraries on it and get some pretty awesome multi-GPU scaling that works for all platforms. (They should probably do the same thing with NVLink) People hate on SLI for gaming, but I've had good luck with it. Since your primary concern is ML, dual GPUs can speed CUDA applications like TensorFlow and MatLab up quite a bit for large sets. CUDA scales on both heterogenous and homogenous GPU architectures, with or without SLI bridges. Probably not for a while since it's a brand new build. That's especially true if science is your primary concern: You want long life and good stability. You may also consider ECC ram.
  4. I have no hope for FBI investigations. They couldn't find over 30,000 emails when an FBI most wanted person was literally posting a searchable database of them online, and they spent 2 years figuring out that Trump had sex 14 years ago, when he has a 14 year old son. I'm convinced that the FBI couldn't find their desks if they were sleeping on them. This will just go away like every other important problem that can't be tied with a nice bow.
  5. straight_stewie

    How to convince teacher?

    Bullied? Doubtful. The kids family won many lawsuits, becoming somewhat wealthy in the process. The child got to meet many prominent figures, including POTUS and astronauts that have flown on the ISS, and was awarded full scholarships to some decent out of country schools. That incident put Ahmed on the fast track to the good life.
  6. straight_stewie

    Basic must-have ICs

    That depends on what stuff you want to work on in your very basic lab. I keep a couple of FPGA boards and a few cheaper microcontrollers around, along with an assortment of basic logic chips, and even some radio and GPS equipment. Of course, my interests are processor design and vehicle guidance. What are your interests in the field?
  7. straight_stewie

    How to convince teacher?

    Make a clock out of a breadboard, discrete 7400 series ICs, and some 7 Segment displays. Make sure it displays time in a 24 hour format such that it counts down the time to 0000, with hours, minutes, and seconds. Take extra special care to make sure it beeps at a somewhat loud volume every second, and make a big deal showing it off. Last, but not least, make sure it uses a large LiPo battery pack for power, and that it is mounted in an obviously viewable location on the clock.
  8. straight_stewie

    Vacuum Chamber

    electricity can be a catalyst for corrosion. So is the simple idea that two metals, not protected by an oxide layer, are in contact with each other. It is not impossible, no one has said that it is. It's just that there are challenges that would have to be overcome to have a long lasting, stable system, even here on Earth. Some of us might even have formal education and professional experience dealing with these kinds of problems... And then, in the end of it all, to cool the thing you have to conduct the heat outside of the vacuum case and into some area where conductive heat transfer can occur, which means you haven't even alleviated the major problem here and that is debris/dust/dirt control, since you have to put the parts most susceptible to those problems, fans, radiators, and/or passive coolers outside of the vacuum anyway. It's an interesting idea, but that's about all it is.
  9. straight_stewie

    Vacuum Chamber

    Yes, they do. There are two types of corrosion that occur in any vacuum, galvanic corrosion and whiskers (google "Tin Whiskers" for the most information). If you're in a high radiation environment, such as outer space, there are even more problems that can occur, some of which even effect plastics.
  10. straight_stewie

    Vacuum Chamber

    Right, except for the GPU will overheat unless you find a way to cool it. As will all other components in the system that produce large amounts of heat, which makes finding ways to get the heat of the VRMs, chipset, RAM, any video/audio circuitry, and any other high speed component including HDD/SDDs out of the case. Having free air in the case, even if it's not circulating, gives these items a chance to dissipate heat.
  11. straight_stewie

    Vacuum Chamber

    I didn't say it wouldn't be possible. I said it would be impossible to passively cool them in the case... And, If I remember correctly, the video you posted wasn't in a vacuum, and that's what this post was about: Doing the same thing, but also in a vacuum.
  12. straight_stewie

    Vacuum Chamber

    Heat still needs to be radiated in a vacuum. By putting the system in a vacuum you create the extra difficulty that you now need to figure out how to cool basically all the other system components because passive cooling of the voltage management systems, ram, chipset, and associated hardware won't be possible.
  13. straight_stewie

    DX12 & Vulkan must be a must going forward.

    I was going off of their whitelist: https://www.geforce.co.uk/hardware/technology/dx12/supported-gpus
  14. straight_stewie

    DX12 & Vulkan must be a must going forward.

    Another big issue is backwards compatibility with graphics cards. Besides a few oddballs, the oldest Nvidia graphics cards capable of running DX12 are the 900 series. AMD cards go back a little further, but not that far. This is a big problem because many gamers still use hardware that doesn't support the technology, and game studios like to support the widest range of hardware possible.
  15. straight_stewie

    Next-Generation Console or new PC build?

    Yes. It's not the prettiest, all glass, full RGB thing in the world, but it supports all the motherboard form factors that you'd be looking at, has more than enough 3.5 and 5 inch bays to do anything you want there, and has way more than enough room to hold multiple large, modern graphics cards. It's ok for custom liquid cooling. You're limited to only three mounting spots, 2 120/140mm rads in the bottom and rear (although I would not recommend putting one on the bottom), and factory it can take up to a 240 radiator on the top (with modifications it could probably support up to a 2x200mm rad, that top area does support 2x200mm fans after all). I've only ever had two complaints with it: the cable management area is just the tiniest bit too narrow for round 24 pin cables, and the side fan is just slightly too high up to be directly blowing on dual GPU setups, the latter problem I've since fixed.
×