If you can't get even Ubuntu 15.04 to boot, you might want to use this fix: http://ubuntuforums.org/showthread.php?t=1613132 The section about "nomodeset" parameter if what you're after. The nouveau driver is almost completely reverse engineered and at least the part that handles desktop CPUs is completely unsupported by nVidia. Adding "nomodeset" parameter will prevent nouveau from loading and you'll have basic but working video output. Once you get to the desktop, you can install proprietary driver developed by nVidia through the "Additional device drivers" tool.
A few extra hints:
- You shouldn't have to add the "nomodeset" parameter permanently, nVidia driver installer will take care of that.
- Do not add any unnecessary kernel parameters unless you have a good reason to think that they will fix a particular problem.
- Always try to use 3rd-party drivers provided by the Ubuntu tool. Do not download and install them manually.
- Do not install 3rd-party drivers for a device that works. A lot of devices have decent open source drivers integrated into the kernel. 3rd-party drivers are usually pretty messy and don't play along nicely with the rest of the kernel infrastructure (yes, Broadcom, I'm looking at you!)
Since I have no intention to use Windows anywhere I thought I'd stay out of the stream mostly worthless babble that floods the web every time new version of Windows is released. However after I looked over some of the statements in EULA and some of the default settings related to "privacy" I can't help but wonder if this is some sort of elaborate joke. I realize that a good portion of it is hyped up a lot but I still find it concerning that people apparently have no problems agreeing to some of the terms mentioned in the EULA
- I get that Cortana needs to gather some information about the user's habits to work efficiently, but why is this an opt-out setting? I suppose that Cortana should also be covered by a separate EULA.
- Backups of the BitLocker recovery key to OneDrive. Come on, really? Besides my natural distrust to an encryption scheme that even supports something like a "recovery key", can one possibly do anything more stupid that upload a decryption key to a cloud storage?
- The paramount of all the "we don't give two shits about your privacy" gems has got to be this:
This whole paragraph can basically translate to anything because the phrasing is so vague. I always thought that my computer was supposed to do what I tell it to, not the other way around.
All of the above plus the fact that one has to jump through all sorts of hoops to tune at least some of the snooping services down is a huge letdown for me.
New optimizations lifted some of the CPU bottleneck so your GPU can now spend more time rendering frames. That's why it's getting hotter. Do what Dave said and fix your heatsink before the GPU burns a hole through your computer.
I noticed a pretty nice FPS increase even under WINE, especially at places where the framerate used to drop to low 30's.
That's not how it works. Without VSync and framerate that is not an integer fraction or multiple of the screen's refresh rate you get tearing - one part of the screen displays Nth frame while the other part of the screen shows N+1th frame. This is annoying to look at but it doesn't introduce any delays.
Small heads up. Kernel 4.1 is starting to hit repositories of rolling update distributions. Issue with some newer revisions of wheels not being picked up by the driver should be resolved there.
Can you post your config file? The parsing function for the config isn't exactly brilliant but there is no obvious reason why this wouldn't work. It definitely did work in older versions of the mod.
Now in LFS_External Intialise insim events are :
InSim.BTC_Received += new LFS_External.InSim.InSimInterface.BTC_EventHandler(BTC_ButtonClicked);
I need to know that for InSimDotNet
Read the section "Receiving Packets" in the tutorial. Seriously, everything you've asked so far is explained there. Take your time and read it, at the end of the day it'll be easier than asking other people for everything.
CSMT is an out-of-tree patch that Fedora includes in its WINE packages. It can obviously be used anywhere as long as you're willing to patch and build WINE yourself.
It's so ridiculous calling it an IGC just because that's what Intel call it.
No, but claiming that IGC is not a GPU is like the essense of nonsense. By the same logic you should not consider "Civic" a car because Honda calls it "Civic" and not a "Car". It's the function that matters and by definition "Graphics Processing Unit" is a piece of computer hardware specifically designed to generate image and display it on a screen. Intel IGCs do exactly that, hence it is perfectly correct to call them GPUs.
That's cus Intel onboard is an IGC, rather than a GPU.
For a little while I considered explaining why is this statement so ridiculous... then I realized that this sort of argument is consistent with most of your posts so I just chuckled and went on about my day.
DX12 or Vulkan make sense only for applications that tax the GPU so heavily that the API overhead and inability to have a fine-grained control over the GPU becomes a problem. I don't believe that LFS is anywhere near that limit. DX12 and Vulkan also require a fairly new GPU because - as far as I gather - both of these APIs make some assupmtions about how a GPU works inside. All this "3D revolution" stuff is just marketing talk. Games won't suddenly get awesome just because their 3D engine is written in Vulkan.
Well, Valve was implementing Vulcan things on existing Intel GPUs (if I recall their presentation correctly), which would make me think that a lot of functionality can be implemented at the driver level, not requiring support at the hardware level (like some OpenGL features require).
I could be wrong, but Vulkan seems to give the greatest "hope" of being backwards compatible. DX12 could be similar, as the last few DX versions haven't seen the hardware changes like some of the DX9 GPU features required.
According to this Vulkan assumes OpenGL 4.3 or OpenGL ES 3.1 compliant GPU.
Which is basically "Any GPU in the last 5 years" after some quick Googling.
Intel GPUs seem to be the exception since official support for OpenGL 4.3 in Windows drivers is available only since Haswell chips.
DX12 or Vulkan make sense only for applications that tax the GPU so heavily that the API overhead and inability to have a fine-grained control over the GPU becomes a problem. I don't believe that LFS is anywhere near that limit. DX12 and Vulkan also require a fairly new GPU because - as far as I gather - both of these APIs make some assupmtions about how a GPU works inside. All this "3D revolution" stuff is just marketing talk. Games won't suddenly get awesome just because their 3D engine is written in Vulkan.
Well, Valve was implementing Vulcan things on existing Intel GPUs (if I recall their presentation correctly), which would make me think that a lot of functionality can be implemented at the driver level, not requiring support at the hardware level (like some OpenGL features require).
I could be wrong, but Vulkan seems to give the greatest "hope" of being backwards compatible. DX12 could be similar, as the last few DX versions haven't seen the hardware changes like some of the DX9 GPU features required.
According to this Vulkan assumes OpenGL 4.3 or OpenGL ES 3.1 compliant GPU.
DX12 or Vulkan make sense only for applications that tax the GPU so heavily that the API overhead and inability to have a fine-grained control over the GPU becomes a problem. I don't believe that LFS is anywhere near that limit. DX12 and Vulkan also require a fairly new GPU because - as far as I gather - both of these APIs make some assupmtions about how a GPU works inside. All this "3D revolution" stuff is just marketing talk. Games won't suddenly get awesome just because their 3D engine is written in Vulkan.
The last log captured with the wheel disconnected doesn't seem to show any obvious pattern. It'd probably be a good idea to run some tests with games other than LFS. At least we need to figure out whether this problem is specific to LFS. Is there anything suspicious in your "dmesg" log?