Search results

That's even better!

I took the liberty of editing it for you. I think there is a box you can click that says "Insert as thumbnail", and will insert a smaller version of the picture that is clickable.

I understand that thought! We have taught ourselves that if there's a problem, it must be at the computer end.It's just in this case, that's not true. I had enough pain with HDMI, TVs and overscan, but stubbornly refused to believe that there wasn't a way of resolving it. Once I discovered it...

It has been a while! I have had my fair share of battles with computers and TVs, so I know the fun of overscan sadly. I'm assuming you're using HDMI from what you've posted. First up, I'd recommend you configure the TV not to overscan. For an LG TV, you need to go into the menu,select "Picture...

Here in "rural" Isle of Wight (UK), I'm paying $71 for an 76/19Mbps DSL connection, and pretty much seeing that throughput too. Like Gizmo, if it wasn't for the fact that this connection doubles up for home and work, I wouldn't pay that much. However, it's pretty nice when you need to sync large...

Makes me wonder how much is down to optimization for the platform it's running on... I remember Valve making a big deal about how Linux OpenGL was faster than Windows OpenGL/DirectX, and it turned out their Windows OpenGL optimizations were not as good as they thought they were. We shall see! :)

DSpurway mentions at least three distros that support POWER8 (including RHEL and Ubuntu), and IBM sell Linux-only POWER8 servers. There are specific POWER8 features that were introduced in the Linux 3.8 (and 3.9) kernel. They look pretty nice machines to be honest, especially if you like the...

Sadly, PlayOnLinux utilizes WINE. As far as I know WINE requires IA32/IA64 architecture as it executes much of the code natively, and only translates API calls. Short of some kind of instruction translation, I'm not sure it'd going to fly. It'd be nice to see code running on non IA32/IA64...

;)

That's why I linked to a suite of benchmarks, rather than a single one. Look how slow the poor 270 is...

That's nice - but only under Windows - how does it look under Linux? (hint: http://openbenchmarking.org/result/1402194-SO-1402176PL32 )

I do wonder if there was a hole that size in a closed source library (like schannel) how quickly would it be found? People seem to like attacking closed source, especially from certain vendors. That doesn't mean open source or closed source has less or more issues. And did you check the code...

Like Danrok says, you could consider the first digit as the generation. So the current generation are 700 series, the previous generation is the 600 series. Inside a series, generally the higher the number the better the performance. There's also a "Ti" modifier that tends to indicate increased...

Slightly different example, but the recent news of the 9 year old gaping hole in GNUTLS highlights that open source doesn't necessarily make things better, and can actually make things much worse. I can also understand that things like copyright licensing can make opensource a complete nightmare...

I just use dump... but then again, this is on a *BSD box.

Sadly, as far as I know, Kepler/Fermi cards cannot be overclocked under Linux directly, as the drivers do not support it. It looks like your only option might be to modify the graphics card BIOS and flash it back.

It's a Uno32 - Arduino-ish, but 32bit MIPS core, rather than an 8bit processor. In this particular case, I'm fuzzing EDID data.

... ;)

The biggest problem is the difference in CPU, so there's a requirement to emulate chunks of the x86 platform - that's going to tend to slow stuff down. Yes, you can recompile x86 instructions at runtime, and that would increase the speed, but it's a much bigger ask than catching OS calls.

Don't expect the standard tools to work on laptops - they usually have their own controllers. People have reported that patching the DSDT in the BIOS can allow you to remap the temperature<->fan speed table. There's details at...