Showing posts from July, 2012

EVGA GTX 260 - the best driver choice under WinXP?

I recently upgraded to a used GTX 260 (EVGA, type SC; model 896-P3-1257-AR to be specific), for the purposes of playing Minecraft, etc. more smoothly. This has been pretty successful - the GTX 260 gets about 2x the frame rate of my old ATI X800 XL, and is only slightly more noisy (because the fan is larger and runs faster). 

But the upgrade was not without problems (since solved, see below). The moment I put the card in I started having problems with hibernating WinXP which had never occurred with the old video card.  About 1/4th of the time hibernation would appear to finish, and the screen would go blank, but the machine would not actually switch off. Worse, upon rebooting, the computer would act like it had hung, and discard the hibernation file, instead proceeding to a regular, full boot. It was like my computer was crashing about once a day. 

Suspecting a driver incompatibility, I tried several different ones:

301.42 - hibernate problems
296.10 - hibernate problems
285.58 - machine somewhat unstable
280.26 - machine very unstable 

275.33 - hibernate OK, but machine is now unstable (most often hanging hard while watching video).

Other recommendations from around the net:

EVGA forum.

I contacted EVGA support, and they gave me a bunch of fairly basic tech support suggestions that I had little hope would fix the problem, and indeed they didn't. Then they suggested I RMA the card, which I did. Problem solved! The card is now completely stable with out any hibernate problems under the 301.42 driver (I didn't test any others). It only cost me about $5 to RMA the card, because of shipping it back to them. Turn around was super fast - they sent the replacement card out the same day it got to them, and it was in my hands the next day (note, however, both EVGA and I are in S. California).

So to conclude: hardware problem.

Update, again:  after about 1 month I had another crash on entering hibernate. That's a lot less than before I sent the card back, but still suggests that the problem isn't actually fixed, as my machine never had this problem with the old ATI card. Not sure what to conclude now.

How much faster have CPUs gotten in the last 15 years?

If you got into the PC business in the 80s or 90s, you can probably remember how each new CPU from Intel was a pretty big performance boost. In those days, a new CPU could offer two kinds of performance advances: clock speed increases, and instructions that took fewer clock cycles to complete. By the time the 486 came out, most instructions took just 1 cycle to complete (down from 4-6), and we had gone from 8mhz to 100+mhz in just a decade. With the first Pentium, Intel introduced the ability to complete instructions in parallel, which meant that a single clock cycle could sometimes result in 2 instructions executed (or more properly, completed, since at this point we had execution pipelines). Later CPUs had even more pipelines, but the gains were diminishing. Clock speeds were still increasing, however. This did not mean that CPUs were getting faster though. Famously, Intel actually took a big step back in performance per clock cycle from the Pentium 3 to the Pentium 4, in order to push into the 4 Ghz range.  In modern times, performance per clock cycle has taken a bit of a back seat to putting more cores into your CPU - great if your software supports multithreading, but even after almost a decade of this approach, a lot of software does not.

So, how do CPUs of the last 15 years compare on single-threaded computation? Not many people have the CPU collection to answer this question, myself included. Tom's CPU charts is the closest I could find, but in 2006 they switched to using a multi-threaded mp3 benchmark, so the best they can offer is the years 1994-2004. To extend this I replicated their benchmark (single threaded) on my desktop, which is shown at the bottom of the following list.

Tom's cpu charts 2004 (seconds to encode a large MP3 file using LAME, so smaller is better):

1994 Intel Pentium                      100mhz     4361
1997 Intel Pentium MMX           233mhz     1926
1999 Intel Pentium III                 600mhz       360
2000 AMD Athlon thunderbird  1ghz             233
2001 AMD Athlon XP 1500+   1.3ghz)         169
2001 Intel Pentium 4                  1.8 ghz         162   
2004 AMD Athlon 64 2800+    1.8ghz          119
2003 AMD Athlon XP 2800+   2 ghz            105
2004 Intel Pentium 4 570           3.8ghz            84
2004 AMD Athlon 64 FX-55    2.6ghz            82
2010 AMD Phenom II 255 X2  3.1ghz            46 (estimated)

The most impressive gains happened in the 90s - 4300 seconds is a long time to encode an MP3 (even a long one). Since then, single threaded performance has gone up, but not so much. The difference between 82 seconds and 46 is almost half, but that's not such a huge change in how long you have to wait. We can see, however, a general trend that more is getting done per clock tick (excepting the P4, of course). 

The weakness of this list, of course, is that there is no attempt to make sure that each of these CPUs is comparable in terms of cost to the user when they were introduced. But it does show that while raw Ghz hasn't gone up much in the last decade, compared to the 90s, single thread performance has still continued to rise somewhat faster than Ghz. 

Radeon incompatibility with Dell LCDs (U2410)

I recently upgraded my video card to an EVGA 260 GTX (SC), for reasons of performance and compatibility. My old video card (Radeon X800 XL) was reasonably quite, fast enough for non-gaming needs, and OK for watching flash video, but not great. Most annoying of all, it has an incompatibility with my Dell U2410 LCD, where it cannot wake the screen back up after it has been put into power saving mode by the screensaver. Apparently the video card sends a signal to the DVI port to see if there is a device there, and the DELL takes too long to respond so the ATI card stops sending a signal, and never discovers that anything is attached. I found I could get around this by opening a full-screen DOS prompt, which causes the ATI drivers to power up the DVI port whether it thinks there is a screen attached or not. I set up a hotkey to open and then close a fullscreen DOS window, thus bringing my display back to life, but what a pain! I don't think this was a problem with all Dell displays - I don't recall the same problem with my older 2408WFP.

To be fair to ATI, this is an old card. But the drivers weren't that old; you'd think this was something that could have been fixed in software. The good news was that the upgrade to an Nvidia card fixed it. Plus, flash video no longer exhibits taring problems. You know, full screen video playback was a problem most software solved in the late 90s. Flash is such a dog - apparently it can only use hardware acceleration with fairly modern video cards - which the ATI card certainly isn't. But then every other video playback program I used was tare-free, so it certainly had the horsepower to do the job if given even half a chance.

Email me


Email *

Message *