Jump to content

azagahl

Member
  • Posts

    437
  • Joined

  • Last visited

  • Donations

    0.00 USD 
  • Country

    United States

About azagahl

azagahl's Achievements

0

Reputation

  1. I was hoping for a bit more detailed advice, perhaps some benchmarks or rules of thumb. Oh well, I should have anticipated a kneejerk "c0mpr3ssion is 4 l0zerS" message. If you think about it for more than a minute, with a fast enough CPU, and slow enough hard disk, compression is guaranteed to increase system performance. On a two- or more core system, where some cores are idle 99% of the time, its worth considering.
  2. Can XP be sped up with Windows File Compression or UPX to compress exe and DLL files? It seems like my dual-core CPU and network are idle almost all the time but XP seems very sluggish. I'm guessing that the harddrive is the bottleneck. Probably its faster to read a Windows-compressed file and decompress it on a fast CPU, than to have to read a file 2x as big from the slow hard disk. I don't think it just this one harddrive that's crappy, the harddrive seems to be a bottleneck on most XP PC's I've seen recently, especially on dual-core systems. Thanks for any suggestions you have.
  3. >1. How hot my video card got at maximum load after 7 straight hours (overnight) If you stress all components (GPU, CPU, RAM, floppies, hard disks, cd roms) simultaneously you can reach the highest temperatures. The reason I'm interested in this, is that by stressing the whole PC, one can find out what the highest stable level of overclocking is. Unfortunately, most GPU tests dumb down the CPU with instructions that do not test stability and generate little heat, meaning that you are basically testing the GPU and CPU independently instead of testing them together. I'll try the Video Card Stability Test. Thanks, noguru. BTW, IMHO the ideal GPU test would also count and log artifacts instead of just trying to crash the system. This goal probably conflicts with the goal of low CPU usage though.
  4. >I don't think that's possible. And why would it not be possible? There's no technical limitation at all. A DirectX app can issue a complex ID3DDevice9::Present() call and then the thread can yield until the GPU finishes. Very little CPU time would be consumed. I've used Rthdribl and Orthos together along with several other stress test programs at the same time; unfortunately Rthdribl consumes lots of CPU time which is basically dumbing down the Orthos torture test and making it easier on the CPU. >You just run it overnight and keep a log of your temperatures with SpeedFan or Everest. Yes, we all know how to use a GPU stress tester. The question is, what good ones are available?
  5. I can log-in on my PC fine, until it goes into standby or resumes from the screensaver. Then my password won't be accepted. I've tried the obvious things like playing with Caps Lock and Shift, as well as Num Lock, Scroll Lock, Alt, and Ctrl. I can enter the password again and again. Nothing works and I am eventually forced to reboot to log in. This problem started happening as soon as I upgraded to a better PC. Maybe my old PS/2 keyboard is incompatible with the new PC? Any ideas? I don't want to give up the password feature, but it's an option I guess.
  6. What good GPU stress testers exist? I know rthdribl, 3dMark06, and Aquamark3 all stress the GPU. But these also consume a lots of CPU time. I want something that stresses the GPU only, while consuming almost no CPU time. Then I can run it at the same time as a CPU / RAM stress tester (e.g. Orthos). Any ideas?
  7. FYI, as I promised to do a while ago, I just tried installing the older version of AVG AntiVirus (7.1 Free I believe) on a clean machine and SURPRISE! it would not function at all, it just popped up "expired" messages. Even the Virus Vault would not open. What happens to the files therein when your AVG AntiVirus suddenly shuts off?? What happens if you have a virus spreading through your documents, and the only antivirus software you have on hand is non-functional AVG software? This isn't a complaint about trialware, which is of course a legitimate practice. The problem is that this AVG AntiVirus product had a bona fide, undisclosed timebomb implemented in it. Do the commercial products also have secret timebombs? I'm not certain but I suspect the answer is yes. Hopefully, such timebombs would not go off at inopportune moments. Please go ahead and attack the messenger though. A lot of insults were hurled and inane rebuttals posted, last time I mentioned this issue.
  8. Thanks for the suggestions. Orthos sounds interesting. Memtest works fine on my PC, but I've seen it work on exremely unstable PC's before. Two instances of Prime95 work but they make a 2.5 GB pagefile so they seem to test the disks more than anything else. Hahahaha.. I'm a total n00b. Two superpis run from the same place cause the same error on my other single core PC. But when run from different directories, no error occurs. So my Pc is stable AFAIK. And I actually have Radeon X1800 XL and my two 400 GB disks form a raid 0 volume. It rocks so far!
  9. I put together this dual-core system, and it can run SuperPi fine, but as soon as I start a calculation in a second SuperPi instance, the second calculation almost INSTANTLY errors out saying ""NOT CONVERGENT IN SQR05". I haven't attempted to overclock anything. Any ideas? Core2Duo E6600 (2.4 GHz 4 MB shared cache) Thermaltake Big Typhoon Asus P5B Deluxe motherboard 2x1GB OCZ DDR2/800 (PC6400) 4-4-4-15 Antec 900 case Antec Phantom 500 power supply ATI Radeon X800 2 x 400 MB Seagate SATA/2 HD's Samsung DVD combo drive Floppy disk
  10. Is it true that the cheap Intel Core2Duo's ($200-$250) are much much better than the same price AMD X2's?
  11. My friend argues that dual cores can help reduce stuttering, jerky frame rates in games. For example, when an e-mail arrives. It's hard to see how even a fast single-core can avoiding this. I guess I'll get a dual-core, although I'm still a little bit skeptical.
  12. >A single thread in the X2 3800+ system sees a 2.0GHz core. A single thread in the A64 3800+ sees a 2.4GHz. The rated clock speed is the speed at which a single thread will execute. You don't divide by two or four depending on the number of cores I understand what you are saying in terms of clock speed. The single-core should be about (2.4-2)/2=20% faster for a single thread, i.e. the dual core is about 1-1/1.2 = 16% slower for a single thread. This slight performance loss is probably well worth it for having the dual-core advantage. Unfortunately caches complicate things a bit. Are the L1 and L2 caches divided amongst the cores? What about memory bandwidth? Suppose that the core on the single-core CPU can hog all of the caches and memory bandwidth entirely to itself, whereas the dual-cores are burdened with half-sized caches and have to compete for memory bandwidth. Then the single-core has an even greater advantage than just its clock speed indicates. BTW, I apologize if I'm asking about things that were figured out long ago.
  13. Another question... Including the associated motherboard, is an AMD or Intel processor more likely to come with a DRM infestation? I haven't kept up much on (un)trusted computing, HDCP, etc...
  14. >The reason why a Athlon X2 3800+ might be slower than a Athlon64 3800+ is because the 64-3800 is clocked higher. Each of the two cores in the X2 will be clocked slower, but AMD gives it the "3800" rating because the second core "makes up" for the lower clock speed. When you write X2 3800 vs. 64 3400, I think from the point of view of a single thread and see 64 1900 vs. 64 3400. I'm assuming that an "X2 3800" essentially equivalent to two "64-1900" cores. Maybe I'm wrong - is the "3800" just an artificial performance number, perhaps measuring the average speed for a few typical situations? Are a fair number of these situations single-threaded? So X2 3800 really is much better than the 64 3400 in practice?? The ability to execute two threads is obviously useful. I guess I'm willing to tolerate a 20% drop in single-threaded performance in exchange for this ability. But a core going from 3400->1900 is too large of a drop. Is there an unbiased source for benchmarks? AnandTech and Tom's Hardware Guide are chock full of ads and seem less informative than a few years ago. I'm not sure whether to trust them.
  15. >MODS? What the... explain... Mods = user modifications (for the game Morrowind I mentioned). It's off-topic so I'll PM that person.
×
×
  • Create New...