Problems? See this thread at archive.org.
So it's just winsat.exe?I thought it was another strange aero dll or something.
-
No offense to either of you but anyone knowledgeable enough to know how WEI works under the lid will probably have better benchmarks in mind, right? I mean, is this one of those "why does a cat lick itself" (get that reference? I had to neuter it a bit!) kind of discussions?
-Matt -
Yup, it's just winsat.exe, you can run it from the command line, and the arguments are all listed in the xml file it produces. In the graphics tests it composits off screen (nothing unusual post-Vista, everything we see on the screen is first prepared in the off screen framebuffer - which incidentally makes Optimus possible), but if I remember it right, it is possible to see its work on screen. I won't bother finding the exact way unless begged.
As a basic benchmark WEI is quite good, even this Optimus ''bug'' shows something that is true - it takes time to wake a device from deep sleep.
Of course, if you want real hard numbers, a real hard benchmark is the only way to go.
-
I guess part of the problem is that almost all the synthetic benchmarks will test something particular that isn't easy to translate into practical performance. So I suppose that's one reason why the wei index is useful to some extent. That you see if the hdd or the graphics card is what is slowing you down, that kind of thing.
But it doesn't tell you about actual performance, and I've seen before that computers with low or dynamic bus-speeds, excellent timing scaling, extremely good dual graphics setups, etc. - that score terrible. Same with performance drops in the test from quirks with the virtual memory setup that actually gives you better performance in practice. Or memory timing that are utterly braindead that gives you better test-results in that particular test, but which you won't get in return anywhere else, etc. The wei index also doesn't tell you if for example your sandy or ivy bridge card only partially supports dx11, and can never run well with certain instruction acceleration. It doesn't expose any bottlenecks either in the sub-systems, and therefore arguably won't tell you anything useful at all.
So I use 3dmark11 for 3d performance. And sandra to look at the individual system-specs. Since then you can tell if it's the bus or the disk, or the memory against the controller that is shutting down, etc.
But then again I also like to look at ipstat spoofs in plain text to search for errors, so I don't know.. -
WEI (Winsat) doesn't really give a damn about the state of DX11 support on any of our cards, since it never gets beyond DX10.
-
hello again, I've just noticed two barely visible spots of a bit darker color on my screen, located at about 10% of height from the bottom and at 40 and 70% of width, about 2 cm in size. They are only visible against dark gray background. When I press slightly on the back of the lid, these spots ripple. Even if I press far from those spots, the rippling only occurs there. It means the lcd is touching the lid (or some supporting structure) in those locations, and I don't know if it's bad or just by design. The lid has no visible depressions or deformations. I wouldn't say it's a huge issue, because it's visible only on occasion, but I'm a bit worried if this could be a premise for dead pixels in future. Has anyone else noticed spots alike? Is it a warranty case? (I'm not sure I'm willing to mess with warranty on this one, but still it's .. well, you know, SOMETHING ISN'T PERFECT
-
-
I never miss a chance to be utterly categorical about something So here goes..
No.. unless you plan to overclock the card to a degree that will make the samsung chassis melt (it leads heat into the chassis), then there's no performance gain to be had. My theory is that the gddr5 ram design itself is intended to have similar performance but at more compact pcb construction. While in any case, the graphics card chips (and low level instruction handling) is identical, so you won't see any difference for normal shader-operation. But the pcb design may make the gddr5 chips more overclockable than the ddr3 chips.
There may be situations where gddr5 performs better though, but I haven't seen any proof of that yet. And I really think it's hearsay - that no one has actually found any example of it, either with aa operations or specific physx routines, or something like that. Gddr5 ram also uses more power than ddr3 ram. So don't go for the higher numbered ram model because of performance.
Actually.. that model should be cheaper, since the ram itself is cheaper to produce in mass.. Could be something else you get with the samsung that works better, though. Even if they have the same annoying locked bios, by all accounts.. -
rashboosh,
I have the same issue and mine is also a May production. Other than that and the subwoofer noises, which I think is caused by the wires or circuit not being shelided as the noises seem to be related to the screen refresh. I am still debating about returning mine and waiting for a later production. My only concern there is I may be exchange one set of problems for another
-
@KLNP93: Seems the usb3 thing works normally. USB isn't stable, and it has variable speeds and so on. But it seems to work as advertised.
Usb 2.0 compatibility is not a problem. Only issue so far seems to be that some of the win7 installs you can find around the net don't include a default usb driver that works. So you would end up booting the install from a usb-stick just fine, but then the install would fail after the init. .. that kind of problem exists. Can't see any other problems.
-
-
-
-
Probably depends on the stream quality. If it only happens on youtube, then that's the problem.
..and if there's distortion because the speakers are about to die, it should be audible on lower volume as well. Same as on any other amplifier. You could amplify a very low signal from a mobile phone, for example, until it's finally audible, and amplify the distortion as well. The speakers and the amplifier works perfectly fine, but the input signal sucks. And you're bound to have some problems like that if an application is amplified up or down with the soundcard before it's sent to the speaker/amplification part.
..er.. what I mean is. If you turn the volume all the way down in the youtube window, or all the way up, you're bound to get distortion. But usually people don't amplify the signal, or the laptop speakers sound like buckets no matter what they play, etc. -
-
-
nipsen, Eduard Khill (R.I.P.) always improves my mood, but I don't hear him singing anything meaningful about nvidia drivers in this song!
-
-
-
..the stock screen is an lg screen, same as the one used in clevos and thinkpads. ..lgd02d9..? 95% rgb range or so. I've no idea if the alternative screen is better, though.
A reinstall isn't extremely painful, since all the drivers you need are put on the driver disc asus puts in the box. See the guide in the sig.
There's a thread elsewhere on the forum where the guy from GentechPC listed a few ram-sticks that work with the current locked ram-timing in the bios. But if you don't need 16Gb now, I wouldn't recommend upgrading until Asus fixes their bios-tweaker a bit, and set the bios-settings to default spd-timing for the ram instead.
Have to say, though - if you're not actually going to use it for games, heavy 3d modeling, simulation runs, stuff that need good cooling and high performance. Or don't do media-file conversion a lot, use hdmi out, run a media-server, etc. -- you really should consider pilfering the piles of unsold llano/apu cheap-tops the shops will want to get rid of now. Track a few of the models, and take a look in the shops. Might be you could get a very good deal on a dual-core e450/apu setup with extra ram. U32u, or similar. Or you could look at the trinity setups that turn up now. A hp dv6 with all the useful extras + dedicated extra card for $700, that kind of thing. These computers will last approximately twice as long as the intel/nvidia combos on normal day to day work. I'm not exaggerating. The u32 versions with an i3/630m last half as long as the same computer with an e450 setup. Tested it myself. The i3 version streams music via spotify for 4.5 hours while surfing via wifi. The other lasted 9 hours (and that was before underclocking and lowering the volt). And they're still fast enough to work extremely well as a work-computer, or to play videos and music on. I saw one of those with 8gb ram on sale for ~$500 a while back. ..No one wants to buy them for some reason.
Just saying that you might end up paying $800 for a two inch larger screen with higher resolution, and a graphics card and processor you're not really using. Doesn't matter if the n56 probably is the most well-made laptop ever made then.
Asus N56VZ-DS71 first impressions
Discussion in 'Asus' started by The_Finder, May 18, 2012.
