no uxga ?? 

|
Originally Posted by Adam@PCTorque
Sorry, I dunno why I was thinking the 9300 had the P4, guess I should spend more time reviews Dell than being up all night putting new options and models on our own site
.Up to 5 hrs? Is that if you buy the upgraded battery, the 1.6GHZ CPU, and the X300 GPU and run with 1 basic program? I believe that's how most companies rate their battery life is with the least demanding options. I like to run a DVD full screen at a volume that I can hear well and time it until it shuts off for my battery life tests. That way it's 100% even playing ground for best comparisons and a decent load for what a lot of people do when on battery. I may have to order one to put them side by side. The 4880 is supposed to really do well with the battery and I won't give it any breaks in the review. |
|
Originally Posted by CarrotTop
Nevermind, i seem to have answered my own question.
According to http://www20.tomshardware.com/mobile...119/index.html When using a dedicated graphics, the difference between Dual-Channel 533 and Single-Channel is almost non-existent, because the extra bandwith is apperantly only used, when using integrated graphics. |
|
Originally Posted by johare
Where did you get that from that article? With the 4880 it isn't just a matter of single channel vs. dual channel....it's 533mhz vs 333mhz and there seems to be a huge speed difference there.
|
|
Originally Posted by CarrotTop
There is a huge difference, almost 4 times as much memory bandwith, but with a dedicated graphic card the computer simply doesnt use it (according to the article).
A couple of points from the article: "As the benchmark results tables reveal, the speed advantage afforded by the new platform's dual-channel DDR2-533 memory is a maximum of 5% compared to single-channel mode. This outcome is not really surprising, because if you look at the peak bandwidths of the new CPU's memory and FSB, you can see that the single-channel configuration is already sufficient to provide enough data to the CPU. The doubling of memory bandwidth thus has only a marginal effect on performance." "The new faster and more frugal DDR2 memory technology turns out to be more of a paper tiger than a real brute when used with the new 533 MHz FSB Pentium M and the 915PM chipset with dedicated graphics chip and memory. The dual-channel memory interface of the 915PM chipset doesn't have much impact either. The performance-limiting factor remains for the time being the too-small maximum FSB bandwidth of Intel's current mobile CPU generation." I agree that DDR2 looks great on paper, but if the difference is 2-3% on average then i dont think its worth the added cost. And DDR333 is not a typo, or something, here is the Clevo specs: http://www.clevo.com.tw/products/M400A.asp |
|
Originally Posted by CarrotTop
Then look at the Benchmarks, sometimes the new platform gets beaten by the old one, and i think the difference in the graphics benchmarks are due to the X600.
All im saying is that DDR2533 isnt gonna be thing that is going to make a huge difference in the performance. And i doubt Clevo made the decision to put DDR333 in this computer without some serious research. But the only thing that will tell if it makes a difference in this laptop is a review of it, especially in the games. Does anyone know how the X700 is clocked in this thing? |
|
Originally Posted by johare
I went back and looked at all those benchmarks and with the single exception of a video encoding benchmark, the 533mhz DDR2 had a very clear and significant advantage in every single benchmark. Even in the synthetic and applications benchmarks, the 533mhz machine was clearly faster than the 333mhz. Go back and look at those benchmarks...it seems pretty obvious that the 533mhz DDR2 makes a very noticable difference. If those specs aren't a misprint then Clevo clearly didn't do their homework on this one.
|

|
Originally Posted by CarrotTop
All the synthetic benchmarks in the world are not going to make a difference. Of course is has more memory bandwith using 533 than 333. I have never said otherwise. But im simply saying that it doesnt matter, if it have 1 GB/s, 10 GB/s or 100 GB/s if it cant use it.
I dont think it can use that much memory bandwith, you do, we can agree to disagree ![]() If this thing can run HL2, Doom3 or WOW on decent settings, run my compilers, then i couldnt care less if it were using DDR133. Im trying to look at the whole package here, which seems decent. |
