Thread: Changing processor worthwhile?

Good, you outline the pc I have, I've changed the graphics do nothing, and I think that makes me a bottle neck, that graph q would have to pay more than you paid, the question is, with this plaque is worth changing the process ? me or the board also restricted?

CPU Type DualCore AMD Athlon 64 X2, 2300 MHz (11.5 x 200) 4400 +
Name Gigabyte motherboard GA-M61PME-S2 (2 PCI, 1 PCI-E x1, 1 PCI-E x16, 2 DDR2 DIMM, Audio, Video, LAN)
System memory 2048 MB (DDR2-800 DDR2 SDRAM)
Placa video ATI Radeon HD 4670 (1024 MB)
Disk ST350032 0AS SCSI Disk Device (500 GB, 7200 RPM, SATA-II)
Generic 450W

I do not think a 4400x2 with a 4670 neck make the truth.

You could spend the 3dmark 06 and give us values. In which bases to say that you would have to pay more?.

Also you can influence this generic, but 4670 seems to me that he had no food or extra ¿?. Te lo digo pq a friend have a 9800gtx + and pays almost like my 9600gt and absence of A in the source.

I on 3dmark 06 with 5400x2, 2x1GB DDR2 800 9600gt. LC 480w Scorpion

2.0 ----- 4286
3.0 ----- 4295
cpu ----- 2129
-------- 9308

Also check and see that there are some drives that ATI is not version 9.2 or 9.3 which came out to say the frog to see improvement and worsening performance.

I base that my friend is some one pc similar to yours, and takes considerably more performance in games that mine with that processor ... Anyway after the way he said and 3dmark results

Thanks for such a prompt response

After talking with better data, but if your friend has a computer as mine is normal to get more performance. Whether by micro pq something and the 9600gt is more than 4670. So there is no way that I like your team so much as micro graph below.

Then also interferes with the operating system you use, monitor "size and quality graphics that you choose.

Here comes too much resolution of the monitor to see if the processor is the bottleneck that makes it or not. Although this bottleneck, without continuing to agree with the way that some use it, but now this.

In theory, if a component does "bottleneck" to the other components because it is put to rest at peak performance. That usually happens when a graphic is mounted on medium-high range with a processor with a time and monitor a few inches (17 "or 19"). In that resolution, the chart is able to compute more fps than what supports the CPU, especially when there are many objects on screen. The graphics are effective with older processors normalitos or something if you get a higher resolution screen, type 22 or 24 "with 1680x1050 or 1920x1080. In that resolution, the graphic touches ceiling before the processor because the processor is almost the same loading process to 1280x1024 to 1920x1080 which, as objects and all the calculations are very similar (except for the calculation of the physics of fragments and little else).

4400x2 oc that processor will not be short any graphical medium / high current range with a 19 "monitor. With a little oc can greatly enhance the rate of fps games in general. So if the processor power is what is top and much better performance because there is still far from the top of the chart.

If you increase the resolution of the monitor without applying oc, the result will be almost the same fps rate, but allow us to apply all kinds of filters which carried out the plot. So until we get to the top of the chart and it is this that makes us the "bottleneck".

To which I referred to not agree with the term "bottleneck" is because there are people who say it makes the processor bottleneck or off the charts when he gives a rate of 70 fps or more! ! Sorry, but this is already a considerable amount of fps and if you are not trying to beat a benchmark record, you can not be called "bottleneck" to something that works so well! Another thing is that the processor or ram, or a slow hard disk, we get to levels that can not play when the rest of the PC meets the specifications and a maximum stake.

Now I have a two-sli 8800GT 512mb with a 3600x2, which I superoceado from 1900mhz to 2850mhz. That brings me some more performance than a 5400x2 and I know that if you do not draw more fps in a game is played because the processor ceiling, because the graphics can still move a lot more cargo. Hence, if the upload processor 3060mhz FPS go through the roof, then a little more ability to make the processes responsible for printing and much more without rechistar (physics calculations for example). But at that speed and voltage can not maintain 24 / 7, so for Bencher very well, but for everyday to keep it going more than 2850 seconds!

My recommendation is that it targets an average oc your processor, you will notice much improvement, but also tell you that 4670 has no cap as far as my 8800GT. So do not hesitate to give Crysis whole host ... but nfs undercover, or prince of persia or CoD5 that can move at maximum rates with good quality (if not a lot of resolution).

I do not think it has anything to do for you, but I will tell you that one source of power less than 500W generic normally does not give more than 25A on the 12v channel therefore often inadequate for these charts. Can you get to start a game, even playing, but if you can not reach that of a painful performance. So look at the label or your source and see if we also play the hot air coming out of the source, these are symptoms of weakness and could do it is based. Not to mention that behind her death may be part of the team ahead!

In theory, your PC may need about 350W with integrated graphics disabled. If you have not disabled that can make the commitment to reach the source at 400w peak, so as your source is generic, not size. It is very justita.