64 3000+ 939 Vs. 3400+ 754

[Hacp]

Dj-Chris, just to point out that the Memory bandwith advantage on going with dual channel is already accounted for. Why do you think a 3000+ 754 is 2.0 GHz while a 3000+ venice is 1.8? Not becuase of the 90nm process but because of dual channel. Yes you get a 50% boost in memory bandwith, but you only get 3-5% boost in real life performance. Also, the Newcastles can def overclock. On average people can at least get 200MHZ out of them. Most can get more, but lets stay on the low side of things. Finally, you blow the energy savings and heat dissipation out context. 130 nm processor do not require that much more energy than 90nm ones. You say ALOT more, but in reality it is only 10 or so more watts. That isn't alot more I would say. So to sum it up, what are the advantages of 939 over 754? Pci-express, which poses no advantage over AGP RIGHT NOW. Dual channel memory, which is already taken accounted for in the modeling numbers by AMD. And finally dual core. Those are the advantages comparing 754 to 939.

Finally DJ-Chris, he OP wanted credible links because it is his money. If you choose not to provide them, thats fine. Just that he'll beleive people who give links more than people who wont, because frankly websites are a little more credible than a poster in a forum.
Btw, here is a detailed look at the sempron. The model that was reviewed is based on the 130nm process, and the article goes into detail about 754 versus 939. Also, there is an overclock mixed in there too.

http://www.xbitlabs.com/articles/cpu/display/sempron-3100-oc.html

 

[Counter-Strike]

Once again, Hacp owns the day.

And stock PSU's vary. You can hear ppl who have had to replace it within a few days, and some people who haven't had any problems w/ it at all.

 

[FghtinIrshNvrDi]

Than you Hacp!

Ryan

 

[Hacp]

Btw, this power supply is reliable. Forton is known to make reliable cheap power supplies. Just don't plan on overclocking if you are to use a 300 watt power supply.

http://www.newegg.com/Product/Product.asp?Item=N82E16817104951

 

[FghtinIrshNvrDi]

I figure if I'm going to buy a new power supply. I'll up the power a little. 400-450ish watts. What do you think about ATADC power supplies? If that's no good, is it plausable to get a better than marginal supply (with that wattage) for under 30?

Ryan

 

[DJ-CHRIS]

Dj-Chris, just to point out that the Memory bandwith advantage on going with dual channel is already accounted for. Why do you think a 3000+ 754 is 2.0 GHz while a 3000+ venice is 1.8? Not becuase of the 90nm process but because of dual channel. Yes you get a 50% boost in memory bandwith, but you only get 3-5% boost in real life performance. Also, the Newcastles can def overclock. On average people can at least get 200MHZ out of them. Most can get more, but lets stay on the low side of things. Finally, you blow the energy savings and heat dissipation out context. 130 nm processor do not require that much more energy than 90nm ones. You say ALOT more, but in reality it is only 10 or so more watts. That isn't alot more I would say. So to sum it up, what are the advantages of 939 over 754? Pci-express, which poses no advantage over AGP RIGHT NOW. Dual channel memory, which is already taken accounted for in the modeling numbers by AMD. And finally dual core. Those are the advantages comparing 754 to 939.

Finally DJ-Chris, he OP wanted credible links because it is his money. If you choose not to provide them, thats fine. Just that he'll beleive people who give links more than people who wont, because frankly websites are a little more credible than a poster in a forum.
Btw, here is a detailed look at the sempron. The model that was reviewed is based on the 130nm process, and the article goes into detail about 754 versus 939. Also, there is an overclock mixed in there too.

http://www.xbitlabs.com/articles/cpu/display/sempron-3100-oc.html

The newer cores also have slightly more peformance, and the difference maybee 3-5% while using windows, but memory demanding programs would have a larger peformance gain from memory.

If he wants to go socket 754 that's up to him, and I gave him my advice. If you want to search google for him to find benchmark's, that's up to you.

And for overclocking and power consumption, I actually do have a newcastle 3400+ (939, however). I got more than a 200mhz boost, up tp 2.66GHZ. However this is a rare speed. Since the 754 3400+ is 2.4 ghz, your likely not going to see 2.66 ghz, let alone stable at that frequency.

And power consumption, you can tell yourself it's 10 watts, but it's closer to half.

 

[gaara]

The 3000+ is faster than the 3400+ clock for clock. Throw in better thermal specifications and a similar pricetag, and the 3000+ comes out on top, end of story.

 

[Hacp]

Edit: Since this is not necessary anymore, I have deleted it.

 

[FghtinIrshNvrDi]

I do now know from sparatic research that the 3400+ 754 will outpreform the 3000+ 939, so I think I'm going to go with it. I'm not planning on doing any overclocking at all, so that doesn't concern me. I think I could figure it out, but why overclock a processor that fast. I have no applications demanding that much speed. I think 400W ought to do it. And ya'll can go back to your 'discussion'

Ryan

 

[Hacp]

Gl with that.