Quicker Sempron 3000+ or 3100+

[badaxxe]

I just bought two Emachines a W3050 and W3115 for 5 dollars. Power supply bad in one. Built one out of two. Any idea which processor is quicker. I realize Emachines aren't the best just trying to squeeze the most out of them for the kids.

 

[toyota]

the 3100 will be faster but if they are both socket 754 but not enough to matter. if thats a older socket A 3000 than the 754 3100 is quite a bit faster. really all them will be just as irrelevant for modern use.

 

[Harvey]

Google is your friend.

emachine w3050 specs.

emachine w3115 specs.

The 3115 has a bigger hard drive and newer mobo and graphic chipsets. You'll want more RAM than the stock 512K no matter which board you use. Max it out to 2 GB for best results. Even then, it's just an emachine, but for what you paid, you'll get your money's worth.

 

[badaxxe]

Thank you for the info!

 

[Plimogz]

wait a minute now, that sempron 3000+ -centric machine is a socket A, correct? I mean, I can't be bothered to check thoroughly, but with the links that Harvey was good enough to post I can see that the 3050 appears to have a 333 FSB (versus the 3115's 1600MHz system bus). So that would likely denote a Barton derivative socket 462 chip and either a Paris or Palermo socket 754 chip...

If this is correct, we might be talking about a dual-channel memory friendly nForce2 system Vs. a single-channel socket 754 system. Now, although Athlon64 (socket 754) was the superior architecture, no question; low-end socket 754 really wasn't all that much faster than the fast socket A systems which came before it... So, we have a 2GHz, 512Kb L2 socket A chip going up against a 1.8GHz, 256Kb L2 socket 754 chip. Sure, the former hails from an older architecture, whereas the latter is technically part of the much-loved K8 familly; I still really don't know that the 200 extra MHz on the 3050 can't compensate for the architecture difference, and that the 2x L2 can't make up for the FSB/IMC variance...

OP: I think you'll find the definitive answer to your question by looking around for what cheap upgrade chips you might be able to find for either socket A or socket 754 -- both are really quite inexpensive at this point, and whatever upgrade chip you can find for either socket would just trounce the other, un-upgraded system.

Furthermore OP, if you happen to have some DDR-400 sticks on hand, the obvious choice would be the socket 754 (3115) system on account of A64's superior memory multiplier handling. On the other hand, if you happen to have some matched pairs of dual-channel PC-2700 DDR laying around, the socket A machine (3050) may well put up some better gaming numbers on account of the possible dual-channel functionality of that nForce2 chipset.

In any case, this talk of the AMD of old has me full of warm and fuzzy feelings; so I'll take some time to drunkenly browse the net for info until I have a firmer grip on what functionalities both those Nvidia chipsets might offer you, and I'll post back later (maybe when I'm not seeing quite so double ).

 

[toyota]

yes that 3000 machine is a socket A. I actually bought my parents that same pc many years ago. of course now they have a much more modern pc. that socket A 3000 Sempron basically performs about like an Athlon XP 2600+.

 

[Plimogz]

slow down there, 2GHz/512Kb L2 wasn't a 2400+. It was either sempron 3000+ or athlonXP 2600+.

damn, you had time to correct your 2400+ for 2600+. well done ninja.

 

[toyota]

slow down there, 2GHz/512Kb L2 wasn't a 2400+. It was either sempron 3000+ or athlonXP 2600+.

damn, you had time to correct your 2400+ for 2600+. well done ninja.

yeah I had changed it. I had an Athlon XP 2400 so I was thinking they were the same for some reason. I looked up the specs and seems that a socket A Sempron 3000 would basically be like a Athlon XP 2600.

 

[Plimogz]

still, do you figure that it would be that much of a win for the s754 machine? from what I can remember, those systems really didn't outshine faster clocked socket A's.

Then again, maybe I'm just remebering all those posts by people with well overclocked Abit boards sporting Athlon-M's laughing at the "under-performance" of the "new" AMD chips...

 

[toyota]

still, do you figure that it would be that much of a win for the s754 machine? from what I can remember, those systems really didn't outshine faster clocked socket A's.

Then again, maybe I'm just remebering all those posts by people with well overclocked Abit boards sporting Athlon-M's laughing at the "under-performance" of the "new" AMD chips...

well they are both very slow but yes the 754 3100 would be faster. I believe even the 754 2800 is faster than the socket A 3000.

 

[Plimogz]

dude, Toyota, you may have been right. This is copy/pasted from an Anandtech article:

AMD Athlon XP Model Numbers
CPU Name Clock Speed
Athlon XP 3000+ (Barton) 2.167GHz
Athlon XP 2800+ (Barton) 2.083GHz
Athlon XP 2800+ (333MHz FSB) 2.25GHz
Athlon XP 2700+ (333MHz FSB) 2.167GHz
Athlon XP 2600+ (333MHz FSB) 2.083GHz
Athlon XP 2600+ 2.13GHz
Athlon XP 2500+ (Barton) 1.83GHz
Athlon XP 2400+ 2.00GHz
Athlon XP 2200+ 1.80GHz
Athlon XP 2100+ 1.73GHz
Athlon XP 2000+ 1.67GHz
Athlon XP 1900+ 1.60GHz
Athlon XP 1800+ 1.53GHz
Athlon XP 1700+ 1.47GHz
Athlon XP 1600+ 1.40GHz
Athlon XP 1500+ 1.33GHz

I'm not done reading yet, but it looks good for your original statement .

Edit: Ok, the 3100+ socket 754 would in fact be faster. After looking through the great numbers provided by Anandtech in this article, I have to admit that I was all wrong.

 

[formulav8]

The 3100+ Sempron is AMD64 based (No 64 bit though!). It is definitely faster than the 3000+ in basically everything. The 3000+ was a pretty rare chip. I don't remember seeing to many of those in the field. It is Barton based with 512KB L2 and on the 333mhz fsb like some have mentioned. I might actually be interested in this chip if you don't use it

Either of those cpu's will handle windows xp for regular users. I still build lower cpu's for people and they are perfectly content. Especially with the money savings


Jason