Newegg has the socket 939 3000 and 3200 now!

ShepsCrook said:
They were released this evening and are in stock!

Third time they've had them in stock.

Prices are absurd. For 3000+:

First time $168
Second time $175
Now $190
 
Whenever these things come out the prices seem to rise dramatically. The 3500 was $350 at first, then flew up to $450, and finally is back down to more reasonable levels. Given time they will go down again.

Then again I think I said that about the 6800 series.
 
Monarch has them as well. I ordered mine yesterday. Grrr, fedex says it won't arrive til monday.
 
etruscan said:
Monarch has them as well. I ordered mine yesterday. Grrr, fedex says it won't arrive til monday.
aren't they more expensive at monarch?
 
Why does newegg always screw-up the info on these things. Here they say it's a newcastle core, and then there's some info about it being a 3000+ or some such.
 
Yeah saw this too - damn used to be $20 less? Still need to decide on mobo though...
 
Yeah, but there's no date on the Release of NF4 boards so you can't count on that.
 
Well, I placed an order for the Newegg bundle.

MSI Neo platinum 2
3400+ 939pin
Kingston 512MB DDR400, the timings on the ram are average, but hey it's just a placeholder anyway. for 454 (free shipping)

I have a nice 80GB SATA drive, and a 6800GT waiting for a new home. Thank god I held out for the 939. I was actually waiting on the 3500+ to drop, but hey got something just as good. For a while I was thinking of going 754, but then I saw all the new goodies that the 939ers' were going to get.
 
Ke0 said:
Well, I placed an order for the Newegg bundle.

MSI Neo platinum 2
3400+ 939pin
Kingston 512MB DDR400, the timings on the ram are average, but hey it's just a placeholder anyway. for 454 (free shipping)

I have a nice 80GB SATA drive, and a 6800GT waiting for a new home. Thank god I held out for the 939. I was actually waiting on the 3500+ to drop, but hey got something just as good. For a while I was thinking of going 754, but then I saw all the new goodies that the 939ers' were going to get.

Wait a sec, from what I've read on here you may want to take a look at that 3400+ 939. I think it's a crippled CPU.
 
EmptyWallet said:
Wait a sec, from what I've read on here you may want to take a look at that 3400+ 939. I think it's a crippled CPU.

Really? Well if it's true, i'll ebay it
 
Crippled? In what sense? I'm not sure what the heck it's supposed to be actually... Newegg lists it as a 130mm part, the slowest s939 part previous to the 90mms was the 3500+ (130mm) at 2.2ghz.

The 3400+ Newegg has listed has the exact same stats (2.2ghz, 512 L2, s939, 130mm) as the 3500+ and says it's made on the 130mm process? Sounds like some massive typ or something... Not sure what chip it could be really since the next step down is 2.0ghz and that's the 3000+ 90mm.

Doesn't make any sense to me, maybe someone else can figure it out... Unless this is somehow their way of seeling the 3500+s cheaper previous to the predicted price cut (since the 3400+ they have listed with identical specs to the 3500+ is like $40 cheaper).
 
Ke0 said:
@ The AMD forums said it's a 3500+ but with 800HTT

Don't think that'd make a lick of difference seeing how little difference people have experienced in performance when playing around with their HT multipliers while OC'ing... What AMD forums?

Also, see this thread on the topic:

http://www.hardforum.com/showthread.php?t=819394&page=1

Haven't read it yet, going to now, two pages worth of replies on it so far tho.
 
Well the conclusion on the above linked thread can be summed up by the Inq's chart on the link below (provided by Kovalen in the thread):

http://www.theinquirer.net/?article=18925

Looks like you're right, it's basically a 3500+ with a slightly lower HTT... Good deal if you ask me, doubt you could tell the difference at all... Seeing as the difference between 754 parts and 939 parts is like 5-10% due mostly to the dual channel bandwith (all the 754s are on the 800 HTT). The HTT is never actually maxed so it really matters very little. Not sure why they've bothered to release this part at all tho.
 
It was reading it over at forums.amd.com. I don't know why they released it either, but it's a steal, I mean either pay 355? or 288 for the same processor with a slightly lower HHT, which you can get back to 1000 anyway
 
yes! finially! Time to get me a 3000+ and MSI Neo2 Platnium :D

wow... only $340
 
Give newegg cudos for hardopc...

I would have missed this.

Got 2 cpus and mbs on the way .90 die...

xman
 
For those of you who have purchased the Socket 939 3000+, could you please post your overclocking results.

There's an article over at Anandtech where they were able to get the 3000+ from 1.8Ghz to 2.6Ghz on air with a slight bump up in the voltage. However, I'm not seeing any other results posted. I don't know if it's because they're so new or what.

Anand Article

I am seriously considering getting one of these with an MSI Neo2 Plat if I start seeing some consistant overclocking results. (I know, I know; there's no guarantees :p )

Thanks All.
 
Madshrimp's article showcased results nearly identical to Anand's, a succesful and stable 2.6 OC on the 3000 90mm. From the three or four 90mm threads I've seen here where people have bought the new parts and pushed them it did seem to indicate they were at 'least as good of an OC part as the older 130mm process chips...

'Course your milage may vary and I doubt every chip will get to 2.6 or even 2.5 (tho I'd bet most will make it to 2.4 with adequate cooling and tweaking)... It depends on the chip as well as the mobo, RAM etc... Nothing new really. Seems like s939 has finally become a viable low-cost upgrade path.
 
See my sig for my results. I don't think I got the best chip off the line and would say that my chip is probably one of the lower end overclockers out there.

Mine rune VERY hot, even at default clocks and voltages, but that could be just the motherboard being off. Either way, right now I'm idling at 50C and load at 55-60C. No matter what voltage I'm at I get the same results, no clock speed above 2.5GHz will go for more than 1 minute in prime. In fact, when I add voltage, it lowers the time, makes the chip idle as high as 55C.

That said, this chip is rock solid at 2.4GHz no matter what my room conditions are. That is at 1.55Vcore. 2.5GHz requires it to be winter in my room to be stable under constant load(I can play Generals for about an hour before generals crashes to desktop stating hardware issues). Which is fine for nights like last night.
 
Dew said:
See my sig for my results. I don't think I got the best chip off the line and would say that my chip is probably one of the lower end overclockers out there.

Mine rune VERY hot, even at default clocks and voltages, but that could be just the motherboard being off. Either way, right now I'm idling at 50C and load at 55-60C. No matter what voltage I'm at I get the same results, no clock speed above 2.5GHz will go for more than 1 minute in prime. In fact, when I add voltage, it lowers the time, makes the chip idle as high as 55C.

That said, this chip is rock solid at 2.4GHz no matter what my room conditions are. That is at 1.55Vcore. 2.5GHz requires it to be winter in my room to be stable under constant load(I can play Generals for about an hour before generals crashes to desktop stating hardware issues). Which is fine for nights like last night.
Nice job man.... one thing though newegg price jumped up .. 190$ ... to 215$
 
Creizai said:
Nice job man.... one thing though newegg price jumped up .. 190$ ... to 215$

Ugh... ya i just noticed that too, i guess ill be holding off a while. :mad:
 
I noticed that also about the 3000 939 going up and down in price. Monarch has it OEM for 179 not in stock though. New Egg has gone from 190 to last price i saw was 219. I thought amd was dropping prices this month. I am looking into the MSI NEO2 PLAT and the 3000 socket 939. Any feedback on that combo? And what do guys think about the antec sonata case. Sorry for all th Questions.

Thanks

Jonesy
 
Both Monarch and Newegg are out of 90nm's. I ordered one from Monarch 2 days ago and they just told me that they didn't have any more in stock. I love it when a company feels it's ok to take your money and then not send out the product you ordered.
 
Not anymore, they prices all went to over $200 a few hours after I bought one...
 
Santos L. Halper said:
For those of you who have purchased the Socket 939 3000+, could you please post your overclocking results.

There's an article over at Anandtech where they were able to get the 3000+ from 1.8Ghz to 2.6Ghz on air with a slight bump up in the voltage. However, I'm not seeing any other results posted. I don't know if it's because they're so new or what.

Anand Article

I am seriously considering getting one of these with an MSI Neo2 Plat if I start seeing some consistant overclocking results. (I know, I know; there's no guarantees :p )

Thanks All.
I saw that same article. I'm interested in seeing what kind of results [H]ardOCP gets from one. Although, I kind of wish that they put the benchmarks for that system in with their review of the 4000+ and FX-55. With both the 90nm 3000+ and 3500+ coming in at 2.6GHz like the FX-55, it would be nice to see how they compare. Especially considering that the 3000+ is around $200 while the FX-55 is over $800.


Impulse said:
Madshrimp's article showcased results nearly identical to Anand's, a succesful and stable 2.6 OC on the 3000 90mm. From the three or four 90mm threads I've seen here where people have bought the new parts and pushed them it did seem to indicate they were at 'least as good of an OC part as the older 130mm process chips...

'Course your milage may vary and I doubt every chip will get to 2.6 or even 2.5 (tho I'd bet most will make it to 2.4 with adequate cooling and tweaking)... It depends on the chip as well as the mobo, RAM etc... Nothing new really. Seems like s939 has finally become a viable low-cost upgrade path.
Impulse, happen to have a link to any of those topics handy?
 
My results (definitely preliminary), for A64 3000+ 90nm.

I have run Prime95 for 2+ hours at 280 MHz FSB, 2520 MHz.

I get BSOD at 290, at +8.3% voltage, 1:1 memory divider.

I seem to be stable at 285, but prime is only 20 minutes in.

I will drop my memory timings massively and see how fast I can get the CPU (or maybe I'm testing my motherboard now).

EDIT:290 is just not prime95 stable. Even at +13%, prime dies after about 10 seconds when I turn the memory down. I haven't found my exact limit, but it's somewhere between 2565 MHz, and 2610. Interestingly, it doesn't seem to be a voltage issue, there's just some point after which the CPU craps out.

Still, even if 2565 is my highest prime stable OC, I'm getting an extra 765 MHz free, out of an 1800 MHz processor.
 
If you look at the Newegg advert on the front of HardOCP, they apparently have the Athlon 64 "30000+" available for only $163!!! Bargain! (it's the same advert at the top of this page but hasn't been corrected from what it seems)
 
Back
Top