AMD Athlon 64 3800+ and FX-53: The First 939 CPUs
by Derek Wilson on June 1, 2004 12:30 AM EST- Posted in
- CPUs
Development Workstation Performance
Interestingly, removing the latency of the registered RAM from the socket 940 based FX processors really helps speed up compile time. Here we see a whole second increase in compile time for the FX-53 processor (which translates to a very long time when only talking about 13.4 seconds to being with). This is interesting because we would expect most of the data we are compiling to be in the L2 cache after the first couple times we compile it, but apparently not even 1MB of cache is enough keep the memory interface from having an impact on performance.
Facebook
Twitter
RSS
38 Comments
View All Comments
Pumpkinierre - Tuesday, June 1, 2004 - link
That earlier french review couldnt get 4 DS sticks to work at DDR400:http://www.x86-secret.com/articles/cpu/s939/s939-3...
Still, I'd sooner have the dual bank memory than the extra cache. But if it is $800 and $600 then the Skt754s become good buys.
intercollector - Tuesday, June 1, 2004 - link
#6 - With memory prices where they're at now, the more dimms the better.For example, if you were to be building a system right now, you'd be hard pressed to justify getting anything more than 1GB (2X512). This fills up 2 slots already, thus only allowing 1 free slot for upgrade in the future. If you were to upgrade with another 512 stick, you'd have 1.5GB, but all your spots filled. If you want this system to last a while, that might just not do.
I say that 3 is definately the minimum you'd ever want, and wanting 4 is definately not out of the question.
Jeff7181 - Tuesday, June 1, 2004 - link
#5 - Is running four unbuffered DIMMs really that necessary? I mean... I haven't ever had more than two DIMMs since my 486, which technically didn't have any DIMMs... but it had 4 one MB SIMMs.mechBgon - Tuesday, June 1, 2004 - link
One potential benefit to the Socket939 platform is that it should open the door to running four unbuffered DDR400 DIMMs.JGF - Tuesday, June 1, 2004 - link
Hmm, now Toms is reporting the $700US+ price of the 3800+ as well. This had better not be true. Roughly $50 seperating the FX line and the 'regular joe' A64 line?? Please AMD, tell me this isnt right...Jeff7181 - Tuesday, June 1, 2004 - link
Well well well... looks like socket 939 isn't really worth waiting for, especially if it's priced a lot higher than socket 754. Socket 754, here I come :)JGF - Tuesday, June 1, 2004 - link
Good read. Couple of disappointments (though not neccessarily wesley's fault):1. Pricing/availability - supplies of 939 parts are rumored to be VERY constrained and some quarters are even uttering the dreaded p-phrase (paper launch). Also early listings for the 3800+ are placing it at an astronomical $700US+ which is obviously bunk. SAtill we are without any official statement regarding price and availability.
2. No 3700+ scores or even a word with what is happening there. With the rumored constraints on 939 and the fact that PCI-E from VIA or nvidia wont be around until late summer, I'm actually seriously looking into a 3700+ 754 system. When will we get info?!?
dvinnen - Tuesday, June 1, 2004 - link
hmmm, 30 mins after the nda ended. Yall need to get on the ball.