mersenneforum.org Launch of the AMD FX 8 Core processor
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

 2011-08-28, 21:23 #1 moebius     Jul 2009 Germany 613 Posts Launch of the AMD FX 8 Core processor http://www.youtube.com/watch?v=qHa7ajE4qVM Hooray, this one should still fit well on my AM3-socket (with Bios update of course)! Who buys it and posts a Prime95 benchmark?
2011-08-29, 01:20   #2
science_man_88

"Forget I exist"
Jul 2009
Dumbassville

26×131 Posts

Quote:
 Originally Posted by moebius http://www.youtube.com/watch?v=qHa7ajE4qVM Hooray, this one should still fit well on my AM3-socket (with Bios update of course)! Who buys it and posts a Prime95 benchmark?
Quote:
 AMD is launched the FX Bulldozer processors on 26th of September. 8 core's with ultimate power for your pc.
odd they are supposedly releasing it on my birthday and the day the world should end according to http://www.mersenneforum.org/showthr...light=birthday

also note the grammar issue it should read "will launch".

Last fiddled with by science_man_88 on 2011-08-29 at 01:23

 2011-10-10, 21:20 #3 Batalov     "Serge" Mar 2008 Phi(4,2^7658614+1)/2 13·733 Posts Is anyone preparing to build a 6GHz 8-core slugger? The launch of 8150 is rumored to be on Wednesday.
2011-10-11, 19:37   #4
moebius

Jul 2009
Germany

613 Posts

Quote:
 Originally Posted by Batalov Is anyone preparing to build a 6GHz 8-core slugger?
The problem is, that Gigabyte probably will not update the Bios of older motherboards to the B2 stepping. Facing of the price of about € 230 in germany for the FX8150, it will take for me so long till my current quad-core is broken.

Last fiddled with by moebius on 2011-10-11 at 19:38

 2011-10-11, 20:18 #5 Batalov     "Serge" Mar 2008 Phi(4,2^7658614+1)/2 13·733 Posts Yeah, for a test of 1055T (and because I am too cheap a bastard to have bought a new MB and DDR3 memory), I've waited for GA to issue a BIOS update - and it took them 3 months iirc. But this time I am thinking... DDR3 16Gb can be had for ~$100 (unlike a year ago; this is down 5 or 10 times!), and a decent board is only ~$100, too. Plus $245 (or so they say) for the 8150? ...doesn't sound too bad. P.S. and oh yes, a liquid cooler for a$100? could be fun. 5GHz should be 'easy' even with air.
 2011-10-12, 02:50 #6 Christenson     Dec 2010 Monticello 34038 Posts If the first of the year goes well, (in my external, non-GIMPS life) I might be willing to plunk down and do this...must remember to put in plenty of power supply overhead for a large CUDA card, like a GT580 or so...
 2011-10-12, 06:53 #7 mdettweiler A Sunny Moo     Aug 2007 USA (GMT-5) 3·2,083 Posts Say, I just noticed reading this article (recently linked in this thread) that the new Bulldozer CPUs are constructed in a rather interesting way: the advertised # of cores actually only represents integer cores, with the floating-point cores being shared at a rate of one FP/two integer cores. As the article indicated, for many applications this is a worthwhile trade-off (why use the transistors to build hefty floating-point firepower when you can use them for more integer power instead), since most are decidedly integer-heavy; but LL testing and the like is the exact opposite, being almost exclusively FP-heavy. If I'm understanding this correctly, this would seem to indicate that the Bulldozer CPUs will be relative crap for LLs (perhaps half as fast as Intels per-core!), though they'll do very well at TFs. Ironically, it harks back to the days when AMD Athlon XP and 64 CPUs were known as being better at TFing and such, while their Pentium 4 competitors were correspondingly better at LL...but now that we have GPUs able to do TF orders of magnitude faster, that's the last thing we need! Last fiddled with by mdettweiler on 2011-10-12 at 06:54
 2011-10-12, 08:38 #8 joblack     Oct 2008 n00bville 23×7×13 Posts http://www.tomshardware.com/reviews/...90fx,3043.html Doesn't look very promising. I think I will go with a 2500K ...
2011-10-12, 11:53   #9
Mini-Geek
Account Deleted

"Tim Sorbera"
Aug 2006
San Antonio, TX USA

10AB16 Posts

Quote:
 Originally Posted by mdettweiler it harks back to the days when AMD Athlon XP and 64 CPUs were known as being better at TFing and such, while their Pentium 4 competitors were correspondingly better at LL
IIRC, all-integer LL code could have been written that would have made the AMD CPUs competitive with the P4s for LLs, but due to the number of P4s vs AMDs, it was deemed not worth the time to write and optimize it to be better than FP. Perhaps it is now worth the time, if this CPU becomes a major part of GIMPS?

Last fiddled with by Mini-Geek on 2011-10-12 at 11:55

2011-10-12, 12:12   #10
axn

Jun 2003

210×5 Posts

Quote:
 Originally Posted by Mini-Geek IIRC, all-integer LL code could have been written that would have made the AMD CPUs competitive with the P4s for LLs
Your recollection is faulty. It was never the case that AMD (or any x86 cpu) integer performance was so good that it would have beat the FPU version on the same processor.

 2011-10-12, 12:35 #11 em99010pepe     Sep 2004 54168 Posts Batalov, For msieve use in LA phase do you recommend one AMD 1100T with 16 GB or one dual Intel® Xeon™ X5650 SixCore 2.26GHz 12M 6,4 QPI 12 MB with 4x Dimm 8GB 1333MHz DDR3 ECC Reg w/Parity CL9 DIMM? Does msieve scale well with increase of number of cores? Sorry for the question but I have not experience running msieve on a CPU with more than 4 cores. Thanks in advance, Carlos

 Thread Tools

 Similar Threads Thread Thread Starter Forum Replies Last Post swl551 Software 1 2012-09-24 23:29 stars10250 Hardware 16 2011-06-25 17:24 kaeptn_kork Hardware 11 2009-03-17 01:58 xilman Factoring 12 2006-07-11 16:23 drew Hardware 5 2006-05-29 07:00

All times are UTC. The time now is 08:02.

Tue Sep 21 08:02:36 UTC 2021 up 60 days, 2:31, 0 users, load averages: 1.60, 1.45, 1.40

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.