It seems like the stats are the same. I want something to replace my FX-53 and have an opportunity to get a opty 185. Just wan't to know what the difference is, if any, between the two. Thanx.
Printable View
It seems like the stats are the same. I want something to replace my FX-53 and have an opportunity to get a opty 185. Just wan't to know what the difference is, if any, between the two. Thanx.
The Opteron 185 is better silicon and the Opteron series excels at Video Encoding actually better than C2D or C2Q does:)
the Opty 185 and FX-60 are pretty much the same thing except the FX has an unlocked multi.
exactly the same except the FX60 has unlocked multiplier...
Brother E - how is the Opty better @ vid encoding? got any linky reads? cheers :up:
http://bztv.typepad.com/moviessquare...otalrecall.jpgQuote:
Originally Posted by Brother Esau
.. anyway. The opty 185 has no unlocked multiplier but if you find the right stepping, it'll probably clock better on air/water than the FX-60. I'm looking for one myself :yepp:
Apparently my mobo does not support this processor but supports the FX-60?!?
http://www.gigabyte.co.nz/Support/Mo...ProductID=1788
I don't get it. :shrug:
Quote:
Originally Posted by biohead
Thats pretty funny man gotta admit I am laughing pretty hard at Arnie Pic steroid power dump with got voltage:rofl: :up:
dont worry the opteron will work perfectly.Quote:
Originally Posted by Rob94hawk
Alot of boards out there dont ''officially'' support the opterons but stil they work.
no consumer mainboard officially supports opteron CPUs, cause they are ment to run in workstations. but itīs just a s939 cpu so it will run in any s939 mainboard that supports dualcore CPUs.
umm.. iīm asking that myself.. isnīt video encoding just crunching numbers?Quote:
Originally Posted by tiro_uspsss
we all know how well core2īs do in crunching numbers...
It will run, but it probably won't support Cool'n'Quiet. If you're an overclocker that probably doesn't matter to you. For me, I rewrote my ACPI DSDT (on Linux) so that it would enable C'n'Q at VIDs and FIDs of my own choosing. Under Windows, no such luck.Quote:
Originally Posted by MadDias
(Running an Opteron 185 on an Asus A8V-Deluxe, latest BIOS...) I have it undervolted and underclocked at the lowest setting (0.8V at 800MHz), and overvolted and overclocked at the top setting. (1.45V at 3.0GHz). I also have 4x1GB registered ECC DDR400 in it; the BIOS automatically limits it to DDR333 but I tweak the memory controller in my boot script to bring it back to DDR400/1T.
:eek: u have 4 dimms @ 1T?? tell me how :D please :DQuote:
Originally Posted by hyc
They're registered DIMMs. Not possible at all with unbuffered.
Although I personally find it difficult to believe, Opterons have 3 HT links and A64s only have 1. I beilieve the extra HT links are primarily for multi socket mobos so that cpus can dirrectly interconnect. I cant really see there being an advantage of the extra 2 links in a 939 board tho.
Anyone?
the opteron 1xx series only have 1 ht link enabled, like 2xx have two - makes sense.Quote:
Originally Posted by bacon612
the only difference i can think of is that the optys come in a variety of steppings that tend to clock better than A64s. i think that's mainly statistically speaking; when chosing between A64/opteron, it's generally down to luck.
Opteron 1xx series are pretty much the same as any other Socket 939 processor - the die has 3 HT links but only one of them is pinned out on the socket. Opteron 2xx and 8xx have all three links present on the Socket 940 pinout, but 2xx only allows one link to support ccHT (cache coherent) mode for communicating with another processor. 8xx allows any/all 3 links to be ccHT.
Talking about any advantage to having 2 more HT links in a 939 board makes no sense. This is the main difference between Socket 939 and Socket 940 - the pins that were used for two of the three HT links in Socket 940 were replaced in the Socket 939 by pins for the memory controller. Since Socket 939 was designed to use unbuffered memory, all of its data lines are doubled to allow the chip to drive up to 4 unbuffered DIMMs. Without that doubling, it would only be able to drive one unbuffered DIMM per channel, which would have been too limiting.
The whole point of Socket 939 was to provide a cheaper option for the desktop, that wouldn't require buffered/registered DIMMs. The only way to support unbuffered was by providing redundant signal lines for the memory interface. The only way to get those signal lines and still make it cheap enough for the desktop market was to remove something else from the socket - the 2 missing HT links.
Interesting to note that about 160 pins are different between Socket 939 and 940, so if you wanted a socket with both (a) all 3 HT links and (b) flexible memory interface support it would take at least 1100 pins. Kinda tells you a little about what's in Socket 1207.